W3C home > Mailing lists > Public > www-talk@w3.org > May to June 1995

Re: Session tracking

From: Gary Adams - Sun Microsystems Labs BOS <gra@labboot.east.sun.com>
Date: Mon, 1 May 1995 08:02:40 +0500
Message-Id: <9505011202.AA05125@labboot.East.Sun.COM>
To: nazgul@utopia.com, www-talk@www10.w3.org
> Date: Sat, 29 Apr 1995 15:13:26 +0500
> From: nazgul@utopia.com (Kee Hinckley)
> Subject: Re: Session tracking
...
> 
> It does seem to me that the magic-cookie design is very closely tied to
> existing password systems, and in that respect I think it's worth
> considering whether the two mechanisms might be tied together more tightly
> (a user password system with expirations makes perfect sense, for
> instance). I haven't delved into that side of the protocol enough to say
> any more.

This is a very good point, that some of the "identifiers" (session, cookie,
whatever) should have a similar life cycle as security credentials (where
passwds are a valid instance of server side authentication).

> 
> Shopping carts embedded in ids is a cute hack, but it's a red herring. The
> real goal in my mind is to find a way to identify a user without requiring
> them to carry a separate ID for every store they walk into.

It seems to me that a "user centric" view of the web would call for
client side generation of the credentials, that could be reused
at many different storefront businesses.i.e. shopping at a mall
rather than a department store for one stop shopping.

eturn-Path: amosley%pubspo.hq.af.mil@hq.hq.af.mil 
Return-Path: <amosley%pubspo.hq.af.mil@hq.hq.af.mil>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA23904; Mon, 1 May 1995 08:41:11 +0500
Received: from LCS.MIT.EDU (mintaka.lcs.mit.edu) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA04185; Mon, 1 May 1995 08:41:10 +0500
Received: from hq.hq.af.mil by MINTAKA.LCS.MIT.EDU id aa10482; 1 May 95 8:40 EDT
Received: from pubspo.hq.af.mil by hq.af.mil (4.1/Mork-1.0)
	id AA02491; Mon, 1 May 95 08:42:53 EDT
Received: by pubspo.hq.af.mil with Microsoft Mail
	id <2FA4FFFC@pubspo.hq.af.mil>; Mon, 01 May 95 08:36:28 PDT
From: "Mosley, Art" <amosley@pubspo.hq.af.mil>
To: www <www-talk@www10.w3.org>
Subject: HTML 2.0/3.o
Date: Mon, 01 May 95 07:30:00 PDT
Message-Id: <2FA4FFFC@pubspo.hq.af.mil>
Encoding: 8 TEXT
X-Mailer: Microsoft Mail V3.0
Content-Length: 142




Does anyone know about HTML 2.0/3.0 - - SGML directly on the Internet? 
 Where I could how it works or how to get information?

Thanks
Art
eturn-Path: mattison@freenet.victoria.bc.ca 
Return-Path: <mattison@freenet.victoria.bc.ca>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA14231; Mon, 1 May 1995 10:50:15 +0500
Received: from vifa1.freenet.victoria.bc.ca ([199.60.222.1]) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA06103; Mon, 1 May 1995 10:50:08 +0500
Received: (from mattison@localhost) by vifa1.freenet.victoria.bc.ca (8.6.9/8.6.9) id HAA21442; Mon, 1 May 1995 07:51:06 -0700
Date: Mon, 1 May 1995 07:51:05 -0700 (PDT)
From: David Mattison <mattison@freenet.victoria.bc.ca>
Subject: Re: HTML 2.0/3.o
To: amosley@pubspo.hq.af.mil
Cc: Multiple recipients of list <www-talk@www10.w3.org>
In-Reply-To: <2FA4FFFC@pubspo.hq.af.mil>
Message-Id: <Pine.2.2.9505010713.A20084@vifa1>
Content-Length: 814

According to Laura Lemay's TEACH YOURSELF WEB PUBLISHING WITH HTML IN A
WEEK (SAMS Publishing, 1995), the HTML 2 spec is at:

http://www.hal.com/users/connolly/html-spec/index.html

The HTML+ 3 draft spec is at:

http://www.w3.org/hypertext/WWW/MarkUp/HTMLPlus/htmlplus_1.html

She gives the above link host as info.cern.ch, but I understand that's
been changed to what I've given.

Lemay also lists a site with the HTML+ 3 draft spec in PostScript format:

http://www.ics.uci.edu/WWWdocs/papers/draft-raggett-www-html-00.ps.gz

David Mattison
Gopher/Web Co-editor
Victoria Free-Net
mattison@freenet.victoria.bc.ca

On Mon, 1 May 1995, Mosley, Art wrote:

> 
> 
> 
> Does anyone know about HTML 2.0/3.0 - - SGML directly on the Internet? 
>  Where I could how it works or how to get information?
> 
> Thanks
> Art
eturn-Path: montulli@netscape.com 
Return-Path: <montulli@netscape.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA01242; Mon, 1 May 1995 16:12:22 +0500
Received: from strumpet.mcom.com (unknown.netscape.com) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA08583; Mon, 1 May 1995 16:09:56 +0500
Received: (from montulli@localhost) by strumpet.mcom.com (8.6.10/8.6.9) id NAA18159; Mon, 1 May 1995 13:03:55 -0700
From: "Lou Montulli" <montulli@netscape.com>
Message-Id: <9505011303.ZM18157@strumpet.mcom.com>
Date: Mon, 1 May 1995 13:03:53 -0700
In-Reply-To: Joerg Rhiemeier <rhiemeir@ibr.cs.tu-bs.de>
        "Re: Dynamic HTML documents with client pull" (Apr 28,  6:57am)
References: <199504281049.MAA03204@kastor.ibr.cs.tu-bs.de>
X-Face: 2%:,dh#NyL=h+${dpRhMgWq_:WU|hRg,gd:q08ah!sc~a^V?8x}:6YSw>)K1n"&>ebiBHx1H5#zr*6qoFw%K,:yyH,'D7^54~HcZ4%CEdVGu#GE3T5WMztPF1GIhXU\-vT2{u61n8cJzcY7SxCzRl'O-`f"93K.yFdPJXrG#|7R
X-Mailer: Z-Mail (3.1.0 22feb94 MediaMail)
To: rhiemeir@ibr.cs.tu-bs.de,
        Multiple recipients of list <www-talk@www10.w3.org>
Subject: Re: Dynamic HTML documents with client pull
Content-Type: text/plain; charset=us-ascii
Mime-Version: 1.0
Content-Length: 1843

On Apr 28,  6:57am, Joerg Rhiemeier wrote:
> Subject: Re: Dynamic HTML documents with client pull
>
> In message <v0151010cabc660f3d752@[130.237.112.5]>,
> Ulf.Kronman@it.ki.se wrote:
>
> >First, sorry if this is some kind of a FAQ or has been debated here
before...
> >
> >I was looking at Web page yesterday, when Netscape (1.1 mac) suddenly
> >started to download a new page by itself. I thougth that I had clicked on a
> >link by mistake, but this repeated several times without my intervention.
> >Very annoying, because I hadn't time to read the first page before the
> >second came.
>

> The <META...> element is legal HTML 3.0, and the HTTP-EQUIV attribute
> also.  Its purpose is to add HTTP headers, for example
>
> <META HTTP-EQUIV=Expires CONTENT="Sat, 6 May 1995 12:00:00 GMT">
>
> creates the following HTTP header:
>
> Expires: Sat, 6 May 1995 12:00:00 GMT
>
> However, AFAIK, there is no such thing as a `Refresh' header.
>
> I am pretty sure that this is Netsc[r]ape-spectific.  I haven't found it
> in the IETF HTTP drafts, and I have to agree with Ulf Kronman that such a
> behaviour is evil, rude and nasty.

This header was suggested by other members of the www-talk list
back in December or January so we implemented it.

It took more than a year for FORMS to make it into any spec, in fact
they still are not in any official spec, so it's no suprise that
the Refresh header isn't in the HTTP spec yet.

>
> This means that Netscrape does not only add extensions to their
> browser, but also to their *server* which is much, much worse!

You should do a little more research.  Extra headers are added by
cgi scripts and require no server modifications.  The refresh header
can be added to any HTTP server.

:lou


-- 
Lou Montulli                 http://www.mcom.com/people/montulli/
       Netscape Communications Corp.
eturn-Path: hartill@ooo.lanl.gov 
Return-Path: <hartill@ooo.lanl.gov>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA05185; Mon, 1 May 1995 16:42:35 +0500
Received: from dxmint.cern.ch by www10.w3.org (5.0/NSCS-1.0S) 
	id AA08861; Mon, 1 May 1995 16:40:20 +0500
Received: from ooo.lanl.gov by dxmint.cern.ch (5.65/DEC-Ultrix/4.3)
	id AA20912; Mon, 1 May 1995 22:40:12 +0200
Received: by ooo.lanl.gov
	(1.37.109.16/16.2) id AA274290807; Mon, 1 May 1995 14:40:07 -0600
From: Rob Hartill <hartill@ooo.lanl.gov>
Message-Id: <199505012040.AA274290807@ooo.lanl.gov>
Subject: centering proposal
To: www-talk@w3.org
Date: Mon, 1 May 95 14:40:06 MDT
X-Organization: Theoretical Division, T-8. Los Alamos National Laboratory
X-Snail: LANL Theoretical Divi' T-8, MS B285, P.O Box 1663, Los Alamos NM 87545
X-Fax: (505) 667 5585
X-Phone: (505) 665-2280 or 667-5336 (T-8 Secretary)
Reply-To: hartill@lanl.gov
Mailer: Elm [revision: 70.85]
Content-Length: 662


Maybe someone out there with influence in the development of HTML
can do something with this...

It'd be useful to have a means of centering a section of text relative
to a specific point. e.g.

text on left ... more text on the right side

could be displayed as,

                       text on left ... more text on the right side

maybe by using tags such as,

<P ALIGN="CENTERED">text on left <MIDPOINT>...</MIDPOINT> more text on the right side</P>

The only way to do this right now is with tables, and that makes it
unfriendly w.r.t non-tables browsers. At least with the above, the text
would remain visible, even if it isn't displayed centered.


rob.
eturn-Path: bert%let.rug.nl@www10.w3.org 
Return-Path: <bert%let.rug.nl@www10.w3.org>
Received: from mintaka.lcs.mit.edu by www19 (5.0/NSCS-1.0S) 
	id AD07467; Mon, 1 May 1995 17:27:48 +0500
Received: from www10.w3.org by MINTAKA.LCS.MIT.EDU id aa14712;
          1 May 95 15:27 EDT
Received: from freya.let.rug.nl by www10.w3.org (5.0/NSCS-1.0S) 
	id AA08272; Mon, 1 May 1995 15:20:16 +0500
Received: from grid.let.rug.nl by freya.let.rug.nl with ESMTP
	(1.37.109.15/16.2) id AA124286006; Mon, 1 May 1995 21:20:06 +0200
Message-Id: <199505011920.AA124286006@freya.let.rug.nl>
Received: by grid.let.rug.nl
	(1.37.109.16/16.2) id AA033196005; Mon, 1 May 1995 21:20:05 +0200
From: Bert Bos <bert@let.rug.nl>
Subject: Re: URLs in saved documents
To: martin@mrrl.lut.ac.uk
Date: Mon, 1 May 1995 21:20:05 +0200 (METDST)
Cc: www-talk@www10.w3.org
In-Reply-To: <199504291334.OAA03658@pride.mrrl.lut.ac.uk> from "Martin Hamilton" at Apr 29, 95 09:36:45 am
X-Mailer: ELM [version 2.4 PL24]
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Content-Length: 1476      

Martin Hamilton writes:

 |It would be nice if WWW browsers gave you the option of including the 
 |URLs of hyperlinks in the saved versions of HTML documents (perhaps 
 |some do already?).  For instance, an easy way of doing this would be 
 |for
 |
 |  <A HREF="http://www.acme.com/">Acme Widgets, Inc</A>
 |
 |to become
 |
 |  Acme Widgets, Inc <URL:http://www.acme.com/>
 |
 |in a saved (say) text or PostScript version of the document.  There 
 |are probably quite a few more, and fancier, ways of doing this - like 
 |footnotes, or an automagically generated list of references at the 
 |end of the document ?

The answer is, again, style sheets: using the notation of my own
proposal -- soon to be replaced with a better one -- you could express
this as, for example:

  *A.insertafter: &lt;!HREF&gt;

maining that the formatter inserts text at the end of every A element,
consisting of "<", followed by the value of the element's HREF
attribute, followed by ">".

The current list of proposed style properties is not
complete. Properties applicable to output to anything other than the
screen are missing. One such property is probably one to put text into
a footnote at the bottom of a printed page.

 |Just a thought :-)

Good thought.



Bert
-- 
                          Bert Bos                      Alfa-informatica
                 <bert@let.rug.nl>           Rijksuniversiteit Groningen
    <http://www.let.rug.nl/~bert/>     Postbus 716, NL-9700 AS GRONINGEN
eturn-Path: gwe3409@drtn009.ca.boeing.com 
Return-Path: <gwe3409@drtn009.ca.boeing.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB11787; Mon, 1 May 1995 18:13:41 +0500
Received: from atc.boeing.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA09387; Mon, 1 May 1995 18:11:26 +0500
Received: by atc.boeing.com (5.57) 
	id AA21539; Mon, 1 May 95 14:56:18 -0700
Message-Id: <9505012156.AA21539@atc.boeing.com>
Received: from drtn009.ca.boeing.com by splinter.boeing.com with SMTP
	(1.37.109.14/16.2) id AA099714986; Mon, 1 May 1995 14:49:46 -0700
Received: by drtn009.ca.boeing.com
	(1.37.109.8/16.2) id AA04045; Mon, 1 May 1995 14:51:09 -0700
From: Gerald W. Edgar <gwe3409@drtn009.ca.boeing.com>
Subject: Re: Private CGI-dir - a security risk
To: www-talk@www10.w3.org
Date: Mon, 1 May 1995 14:51:09 -0700 (PDT)
Cc: wendy@drtn001.ca.boeing.com, cingalls@drtn001.ca.boeing.com
In-Reply-To: <9504292302.AA12734@atc.boeing.com> from "www-talk@www10.w3.org" at Apr 29, 95 04:02:36 pm
X-Mailer: ELM [version 2.4 PL23]
Content-Type: text
Content-Length: 650       


In one note  about CGI from Vidar Madsen a mension that a CGI may 
overwrite files that the "webmaster" account owns. This may include 
configuration files. 

There is a simple solution. Have the httpd run under a second 
userid/groupid. 

Permission could be given to read the configuration 
file, but since the daemon executes under another account
it would not have permission by default to destroy the files.

In this situation one must be careful to give appropriate 
permission to directories and files for public read and execute
permission only as needed. One must exercize caution in giving write 
privilages.

Gerald Edgar
"My opinions"


eturn-Path: drtr1@cus.cam.ac.uk 
Return-Path: <drtr1@cus.cam.ac.uk>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA18300; Tue, 2 May 1995 06:36:58 +0500
Received: from bootes.cus.cam.ac.uk by www10.w3.org (5.0/NSCS-1.0S) 
	id AA14576; Tue, 2 May 1995 06:36:48 +0500
Received: from grus.cus.cam.ac.uk [131.111.8.3] (ident = root) 
	by bootes.cus.cam.ac.uk with smtp 
	(Smail-3.1.29.0 #36) id m0s6FIf-000BzaC; Tue, 2 May 95 11:35 BST
Received: by grus.cus.cam.ac.uk 
	(Smail-3.1.29.0 #36) id m0s6FIe-0007b0C; Tue, 2 May 95 11:35 BST
Message-Id: <m0s6FIe-0007b0C@grus.cus.cam.ac.uk>
Date: Tue, 2 May 95 11:35 BST
Sender: drtr1@cus.cam.ac.uk (David Robinson)
To: www-talk@www10.w3.org
Subject: Re: CGI spec revisited
Cc: drtr1@cam.ac.uk, paulp@cerf.net
From: drtr1@cam.ac.uk (David Robinson)
Reply-To: drtr1@cam.ac.uk (David Robinson)
Content-Length: 475

On Thu, 27 Apr 1995 01:27:42 +0500 Paul Phillips wrote:
>Is there a more definitive copy of the CGI 1.1 spec than the one at
>hoohoo.ncsa.uiuc.edu? It has more the sense of a tutorial than a rigorous
>specification.

I have attempted to write such a specification. So for, I have defined the
base spec, without any operating-system specifics. For example,
PATH_TRANSLATEED is an optional feature.

The URL is <URL:http://www.ast.cam.ac.uk/%7Edrtr/cgi.html>

 David Robinson.
eturn-Path: wmperry@spry.com 
Return-Path: <wmperry@spry.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB15509; Tue, 2 May 1995 09:32:51 +0500
Received: from monolith (monolith.spry.com) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA16313; Tue, 2 May 1995 09:32:45 +0500
Message-Id: <m0s6I4i-00000EC@monolith>
Date: Tue, 2 May 95 06:33 PDT
From: wmperry@spry.com
To: hartill@ooo.lanl.gov
Errors-To: wmperry@spry.com
Reply-To: wmperry@spry.com
X-Face: O~Rn;(l][/-o1sALg4A@xpE:9-"'IR[%;,,!m7</SYF`{vYQ(&RI1&EiH[FvT;J}@f!4kfz
 x_!Y#=y{Uuj9GvUi=cPuajQ(Z42R[wE@{G,sn$qGr5g/wnb*"*ktI+,CD}1Z'wxrM2ag-r0p5I6\nA
 [WJopW_J.WY;
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Re: centering proposal
In-Reply-To: <199505012040.AA274290807@ooo.lanl.gov>
References: <199505012040.AA274290807@ooo.lanl.gov>
Content-Length: 813

Rob Hartill writes:
> 
> Maybe someone out there with influence in the development of HTML
> can do something with this...
> 
> It'd be useful to have a means of centering a section of text relative
> to a specific point. e.g.
> 
> text on left ... more text on the right side
> 
> could be displayed as,
> 
>                        text on left ... more text on the right side
> 
> maybe by using tags such as,
> 
> <P ALIGN="CENTERED">text on left <MIDPOINT>...</MIDPOINT> more text on the right side</P>
> 
> The only way to do this right now is with tables, and that makes it
> unfriendly w.r.t non-tables browsers. At least with the above, the text
> would remain visible, even if it isn't displayed centered.

  I believe you can do this with 

<tab align=right> in the current draft of HTML 3.0.

-Bill P.
eturn-Path: rick@cts.com 
Return-Path: <rick@cts.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB22088; Tue, 2 May 1995 11:01:34 +0500
Received: from crash.cts.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA01038; Tue, 2 May 1995 11:01:26 +0500
Received: from stout1 by crash.cts.com with smtp
	(Smail3.1.28.1 #23) id m0s6JR6-0000zpC; Tue, 2 May 95 08:00 PDT
Message-Id: <m0s6JR6-0000zpC@crash.cts.com>
Date: Tue, 2 May 95 08:00 PDT
X-Sender: rick@crash.cts.com
X-Mailer: Windows Eudora Version 2.0.3
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
To: hartill@ooo.lanl.gov, Multiple recipients of list <www-talk@www10.w3.org>
From: rick@cts.com (Rick Stout)
Subject: Re: centering proposal
Content-Length: 687

>text on left ... more text on the right side
>
>could be displayed as,
>
>                       text on left ... more text on the right side
>
>maybe by using tags such as,
>
><P ALIGN="CENTERED">text on left <MIDPOINT>...</MIDPOINT> more text on the
right side</P>
>
>The only way to do this right now is with tables, and that makes it
>unfriendly w.r.t non-tables browsers. At least with the above, the text
>would remain visible, even if it isn't displayed centered.


Tabs are supposed to do this in HTML 3.0.

See http://www.hpl.hp.co.uk/people/dsr/html/tabs.html

-Rick Stout
--------------------------------------------------------------------------
-Rick Stout
(rick@cts.com)

eturn-Path: charlab@embratel.net.br 
Return-Path: <charlab@embratel.net.br>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA15937; Tue, 2 May 1995 13:10:42 +0500
Received: from rjo04 ([200.255.253.244]) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA02492; Tue, 2 May 1995 13:10:34 +0500
Received: from 200.255.254.249 (drjo022A249.embratel.net.br [200.255.254.249]) by rjo04 (8.6.12/8.6.9) with SMTP id OAA06965 for <www-talk@www10.w3.org>; Tue, 2 May 1995 14:10:26 -0300
Date: Tue, 2 May 1995 14:10:26 -0300
Message-Id: <199505021710.OAA06965@rjo04>
From: charlab@embratel.net.br
Subject: Avoiding a blank line.
To: www-talk@www10.w3.org
X-Mailer: AIR Mail 3.X (SPRY, Inc.)
Content-Length: 614

Hi folks,

If I write according to specifications, and use <P ALIGN=CENTER>
instead of <CENTER>, I get things centered, but also get a
blank line, like here:

<HR ALIGN=CENTER SIZE=5 WIDTH=80%>
<I><P ALIGN=CENTER>2 de maio</I>

How can I avoid this line? I would prefer to have the HR line and
the text closer. Also, how could I define a blank line for browsers
other than Netscape, which seems one of the few to accept <P> as
blank lines, like here:

<P>
<P>

_______________________________
c    h    a    r    l    a    b

Sergio Charlab
Jornal do Brasil
*JB no WEB* --- > http://www.ibase.br/~jb/index.html 


eturn-Path: hartill@ooo.lanl.gov 
Return-Path: <hartill@ooo.lanl.gov>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA22665; Tue, 2 May 1995 14:20:41 +0500
Received: from ooo.lanl.gov by www10.w3.org (5.0/NSCS-1.0S) 
	id AA03201; Tue, 2 May 1995 14:20:39 +0500
Received: by ooo.lanl.gov
	(1.37.109.16/16.2) id AA132998826; Tue, 2 May 1995 12:20:26 -0600
From: Rob Hartill <hartill@ooo.lanl.gov>
Message-Id: <199505021820.AA132998826@ooo.lanl.gov>
Subject: Re: centering proposal
To: www-talk@www10.w3.org
Date: Tue, 2 May 95 12:20:25 MDT
In-Reply-To: <m0s6JR6-0000zpC@crash.cts.com>; from "Rick Stout" at May 2, 95 8:00 am
X-Organization: Theoretical Division, T-8. Los Alamos National Laboratory
X-Snail: LANL Theoretical Divi' T-8, MS B285, P.O Box 1663, Los Alamos NM 87545
X-Marks-The-Spot: Doh !
Reply-To: hartill@lanl.gov
Mailer: Elm [revision: 70.85]
Content-Length: 260



Re: centering about a given point..
 
> Tabs are supposed to do this in HTML 3.0.
> 
> See http://www.hpl.hp.co.uk/people/dsr/html/tabs.html

Are there any browsers which handle "tabs" yet ?
I'd like to experiment. I just tried it with Arena - no luck.

rob
eturn-Path: wmperry@spry.com 
Return-Path: <wmperry@spry.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA23830; Tue, 2 May 1995 14:38:52 +0500
Received: from monolith (monolith.spry.com) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA03385; Tue, 2 May 1995 14:38:51 +0500
Message-Id: <m0s6Mqw-00000HC@monolith>
Date: Tue, 2 May 95 11:39 PDT
From: wmperry@spry.com
To: hartill@ooo.lanl.gov
Errors-To: wmperry@spry.com
Reply-To: wmperry@spry.com
X-Face: O~Rn;(l][/-o1sALg4A@xpE:9-"'IR[%;,,!m7</SYF`{vYQ(&RI1&EiH[FvT;J}@f!4kfz
 x_!Y#=y{Uuj9GvUi=cPuajQ(Z42R[wE@{G,sn$qGr5g/wnb*"*ktI+,CD}1Z'wxrM2ag-r0p5I6\nA
 [WJopW_J.WY;
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Re: centering proposal
In-Reply-To: <199505021820.AA132998826@ooo.lanl.gov>
References: <199505021820.AA132998826@ooo.lanl.gov>
Content-Length: 409

Rob Hartill writes:
> 
> 
> Re: centering about a given point..
>  
> > Tabs are supposed to do this in HTML 3.0.
> > 
> > See http://www.hpl.hp.co.uk/people/dsr/html/tabs.html
> 
> Are there any browsers which handle "tabs" yet ?
> I'd like to experiment. I just tried it with Arena - no luck.

  Emacs-w3 handles them partially.  Its next on the block for complete
  implementation before tables.

-Bill P.
eturn-Path: rick@cts.com 
Return-Path: <rick@cts.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA28995; Tue, 2 May 1995 15:21:08 +0500
Received: from crash.cts.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA03734; Tue, 2 May 1995 15:21:04 +0500
Received: from stout1 by crash.cts.com with smtp
	(Smail3.1.28.1 #23) id m0s6NVF-0001iKC; Tue, 2 May 95 12:21 PDT
Message-Id: <m0s6NVF-0001iKC@crash.cts.com>
Date: Tue, 2 May 95 12:21 PDT
X-Sender: rick@crash.cts.com
X-Mailer: Windows Eudora Version 2.0.3
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
To: hartill@ooo.lanl.gov, Multiple recipients of list <www-talk@www10.w3.org>
From: rick@cts.com (Rick Stout)
Subject: Re: centering proposal
Content-Length: 549

At 02:28 PM 5/2/95 +0500, Rob Hartill wrote:
>
>
>Re: centering about a given point..
> 
>> Tabs are supposed to do this in HTML 3.0.
>> 
>> See http://www.hpl.hp.co.uk/people/dsr/html/tabs.html
>
>Are there any browsers which handle "tabs" yet ?
>I'd like to experiment. I just tried it with Arena - no luck.
>
>rob

I haven't found one yet.  I've tried Arena, Netscape (Win and X), 
Mosaic (Win and X),and WinWeb. No go on these either.

-Rick
--------------------------------------------------------------------------
-Rick Stout
(rick@cts.com)

eturn-Path: alice@datalib.ubc.ca 
Return-Path: <alice@datalib.ubc.ca>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB29256; Tue, 2 May 1995 15:22:56 +0500
Received: from hub.ubc.ca by www10.w3.org (5.0/NSCS-1.0S) 
	id AA03760; Tue, 2 May 1995 15:22:48 +0500
Received: from  datalib.ubc.ca by hub.ubc.ca (4.1/1.14)
	id AA10391; Tue, 2 May 95 12:22:46 PDT
Received: by  datalib.ubc.ca (5.0/SMI-SVR4)
	id AA25476; Tue, 2 May 1995 12:14:40 -0700
Date: Tue, 2 May 1995 12:03:53 -0700 (PDT)
From: Alice Iordache <alice@datalib.ubc.ca>
Subject: Httpd-log files
To: www-talk@www10.w3.org
Message-Id: <Pine.3.85.9505021253.A25428-0100000@mccoy>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 849

I'm using getstats for my monthly statistics in Data Library. For using 
this program I have to specify the log file which for WWW is httpd-log. 
This file used to include all the accesses starting with the earliest 
date and ending at the most recent one - all the information in one file. 
Some time ago it started to break the information and automatically 
create a separate httpd-log file for each week or even worse for one day.
Is there something I can do to prevent that and what is the reason of 
suddenly creating these files. Thank you for your help!

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Alice Iordache             Internet:  alice@datalib.ubc.ca
UBC Data Library           
Data Services Assistant    Phone: (604)822-5587    
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -

                  



eturn-Path: brian@organic.com 
Return-Path: <brian@organic.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AC14566; Tue, 2 May 1995 17:54:14 +0500
Received: from eat.organic.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA05109; Tue, 2 May 1995 17:54:13 +0500
Received: (from brian@localhost) by eat.organic.com (8.6.11/8.6.9) id OAA23772; Tue, 2 May 1995 14:54:02 -0700
Date: Tue, 2 May 1995 14:54:01 -0700 (PDT)
From: Brian Behlendorf <brian@organic.com>
Subject: Re: Session tracking
To: Kee Hinckley <nazgul@utopia.com>
Cc: Multiple recipients of list <www-talk@www10.w3.org>
In-Reply-To: <v02110137abcc1310d4c6@[204.57.39.6]>
Message-Id: <Pine.3.89.9505021452.X27478-0100000@eat.organic.com>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 4910

On Tue, 2 May 1995, Kee Hinckley wrote:
> >> It also needs to be possible to
> >> defeat cacheing - and in that respect I see the cookie proposal as very
> >> much akin to the standard name/password mechanism.
> >
> >Er, I hope that's a typo.  Any proposal made from here on out needs to be
> >able to support caching to some extent - publishing on the web is going
> 
> Not a typo. And I'm not saying that caching shouldn't be possible, just
> that there need to be times when I want to defeat it.  

Agreed - I was saying that any proposal for web mechanisms these days 
must not make caching impossible, so we're not conflicting here.  The 
problem with the cookie proposal was that it made proxy-defeating the 
*norm*. 

> It's necessary when
> I provide a page that is tailored in some way for a particular user.

Yes - if someone is connecting to a financial company and wants to see how
their portfolio is doing, sure, that can't be cached (it'll most likely be
encrypted anyways).  But the question is if you should base that a
Session-ID or a real user-password system.  I would vote for the latter.
The real solution is one where the financial company sends the investor a
portfolio-app (Java or otherwise) which queries dowjones.com via HTTP (or
better yet, dowjones.com broadcasts stats via multicasting, which the
portfolio app listens to giving you info real time on your portfolio).  

But the debate here is whether using Session-ID is appropriate for 
tailoring information.... and I still say no, because I think for every 
application where someone can show it's useful, there's a much better 
long term solution.

> >In fact, I'd support language in the spec that said "it is strongly
> >recommended that servers not generate content that depends on a
> >particular session-ID value" - what would be ideal is if proxy caches
> 
> We differ fundamentally there. I see the Web as a distributed application
> mechanism with the ability to personalize pages to any individual's taste.
> Generating session-id specific content is one of my primary goals.

Yes.  Distributed.  Absolutely.  Custom-creating content for every access
is the opposite.  Distribution works *best* when the objects being flung
around the net are as generic as possible (thus highly cacheable), and the
customization happens as close to the client as possible.  Make the 
clients just as smart, if not smarter than, the servers.  

I will admit this doesn't solve content-provider's problems and needs 
right now.  Those of us who want to provide custom content really do 
have to rely on password-based mechanisms or URL hacks for now - but I 
don't think the tool-builders need to waste their time with short term 
solutions which will just make the content-providers suffer longer (if a 
little less).

> >did forward GET If-modified-since requests, *with* the session-ID's, so
> >servers who want it can still get good data without having to dish out
> >the whole file for every request.  HTTP-NG sounds like it'll have a way
> 
> This gets into where I feel session-ids overlap with security. I'd like not
> to serve certain pages to a person with a different session-id. And perhaps
> not even let them be cached. 

So *don't* use Session-ID's for sensitive information!  There are lots of
other security work going on, the last thing we need is a parallel
mechanism. 

Besides, Session-Id *can't* provide authentication information by itself, 
since it's the server issuing them.  The server could issue them when a 
page is accessed via usual HTTP user authentication of course, but the 
thought was to eliminate that.

> >Agreed... which is why I don't think they should carry a session-ID for
> >every store they visit either.  :)
> 
> That gets to the one major advantage of client-generated session-ids. Of
> course then one needs to determine how to generate a unique user id. There
> are two options that immediately come to mind. Email addresses work pretty
> well (that's what we use to search for an ID if someone loses theirs - we
> just mail the result of the search to them - First Virtual-style security
> :-). The most useful option though would really be a public-key.  If every
> user had (a unique) one (or every browser, if must be) then we would have
> an *extremely* useful session-id. If nothing else, it would give the Feds
> nightmares!

Again, *authentication* is entirely separate and should be out of the 
scope of Session-ID.  Eventually, yes, public-key crypto will allow 
people to authenticate themselves without requiring huge user databases 
on the server end, but that's not quite the same issue here.  Server X 
should be able to discern my path through their service without getting 
any personal info about me. 

	Brian

--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--
brian@organic.com  brian@hyperreal.com  http://www.[hyperreal,organic].com/

eturn-Path: clapp@engr.orst.edu 
Return-Path: <clapp@engr.orst.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA19425; Tue, 2 May 1995 22:23:21 +0500
Received: from engr.orst.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA07509; Tue, 2 May 1995 22:23:15 +0500
Received: from zero (zero-atm.ENGR.ORST.EDU [198.106.199.103]) by engr.orst.edu (8.6.10/8.6.10) with SMTP id TAA02978 for <www-talk@www10.w3.org>; Tue, 2 May 1995 19:22:48 -0700
Received: by zero (5.x/ENGR-Client)
	id AA01059; Tue, 2 May 1995 19:21:27 -0700
Message-Id: <9505030221.AA01059@zero>
To: www-talk@www10.w3.org
From: "Andrew S. Clapp" <clapp@engr.orst.edu>
Date: Tue, 02 May 1995 19:21:26 -0700
Sender: clapp@engr.orst.edu
Content-Length: 1740

Where could I find a list of these codes?
-ASC

 > Mike Meyer wrote:
 > 
 > >> Can someone explain where one should use a 403 response versus a 400
 > >> response? Is using 400 only for mailformed requests, and 400 for
 > >> requests with a command that isn't understood a reasonable
 > >> interpretation?
 > 
 > and Paul Phillips responded:
 > 
 > > My spec indicates that 403 implies greater server understanding than 400 
 > > does.  A 403 means the server tried to service the request, and failed, 
 > > while a 400 means that the server knew based on the request that it would 
 > > fail.
 > 
 > Ummmm, almost.  400 Bad Request indicates that the server was unable
 > to understand the request due to it being malformed.  403 Forbidden
 > indicates that the server *did* understand the request, but refuses to
 > service it for some reason that remains unknown to the client.
 > 
 > > There does seem to be some abiguity here, but both codes instruct the 
 > > client not to repeat the request, so I don't think it's critical.
 > 
 > There is a certain amount of overlap between 400 and all 4xx responses,
 > but I don't consider that to be ambiguous.  I'll change the spec so
 > that the purpose of the two codes is clarified.
 > 
 > Hmmmm, I could just change the example Reason Phrases to
 > 
 >      400 You screwed up
 >      403 Piss off
 > 
 > ;-)
 > 
 >  ....Roy T. Fielding  Department of ICS, University of California, Irvine USA
 >                                        <fielding@ics.uci.edu>
 >                       <URL:http://www.ics.uci.edu/dir/grad/Software/fielding>

Andrew S. Clapp  &_   ______  /\  /\  /\  ______\  Oregon State University
  www.nero.net  O7/O        \/  \/  \/  \/      /  COE Computer Services
eturn-Path: namita@henna.iitd.ernet.in 
Return-Path: <namita@henna.iitd.ernet.in>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB01834; Wed, 3 May 1995 00:39:28 +0500
Received: from sangam.ncst.ernet.in by www10.w3.org (5.0/NSCS-1.0S) 
	id AA08416; Wed, 3 May 1995 00:39:23 +0500
Received: (from uucp@localhost) by sangam.ncst.ernet.in (8.6.8.1/8.6.6) with UUCP id KAA23419 for www-talk@www10.w3.org; Wed, 3 May 1995 10:09:35 +0530
Received: from niyati.ernet.in by henna.iitd.ernet.in (4.1/SMI-4.1-MHS-7.0
 )
	id AA05853; Wed, 3 May 95 09:49:10 IST
X-Organisation: Indian Institute of Technology, New Delhi.
Received: by niyati.ernet.in (5.x/SMI-SVR4)
	id AA00533; Wed, 3 May 1995 09:47:42 +0530
Date: Wed, 3 May 1995 09:47:42 +0530
From: namita@henna.iitd.ernet.in (Namita)
Message-Id: <9505030417.AA00533@niyati.ernet.in>
To: www-talk@www10.w3.org
X-Sun-Charset: US-ASCII
Content-Length: 10

subscribe
eturn-Path: verga@ercole.cefriel.it 
Return-Path: <verga@ercole.cefriel.it>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA26702; Wed, 3 May 1995 04:55:27 +0500
Received: from mailer.cefriel.it (ercole.cefriel.it) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA10242; Wed, 3 May 1995 04:55:23 +0500
Received: from punto.cefriel.it by mailer.cefriel.it (4.1/SMI-4.1)
	id AA18267; Wed, 3 May 95 10:51:19+010
Received: by punto.cefriel.it (4.1/SMI-4.1)
	id AA15332; Wed, 3 May 95 09:52:07+010
From: verga@ercole.cefriel.it (Alberto Verga)
Message-Id: <9505030852.AA15332@punto.cefriel.it>
Subject: Re: HTML 2.0/3.o
To: amosley@pubspo.hq.af.mil
Date: Wed, 3 May 1995 09:52:06 +0100 (GMT+0100)
Cc: www-talk@www10.w3.org
In-Reply-To: <2FA4FFFC@pubspo.hq.af.mil> from "Mosley, Art" at May 1, 95 09:36:01 am
X-Mailer: ELM [version 2.4 PL23]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 697       

> Does anyone know about HTML 2.0/3.0 - - SGML directly on the Internet? 
>  Where I could how it works or how to get information?
> 
> Thanks
> Art
> 
Here!

http://www.w3.org/hypertext/WWW/MarkUp/MarkUp.html  for specs
http://www.w3.org/hypertext/WWW/Tools/Overview.html  for tools

These are good starting points....

-- 
----------------------------------------------------------------------
Alberto Verga                         e-mail : verga@mailer.cefriel.it

CEFRIEL - Politecnico di Milano      
Via Emanueli, 15		      voice  :          +39-2-66100083
20126 Milano (Italy)		      fax    :          +39-2-66100448 
----------------------------------------------------------------------

eturn-Path: Emmanuel.Ponsardin@edelweb.fr 
Return-Path: <Emmanuel.Ponsardin@edelweb.fr>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA27157; Wed, 3 May 1995 04:58:29 +0500
Received: from edelweb.fr by www10.w3.org (5.0/NSCS-1.0S) 
	id AA10282; Wed, 3 May 1995 04:58:25 +0500
Received: from champagne.edelweb.fr (champagne.edelweb.fr [193.51.12.33]) by edelweb.fr (8.6.10/8.6.9) with ESMTP id KAA10950 for <www-talk@www10.w3.org>; Wed, 3 May 1995 10:58:23 +0200
Received: from moet.edelweb.fr (moet.edelweb.fr [193.51.12.38]) by champagne.edelweb.fr (8.6.10/8.6.6) with ESMTP id KAA12747 for <www-talk@www10.w3.org>; Wed, 3 May 1995 10:58:23 +0200
From: Emmanuel Ponsardin <Emmanuel.Ponsardin@edelweb.fr>
Received: (ponsardi@localhost) by moet.edelweb.fr (8.6.10/8.6.6) id KAA00493 for www-talk@www10.w3.org; Wed, 3 May 1995 10:58:22 +0200
Date: Wed, 3 May 1995 10:58:22 +0200
Message-Id: <199505030858.KAA00493@moet.edelweb.fr>
To: www-talk@www10.w3.org
Subject: imagemaps
Content-Length: 392

I use imagemaps on my web pages (imagemap.c v1.2) and i have some trouble:

- first of all, my error_log file (ncsa web server) is full of "killing CGI
  process XXXX". Why?

- and as i test my pages on local network, when i click on my imagemap, it s
  very long to connect to the url requested... are all imagemap scripts so
  unperformants?

Thanks for your help...


Emmanuel Ponsardin  
eturn-Path: brianm@itthartford.com 
Return-Path: <brianm@itthartford.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA04391; Wed, 3 May 1995 05:58:29 +0500
Received: from interlock.itthartford.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA10854; Wed, 3 May 1995 05:58:26 +0500
Received: by interlock.itthartford.com id AA02273
  (InterLock SMTP Gateway 3.0 for www-talk@www10.w3.org);
  Wed, 3 May 1995 05:58:12 -0400
Message-Id: <199505030958.AA02273@interlock.itthartford.com>
Received: by interlock.itthartford.com (Internal Mail Agent-1);
  Wed, 3 May 1995 05:58:12 -0400
To: www-talk@www10.w3.org
Subject: Support for Multiple Submit buttons on Forms 
Date: Tue, 02 May 95 06:06:56 -0400
From: brianm@itthartford.com
X-Mts: smtp
Content-Length: 288


Does there exist a way to support multiple submit buttons within a single form 
????. I would like to have the executable associated with the form take a 
different logic path based on which button is selected... eg. help button. 
Thanks in advance for your help...     	 
						Brian


eturn-Path: lilley@afs.mcc.ac.uk 
Return-Path: <lilley@afs.mcc.ac.uk>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA07234; Wed, 3 May 1995 06:22:22 +0500
Received: from norse.mcc.ac.uk by www10.w3.org (5.0/NSCS-1.0S) 
	id AA11059; Wed, 3 May 1995 06:22:19 +0500
Received: from afs.mcc.ac.uk (actually cguhpc.cgu.mcc.ac.uk) by norse.mcc.ac.uk 
          with SMTP (PP); Wed, 3 May 1995 11:21:47 +0100
From: lilley <lilley@afs.mcc.ac.uk>
Message-Id: <1362.9505031021@afs.mcc.ac.uk>
Subject: Re: HTML 2.0/3.o
To: mattison@freenet.victoria.bc.ca
Date: Wed, 3 May 1995 11:21:36 +0100 (BST)
Cc: www-talk@www10.w3.org
In-Reply-To: <Pine.2.2.9505010713.A20084@vifa1> from "David Mattison" at May 1, 95 10:51:41 am
Organisation: Computer Graphics Unit, University of Manchester, UK
Phone: +44 0161 275 6045
Fax: +44 0161 275 6040
Operating-System: some HP unix thingy
X-Uri: http://info.mcc.ac.uk/CGU/staff/lilley/lilley.html
X-Mailer: ELM [version 2.4 PL24alpha3]
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Content-Length: 1452

David Mattison said:

> According to Laura Lemay's TEACH YOURSELF WEB PUBLISHING WITH HTML IN A
> WEEK (SAMS Publishing, 1995), the HTML 2 spec is at:
> 
> http://www.hal.com/users/connolly/html-spec/index.html
> 
> The HTML+ 3 draft spec is at:
> 
> http://www.w3.org/hypertext/WWW/MarkUp/HTMLPlus/htmlplus_1.html

Beware anything that refers to HTML+, that is a sure sign it is out of date.

For HTML 2.0:
 http://www.w3.org/hypertext/WWW/MarkUp/html-spec/index.html
 
For HTML 3.0:
 http://www.hpl.hp.co.uk/people/dsr/html/CoverPage.html

For SGML as a MIME type:
 ftp://ifi.uio.no/pub/SGML/MIME-SGML/draft-levinson-sgml-02.txt

--
Chris Lilley
+----------------------------------------------------------------------+
|Technical Author, Manchester and North HPC Training & Education Centre|
+----------------------------------------------------------------------+
| Computer Graphics Unit,             |  Email: Chris.Lilley@mcc.ac.uk |
| Manchester Computing Centre,        |  Voice: +44 61 275 6045        |
| Oxford Road, Manchester, UK.M13 9PL |    Fax: +44 61 275 6040        |
+-------------------------------------+ BioMOO: ChrisL                 |
|       URI: http://info.mcc.ac.uk/CGU/staff/lilley/lilley.html        | 
+----------------------------------------------------------------------+
|       "The first W in WWW will not wait."   François Yergeau         |
+----------------------------------------------------------------------+
eturn-Path: news@reznor.larc.nasa.gov 
Return-Path: <news@reznor.larc.nasa.gov>
Received: from dxmint.cern.ch by www19 (5.0/NSCS-1.0S) 
	id AA06962; Wed, 3 May 1995 10:55:05 +0500
Received: from www0.cern.ch by dxmint.cern.ch
	id AA23749; Wed, 3 May 1995 16:54:57 +0200
Received: from dxmint.cern.ch by www0.cern.ch (5.0/SMI-4.0)
	id AA19465; Wed, 3 May 1995 16:54:54 --100
Received: from reznor.larc.nasa.gov by dxmint.cern.ch
	id AA23722; Wed, 3 May 1995 16:54:52 +0200
Received: by reznor.larc.nasa.gov (8.6.10/server2.4)
	id OAA07016; Wed, 3 May 1995 14:54:48 GMT
To: www-talk@www0.cern.ch
Path: usenet
From: Larry Matthias <l.e.matthias@larc.nasa.gov>
Newsgroups: mail.www-talk
Subject: HTTPD,Informix and the Web
Date: 3 May 1995 14:54:47 GMT
Organization: NASA Langley Research Center, Hampton, VA  USA
Lines: 47
Message-Id: <3o85fn$5df@reznor.larc.nasa.gov>
Nntp-Posting-Host: artos.larc.nasa.gov
Content-Length: 1452

Greetings Fellow Earthlings,

        I am trying to setup a Web site that accesses an Informix
database. I am using the NCSA server and their CGI interface. I have 
written my applications in ESQL/C. Now the tricky part. I can
execute my program at the command line and enter data as HTTPD will
The database is accessed and I can actually see data records 
fly by (amazing my program works). BUT when the HTTPD demon
tries to do this it fails with:

       
        There is problem with your query!
        The following SQL error has occured at open_database_esql!! 
        The Error Code is -457 in database: Database server
        terminated unexpectedly. 
        
        
Ok, so I talked with tech support and we all think it must have 
something to do with the environment that the demon is executing
under. I modified the C code to set the following:

        TBCONFIG
        DBPATH
        INFORMIXDIR
        SQLEXEC

After I set these values I printed them. They look ok, but the
program STILL fails.

Therefore, I am interested in talking to, emailing, faxing, or
whatever to people who have setup web sites that integrate the
HTTPD demon with Informix Online.


Any help that you give would be greatly appreciated.

Larry


--------------------------
Larry Matthias
Lockheed Engineering and Sciences Company
Phone:  (804) 766-9726   --> Yes this old technology still works!
Fax:    (804) 766-9601
Email:  l.e.matthias@larc.nasa.gov


eturn-Path: lilley@afs.mcc.ac.uk 
Return-Path: <lilley@afs.mcc.ac.uk>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA08825; Wed, 3 May 1995 11:06:39 +0500
Received: from norse.mcc.ac.uk by www10.w3.org (5.0/NSCS-1.0S) 
	id AA13725; Wed, 3 May 1995 11:06:30 +0500
Received: from afs.mcc.ac.uk (actually cguhpc.cgu.mcc.ac.uk) by norse.mcc.ac.uk 
          with SMTP (PP); Wed, 3 May 1995 16:05:49 +0100
From: lilley <lilley@afs.mcc.ac.uk>
Message-Id: <2617.9505031505@afs.mcc.ac.uk>
Subject: Re: Avoiding a blank line.
To: charlab@embratel.net.br
Date: Wed, 3 May 1995 16:05:45 +0100 (BST)
Cc: www-talk@www10.w3.org
In-Reply-To: <199505021710.OAA06965@rjo04> from "charlab@embratel.net.br" at May 2, 95 01:22:21 pm
Organisation: Computer Graphics Unit, University of Manchester, UK
Phone: +44 0161 275 6045
Fax: +44 0161 275 6040
Operating-System: some HP unix thingy
X-Uri: http://info.mcc.ac.uk/CGU/staff/lilley/lilley.html
X-Mailer: ELM [version 2.4 PL24alpha3]
Mime-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Transfer-Encoding: 8bit
Content-Length: 1772

Sergio Charlab said:

> If I write according to specifications, and use <P ALIGN=CENTER>
> instead of <CENTER>, I get things centered, but also get a
> blank line, like here:
> 
> <HR ALIGN=CENTER SIZE=5 WIDTH=80%>
> <I><P ALIGN=CENTER>2 de maio</I>

You have nearly written to specification, but not quite. You have an hr, so 
following text starts on a new line. A new paragraph is implied by your text. 
You go into italics. Then you start another new paragraph!

Instead, try
<HR ALIGN=CENTER SIZE=5 WIDTH=80%>
<P ALIGN=CENTER><I>2 de maio</I></P>

> How can I avoid this line? I would prefer to have the HR line and
> the text closer.

The exact spacing depends on how your browser chooses to render this. 

> Also, how could I define a blank line for browsers
> other than Netscape, which seems one of the few to accept <P> as
> blank lines, like here:

I think that browsers are at liberty to coalesce multiple empty paragraphs.
You probably want to use <BR>

--
Chris Lilley, Technical Author
+-------------------------------------------------------------------+
|       Manchester and North HPC Training & Education Centre        |
+-------------------------------------------------------------------+
| Computer Graphics Unit,             Email: Chris.Lilley@mcc.ac.uk |
| Manchester Computing Centre,        Voice: +44 61 275 6045        |
| Oxford Road, Manchester, UK.          Fax: +44 61 275 6040        |
| M13 9PL                            BioMOO: ChrisL                 |
|     URI: http://info.mcc.ac.uk/CGU/staff/lilley/lilley.html       | 
+-------------------------------------------------------------------+
|     "The first W in WWW will not wait."   François Yergeau        |
+-------------------------------------------------------------------+
eturn-Path: riyer@seta.com 
Return-Path: <riyer@seta.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA16775; Wed, 3 May 1995 12:08:27 +0500
Received: from dxmint.cern.ch by www10.w3.org (5.0/NSCS-1.0S) 
	id AA14468; Wed, 3 May 1995 12:08:12 +0500
Received: from seta.com by dxmint.cern.ch
	id AA09169; Wed, 3 May 1995 18:08:07 +0200
Date: Wed, 3 May 95 12:09:27 EDT
From: riyer@seta.com (Ramani K. Iyer)
Received: by seta.com (4.1/3.1.012693-SETA Corporation);
        id AA04997 for www-talk@info.cern.ch; Wed, 3 May 95 12:09:27 EDT
Message-Id: <9505031609.AA04997@seta.com>
To: www-talk@w3.org
Subject: HTTPD server question
Content-Length: 669

I am new to this mailing list and am not sure if this
has been discussed. But here it goes.
    Is it possible to have 2 different names resolve to 
    the same server but point to two different home pages.
     For example : x.y.z should point to page 1 in my sun and
                   address http://a.b.c should point to page 2
                   in my sun.
If yes, how do I do that. We use DNS and I can setup CNAME aliases
for my sun so that x.y.z and a.b.c both point to my sun platform. 
How do I make httpd to resolve these different names and serve
the proper home page.
If this is not possible if there a workaround.
Thanks a lot in advance.

- Ramani Iyer
eturn-Path: amosley@pubspo.hq.af.mil 
Return-Path: <amosley@pubspo.hq.af.mil>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA18596; Wed, 3 May 1995 12:26:52 +0500
Received: from hq.af.mil (hq.hq.af.mil) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA14631; Wed, 3 May 1995 12:26:50 +0500
Received: from pubspo.hq.af.mil by hq.af.mil (4.1/Mork-1.0)
	id AA21343; Wed, 3 May 95 12:29:05 EDT
Received: by pubspo.hq.af.mil with Microsoft Mail
	id <2FA7D7F8@pubspo.hq.af.mil>; Wed, 03 May 95 12:22:32 PDT
From: "Mosley, Art" <amosley@pubspo.hq.af.mil>
To: www-talk <www-talk@www10.w3.org>
Subject: Re: HTML 2.0/3.o
Date: Wed, 03 May 95 11:18:00 PDT
Message-Id: <2FA7D7F8@pubspo.hq.af.mil>
Encoding: 34 TEXT
X-Mailer: Microsoft Mail V3.0
Content-Length: 1023


I checked to various web locations...thanks.   Does anyone know or have any 
experience with any SGML Browsers for the internet?  Or about the efforts to 
make SGML browsers for the Internet.

Thanks
 ----------
From: www-talk
To: Multiple recipients of list
Subject: Re: HTML 2.0/3.o
Date:  03 May 95 05:21

> Does anyone know about HTML 2.0/3.0 - - SGML directly on the Internet?
>  Where I could how it works or how to get information?
>
> Thanks
> Art
>
Here!

http://www.w3.org/hypertext/WWW/MarkUp/MarkUp.html  for specs
http://www.w3.org/hypertext/WWW/Tools/Overview.html  for tools

These are good starting points....

 --
 ----------------------------------------------------------------------
Alberto Verga                         e-mail : verga@mailer.cefriel.it

CEFRIEL - Politecnico di Milano
Via Emanueli, 15                      voice  :          +39-2-66100083
20126 Milano (Italy)                  fax    :          +39-2-66100448
 ----------------------------------------------------------------------

eturn-Path: nsb%nsb.fv.com@www10.w3.org 
Return-Path: <nsb%nsb.fv.com@www10.w3.org>
Received: from mintaka.lcs.mit.edu by www19 (5.0/NSCS-1.0S) 
	id AE15602; Wed, 3 May 1995 17:13:46 +0500
Received: from www10.w3.org by MINTAKA.LCS.MIT.EDU id aa04631;
          3 May 95 14:15 EDT
Received: from fv.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA15446; Wed, 3 May 1995 14:11:17 +0500
Received: from nsb.fv.com (guppy.jvnc.net [192.67.238.69]) by fv.com (8.6.10/8.6.10) with SMTP id OAA18590; Wed, 3 May 1995 14:11:16 -0400
Received: by  nsb.fv.com (4.1/SMI-4.1)
	id AA18632; Wed, 3 May 95 14:09:39 EDT
Received: from Messages.8.5.N.CUILIB.3.45.SNAP.NOT.LINKED.nsb.fv.com.sun4.41
          via MS.5.6.nsb.fv.com.sun4_41;
          Wed,  3 May 1995 14:09:38 -0400 (EDT)
Message-Id: <wjdwPWP0Eyt5AmhdYK@nsb.fv.com>
Date: Wed,  3 May 1995 14:09:38 -0400 (EDT)
From: Nathaniel Borenstein <nsb@nsb.fv.com>
To: Multiple recipients of list <www-talk@www10.w3.org>, brian@organic.com
Subject: Re: Session tracking
In-Reply-To: <14749.799495825.34@nsb.fv.com>
References: <14749.799495825.34@nsb.fv.com>
Content-Length: 1967

Excerpts from mail: 3-May-95 Re: Session tracking brian@organic.com (4912)

> > Email addresses work pretty
> > well (that's what we use to search for an ID if someone loses theirs - we
> > just mail the result of the search to them - First Virtual-style security
> :-). 

I know this isn't really a discussion about First Virtual, but I really
have to correct something here.  Not only is this not "First
Virtual-style security", it completely misrepresents how we do business.
 In particular, if one of our users loses their ID, our policy is that
we will NOT email their ID back to them -- in fact, with rare
exceptions, their account is permanently lost in such situations, and
they have to set up a new one.  This is because part (by no means all)
of our transactional security comes from the lack of direct correlation
between an email address and the associated FV ID.  If we let just
anyone forge mail to us saying they'd forgotten their ID, and then use a
sniffer on the resulting traffic back to the real user, we'd open up a
somewhat easier path to fraud than we are willing to tolerate.

You don't have to love our model, but you shouldn't criticize it without
understanding it.  We take security *extremely* seriously, and we have
lots of very happy buyers and sellers who are grateful for it.  (And our
user community and transaction volumes are both growing at a very steady
15% per week, by the way, so we must be doing *something* right.)  --
Nathaniel
--------
Nathaniel S. Borenstein <nsb@fv.com>
Chief Scientist, First Virtual Holdings Incorporated
Phone: +1 201 540-8967  (fax 993-3032)
FREQUENTLY ASKED QUESTIONS (& PGP key):  nsb+faq@nsb.fv.com

-----VIRTUAL YELLOW RIBBON----zldf@clark.net----VIRTUAL YELLOW RIBBON----

> When privacy is outlawed, only   Support the Zimmerman Legal Defense! <
> outlaws will have privacy!       http://www.netresponse.com/zldf      <

-----VIRTUAL YELLOW RIBBON----zldf@clark.net----VIRTUAL YELLOW RIBBON----
eturn-Path: brian@organic.com 
Return-Path: <brian@organic.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA20607; Wed, 3 May 1995 23:30:09 +0500
Received: from eat.organic.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA19442; Wed, 3 May 1995 23:30:07 +0500
Received: (from brian@localhost) by eat.organic.com (8.6.11/8.6.9) id UAA05130; Wed, 3 May 1995 20:30:04 -0700
Date: Wed, 3 May 1995 20:30:03 -0700 (PDT)
From: Brian Behlendorf <brian@organic.com>
Subject: Re: CGI spec revisited
To: Marc Hedlund <marc@precipice.org>
Cc: Multiple recipients of list <www-talk@www10.w3.org>
In-Reply-To: <v02110101abc761d164c5@[38.10.109.32]>
Message-Id: <Pine.3.89.9505032000.a16415-0100000@eat.organic.com>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 4288

On Sat, 29 Apr 1995, Marc Hedlund wrote:
> Is there any interest for a new look at the CGI spec?  I don't just mean at
> NCSA....

Yes, definitely.  Though the limits of the CGI interface may dictate that 
there's only so far we can go with this, at which point a real runtime 
API might be a good thing to look at (like NetScape's).  Perhaps someone 
should start www-servers?

> A few issues to kick around:
> 
> * I like NCSA's DOCUMENT_ROOT idea (which Paul mentions).  A number of
> people have bitched about not being able to reliably determine the document
> root or server root across a variety of servers without asking for help
> from the humans.

I added this many moons ago to my hacked version of httpd, and it found 
its way to the Apache team which is where I presume NCSA picked it up 
from.  Basically I use it because a lot of my site creation has to be 
self-contained, where CGI scripts and libraries and data has to sit 
within one subdirectory.  It's also just good programming to abstract 
away as much as you can, so cases where I had scripts that had to 
assemble a composite page from a bunch of sub objects which were 
themselves accessible individually through the web server had to know 
where to open() files.  

Anyways, if it makes its way into a spec I would be happy to condition it 
on "this is only appropriate for those web servers that employ the 
concept of a document root".  For those web sites that change the file 
system they point to based on IP number or path mapping, the concept of 
the "document root" still holds.

> * Is there any consensus about what should happen to POSTed data if the
> client receives a redirect?  I remember reading somewhere that POSTs get
> turned into GETs if redirected; and a couple of browsers mangle POSTs into
> PATH_INFO (!?!) if passed through a proxy, as I recall.  Why, I ask you,
> why?  Shouldn't POSTs stay POSTs?

Hmm... I think the problem is that the HTTP method is not expressible in 
a URI, so I don't think the browser has any choice but to do a GET on a 
redirect.  I don't think this is a CGI issue.

> * A couple of people have suggested to me hashing out the horrible ACCEPT
> issue in a new CGI spec.  I'm not fond of that idea; I think that's an
> HTTP-wg problem.  However, maybe something could be done to improve the
> amount of information scripts receive from the client, apart from MIME-type
> content negotiation.  If a server and a client are negotiating directly,
> the HTTP spec would govern; if a gateway stands between the two, content
> negotiation can also include the following.... etc.

Negotiation, horrible?  :) Nah, what's needed is a common function that 
the CGI script can call that takes as input the Accept: string and a list 
of possible data types the script can return, and returns the most 
appropriate data type (text/html vs. text/html3 for example).  The server 
can't do this ahead of time because only the CGI script knows what data 
formats it can return.  Hmm - it would seem to me that the server should 
tell the CGI script what its "qs" values are, though, unless it wants to 
handle the q*qs operations on the Accept: string before it gets passed 
to the CGI script.  Comments?

Also for the CGI 1.2 stew: NCSA and Apache diverged unfortunately in the 
issue of CGI variables with internal redirects.  In both, you can point 
to another URI to be accessed when an error occurs - like having an 
access which results in a 404 get pointed to /404.html or even /404.cgi, 
which could potentially return some cool info.  404.cgi needs to know 
some information about the original access to make some intelligent 
decisions, but it would be incorrect to just make its CGI environment 
exactly that of the error-causing access.  Thus, Apache introduces a few 
more variables using REDIRECT_ as the base, and NCSA used ERROR_ as the 
base.  We chose the former as they aren't necessarily the result of an 
error (like 401 responses for example), and when HTTP/1.1 allows us to 
send a Base: header then scripts like imagemap can be made one access 
instead of two using internal redirection.

	Brian

--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--
brian@organic.com  brian@hyperreal.com  http://www.[hyperreal,organic].com/

eturn-Path: zhang@welchgate.welch.jhu.edu 
Return-Path: <zhang@welchgate.welch.jhu.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA21160; Thu, 4 May 1995 10:55:01 +0500
Received: from welchgate.welch.jhu.edu (gateway.welch.jhu.EDU) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA26166; Thu, 4 May 1995 10:54:59 +0500
Received: by welchgate.welch.jhu.edu (4.1/1.34)
	id AA19498; Thu, 4 May 95 10:55:45 EDT
From: zhang@welchgate.welch.jhu.edu (Dongming Zhang)
Message-Id: <9505041455.AA19498@welchgate.welch.jhu.edu>
Subject: MIME type help.
To: www-talk@www10.w3.org
Date: Thu, 4 May 95 10:55:44 EDT
X-Mailer: ELM [version 2.3 PL11]
Content-Length: 449

Greeting,

   I was trying to get one of my local applications launched and was
working on "Helper application" with windows. 

   One question is can I launch application without extension, i.e.
not specifying the file extention?  It was not successful. 
If someone have such experience, I would appreciate it if you can send
some tips to me.

   Dongming Zhang
   Welch Information Technology
   Welch Medical Library
   Johns Hopkins University

eturn-Path: GAYLA.HISS%MSFC34PO@x400gw.msfc.nasa.gov 
Return-Path: <GAYLA.HISS%MSFC34PO@x400gw.msfc.nasa.gov>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA28103; Thu, 4 May 1995 15:00:54 +0500
Received: from dxmint.cern.ch by www10.w3.org (5.0/NSCS-1.0S) 
	id AA29138; Thu, 4 May 1995 15:00:49 +0500
Received: from hpx400.msfc.nasa.gov by dxmint.cern.ch
	id AA10077; Thu, 4 May 1995 21:00:45 +0200
Received: by hpx400.msfc.nasa.gov
	(1.38.193.5/16.2) id AA02249; Thu, 4 May 1995 13:22:45 -0500
From: GAYLA.HISS%MSFC34PO@x400gw.msfc.nasa.gov
Received: by x400gw.msfc.nasa.gov via Worldtalk with X400 (3.0.3/1.55)
	id WT09128.494; Thu, 04 May 1995 13:22:44 CDT
Date: 4 May 95 13:23:00 -0500
To: www-talk@w3.org
Subject: WWW/HTML/HTTP Training
Message-Id: <M933764.001.2gb40.4133.950504182103Z.CC-MAIL*/O=CCMAIL/PRMD=MSFC/ADMD=TELEMAIL/>
Content-Length: 108

         Does anyone there know of any good training courses on the WWW or 
         HTML or HTTP?  Thanks.
eturn-Path: reverman@ka.reg.uci.edu 
Return-Path: <reverman@ka.reg.uci.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB10812; Thu, 4 May 1995 16:51:07 +0500
Received: from ka.reg.uci.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA00306; Thu, 4 May 1995 16:51:05 +0500
Received: from [128.200.148.50] (cybernaut.reg.uci.edu) by ka.reg.uci.edu (4.1/SMI-4.1)
	id AA16565; Thu, 4 May 95 13:50:46 PDT
Date: Thu, 4 May 95 13:50:45 PDT
Message-Id: <9505042050.AA16565@ka.reg.uci.edu>
From: "Richard Everman"  <reverman@ka.reg.uci.edu>
Reply-To: "Richard Everman"  <reverman@uci.edu>
To: www-talk@www10.w3.org
Subject: Are there any browsers which handle "tabs" yet ?
Content-Length: 616

rob

MacWeb 1.00ALPHA3.2 does.

Richard

==================================================

To:      Multiple recipients of list <www-talk@www10.w3.org>

From:    Rob Hartill <hartill@ooo.lanl.gov>
Subject: Re: centering proposal

Sender:  www-talk@www10.w3.org
X-Comment: To sign off, send mail to listproc@mail.w3.org with body DEL WWW-TAL
     ***K



Re: centering about a given point..

> Tabs are supposed to do this in HTML 3.0.
>
> See http://www.hpl.hp.co.uk/people/dsr/html/tabs.html

Are there any browsers which handle "tabs" yet ?
I'd like to experiment. I just tried it with Arena - no luck.

rob




eturn-Path: web@sowebo.charm.net 
Return-Path: <web@sowebo.charm.net>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA06259; Thu, 4 May 1995 19:59:15 +0500
Received: from sowebo.charm.net by www10.w3.org (5.0/NSCS-1.0S) 
	id AA01986; Thu, 4 May 1995 19:59:14 +0500
From: web@sowebo.charm.net
Message-Id: <9505041959.AA09955@sowebo.charm.net>
Received: from web by sowebo.charm.net; Thu,  4 May 95 19:59 EDT
Subject: Re: WWW/HTML/HTTP Training
To: GAYLA.HISS%MSFC34PO@x400gw.msfc.nasa.gov
Date: Thu, 4 May 1995 19:59:47 -0400 (EDT)
>From: "CyberWeb" <web@sowebo.charm.net>
Cc: www-talk@www10.w3.org
In-Reply-To: <M933764.001.2gb40.4133.950504182103Z.CC-MAIL*/O=CCMAIL/PRMD=MSFC/ADMD=TELEMAIL/> from "GAYLA.HISS%MSFC34PO@x400gw.msfc.nasa.gov" at May 4, 95 03:03:31 pm
>From: Dr.Web@Stars.com
Url: http://WWW.Stars.com/
X-Mailer: ELM [version 2.4 PL24]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 574

GAYLA.HISS%MSFC34PO@x400gw.msfc.nasa.gov wrote:
> 
>          Does anyone there know of any good training courses on the WWW or 
>          HTML or HTTP?  Thanks.
> 
	Try	http://WWW.Stars.com/Tutorial/	which is the slides
		I used for the WebWorlds and WWW3 conferences.
	Alan.
      ___________________________________________________________________
      Dr.Web@Stars.com -=*<URL:http://WWW.Stars.com/>*=- 1 (301) 552 0272
      Web Developer's Virtual Library * CyberWeb SoftWare * WWW Databases
      HTML * CGI * Training * Transatlantic Liaison * Per Ardua, Ad Astra
eturn-Path: hemang@bcpsparc.ucdavis.edu 
Return-Path: <hemang@bcpsparc.ucdavis.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB08758; Thu, 4 May 1995 20:44:45 +0500
Received: from franc.ucdavis.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA02220; Thu, 4 May 1995 20:44:42 +0500
Received: from bcpsparc.ucdavis.edu by franc.ucdavis.edu (8.6.12/UCD3.4)
	id RAA26634; Thu, 4 May 1995 17:44:30 -0700
Received: by bcpsparc.ucdavis.edu (4.1/UCD2.03)
	id AA00254; Thu, 4 May 95 17:54:04 PDT
Date: Thu, 4 May 1995 17:54:03 -0700 (PDT)
From: Hemang Patel <hemang@bcpsparc.ucdavis.edu>
To: GAYLA.HISS%MSFC34PO@x400gw.msfc.nasa.gov
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Re: WWW/HTML/HTTP Training
In-Reply-To: <M933764.001.2gb40.4133.950504182103Z.CC-MAIL*/O=CCMAIL/PRMD=MSFC/ADMD=TELEMAIL/>
Message-Id: <Pine.SUN.3.91.950504175306.235C-100000@bcpsparc>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 582

On Thu, 4 May 1995 GAYLA.HISS%MSFC34PO@x400gw.msfc.nasa.gov wrote:

>          Does anyone there know of any good training courses on the WWW or 
>          HTML or HTTP?  Thanks.
> 

Start here:

http://www-mcb.ucdavis.edu/people/hemang/bookmarks/www.html


I have links to all kinds of things:

Learn to write html
Learn to write CGIs
Server software & setup tools
and more

__________________________________________________
Hemang Patel
Section of Molecular and Cellular Biology
Univ. of Ca. Davis
hemang@bcpsparc.ucdavis.edu
http://www-mcb.ucdavis.edu/people/hemang/home.html

eturn-Path: pp%plymouth.ac.uk@www10.w3.org 
Return-Path: <pp%plymouth.ac.uk@www10.w3.org>
Received: from mintaka.lcs.mit.edu by www19 (5.0/NSCS-1.0S) 
	id AJ18367; Fri, 5 May 1995 12:45:00 +0500
Received: from www10.w3.org by MINTAKA.LCS.MIT.EDU id aa24264; 5 May 95 5:40 EDT
Received: from sun2.nsfnet-relay.ac.uk by www10.w3.org (5.0/NSCS-1.0S) 
	id AB05793; Fri, 5 May 1995 05:36:34 +0500
Message-Id: <9505050936.AB05793@www10.w3.org>
Via: uk.ac.plymouth; Fri, 5 May 1995 10:03:44 +0100
Received: from plymouth.ac.uk by argus.plymouth.ac.uk 
          id <02833-0@argus.plymouth.ac.uk>; Fri, 5 May 1995 10:02:11 +0100
From: pp@plymouth.ac.uk
To: www-talk@www10.w3.org
Date: Fri, 5 May 1995 10:02:05 +0100
Subject: WARNING: message delayed at "plymouth.ac.uk"
Content-Length: 377

  Your message to "s.barnes@plymouth.ac.uk" (Message Id: msg.05984-0)
  is currently awaiting delivery by argus.plymouth.ac.uk

  There is a delay in delivering the message, which, if not
  resolved, will be returned to you in 63 hours.

  No further action is required from you at this time since
  the mail system will automatically attempt to deliver the
  message for you.
eturn-Path: pp%plymouth.ac.uk@www10.w3.org 
Return-Path: <pp%plymouth.ac.uk@www10.w3.org>
Received: from mintaka.lcs.mit.edu by www19 (5.0/NSCS-1.0S) 
	id AV29917; Fri, 5 May 1995 14:25:02 +0500
Received: from www10.w3.org by MINTAKA.LCS.MIT.EDU id aa24712; 5 May 95 6:23 EDT
Received: from sun2.nsfnet-relay.ac.uk by www10.w3.org (5.0/NSCS-1.0S) 
	id AA06110; Fri, 5 May 1995 06:18:49 +0500
Message-Id: <9505051018.AA06110@www10.w3.org>
Via: uk.ac.plymouth; Fri, 5 May 1995 10:45:57 +0100
Received: from plymouth.ac.uk by argus.plymouth.ac.uk 
          id <05177-0@argus.plymouth.ac.uk>; Fri, 5 May 1995 10:44:31 +0100
From: pp@plymouth.ac.uk
To: www-talk@www10.w3.org
Date: Fri, 5 May 1995 10:44:24 +0100
Subject: WARNING: message delayed at "plymouth.ac.uk"
Content-Length: 377

  Your message to "s.barnes@plymouth.ac.uk" (Message Id: msg.05285-0)
  is currently awaiting delivery by argus.plymouth.ac.uk

  There is a delay in delivering the message, which, if not
  resolved, will be returned to you in 59 hours.

  No further action is required from you at this time since
  the mail system will automatically attempt to deliver the
  message for you.
eturn-Path: frystyk@w3.org 
Return-Path: <frystyk@w3.org>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB10753; Fri, 5 May 1995 15:40:19 +0500
Received: from www20 (www20.w3.org) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA10327; Fri, 5 May 1995 15:40:14 +0500
Received: by www20 (5.0/NSCS-1.0S) 
	id AA15259; Fri, 5 May 1995 15:40:13 +0500
Date: Fri, 5 May 1995 15:40:13 +0500
From: frystyk@w3.org (Henrik Frystyk Nielsen)
Message-Id: <9505051940.AA15259@www20>
To: www-talk@w3.org, www-lib@w3.org
Subject: Strange problem with GMT on Solaris
Reply-To: Henrik Frystyk Nielsen <frystyk@w3.org>
X-Sun-Charset: US-ASCII
Content-Length: 774


I have a strange problem on Solaris:

The gmtime() function on Solaris seems to return the wrong value. It is one
hour ahead of GMT which means that all date header values in a HTTP request
or response is one hour ahead of GMT. This is for example the case on our
www.w3.org servers which certainly confuses a lot of caches!

It is simple to test:

	time_t calendar = time(NULL);
	struct tm *gmt = gmtime(&calendar);
    	strftime(buf, 40, "%a, %d %b %Y %H:%M:%S GMT", gmt);

Has anybody else heard of this problem?


-- cheers --

Henrik Frystyk                                          frystyk@W3.org
World-Wide Web Consortium,                              Tel + 1 617 258 8143
MIT/LCS, NE43-356					Fax + 1 617 258 8682
77 Massachusetts Avenue
Cambridge MA 02154, USA


eturn-Path: rupesh@altair.shaktiweb.com 
Return-Path: <rupesh@altair.shaktiweb.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA15739; Fri, 5 May 1995 16:19:09 +0500
Received: from altair.shaktiweb.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA10644; Fri, 5 May 1995 16:19:02 +0500
Received: (from rupesh@localhost) by altair.shaktiweb.com (8.6.9/8.6.9) id NAA04494; Fri, 5 May 1995 13:18:15 -0700
Date: Fri, 5 May 1995 13:18:14 -0700 (PDT)
From: Rupesh Kapoor <rupesh@altair.shaktiweb.com>
To: www-talk@www10.w3.org
Subject: Re: your mail
In-Reply-To: <199505042328.QAA10898@gaia.internex.net>
Message-Id: <Pine.BSD/.3.91.950505131450.4474B-100000@altair.shaktiweb.com>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 1223

On Thu, 4 May 1995 www-talk@www10.w3.org wrote:

> I am new to this mailing list and am not sure if this
> has been discussed. But here it goes.
>     Is it possible to have 2 different names resolve to 
>     the same server but point to two different home pages.
>      For example : x.y.z should point to page 1 in my sun and
>                    address http://a.b.c should point to page 2
>                    in my sun.
> If yes, how do I do that. We use DNS and I can setup CNAME aliases
> for my sun so that x.y.z and a.b.c both point to my sun platform. 
> How do I make httpd to resolve these different names and serve
> the proper home page.
> If this is not possible if there a workaround.
> Thanks a lot in advance.
> 
> - Ramani Iyer
> 

	That's not how it works. You have to run two different httpd's (with 
different root areas, of course). They would run on diff ports, say one 
on 80 (default) and the other on 2000. URL for the first is

http://xyz.com/

and that for the second

http://xyz.com:2000/
or http://abcd.com:2000/

assuming that abcd.com is the DNS equivalent of xyz.com, and is a more 
logical name for the second server.

The port number has to be there. Am I missing a point?


-- Rupesh

eturn-Path: eshyjka%dc.isx.com@www10.w3.org 
Return-Path: <eshyjka%dc.isx.com@www10.w3.org>
Received: from mintaka.lcs.mit.edu by www19 (5.0/NSCS-1.0S) 
	id AS23529; Fri, 5 May 1995 17:34:02 +0500
Received: from www10.w3.org by MINTAKA.LCS.MIT.EDU id aa29185;
          5 May 95 10:26 EDT
Received: from dc.isx.com ([204.145.242.33]) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA07623; Fri, 5 May 1995 10:21:55 +0500
Received: from [204.145.242.135] (mac135.dc.isx.com) by dc.isx.com (4.1/SMI-4.1)
	id AA15132; Fri, 5 May 95 10:18:15 EDT
Message-Id: <v02110101abcfe4b41c0b@[204.145.242.135]>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Fri, 5 May 1995 10:23:53 -0400
To: www-talk@www10.w3.org
From: eshyjka@dc.isx.com (Elisabeth Shyjka)
Subject: Server side browser identification 
Content-Length: 459

I know that there is a way to check what browser is accessing your server.
I remember seeing this on a URL somewhere.  Does anyone know how this works
or where I could get more information?  I would like to be able to send a
page that is optimized for the accessing browser. Anyone have a good idea
where I should start?

-Elisabeth

------------------------
Elisabeth Shyjka
ISX Corporation
(703)351-6420   fax:(703)351-6429       internet:eshyjka@isx.com


eturn-Path: ldaly@acs.bu.edu 
Return-Path: <ldaly@acs.bu.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA01881; Fri, 5 May 1995 18:41:28 +0500
Received: from acs4.bu.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA11945; Fri, 5 May 1995 18:41:27 +0500
Received: by acs4.bu.edu (8.6.11/BU_SmartClient-1.0)
	id SAA21007; Fri, 5 May 1995 18:37:04 -0400
From: ldaly@acs.bu.edu
Message-Id: <199505052237.SAA21007@acs4.bu.edu>
Subject: Re: Server side browser identification
To: eshyjka@dc.isx.com
Date: Fri, 5 May 1995 18:37:04 -0400 (EDT)
Cc: www-talk@www10.w3.org
In-Reply-To: <v02110101abcfe4b41c0b@[204.145.242.135]> from "Elisabeth Shyjka" at May 5, 95 05:56:22 pm
X-Mailer: ELM [version 2.4 PL23]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 602       


> I know that there is a way to check what browser is accessing your server.
> I remember seeing this on a URL somewhere.

The URL very well might have been http://hoohoo.ncsa.uiuc.edu/cgi/env.html,
which is a listing of most (but not all) CGI environment variables.
The variable HTTP_USER_AGENT is the one you're after.

--Liza

-- 
There was nothing left to do but lift the lid of the piano
and lay the dead Chihuahua inside on the wires
                                                                   ldaly@bu.edu
                                              http://metro.turnpike.net/G/gecko

eturn-Path: ann@sonic.net 
Return-Path: <ann@sonic.net>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA02114; Fri, 5 May 1995 18:49:32 +0500
Received: from sonic.net by www10.w3.org (5.0/NSCS-1.0S) 
	id AA11968; Fri, 5 May 1995 18:49:29 +0500
Received: from @mail.sonic.net (pm00 [199.4.118.100]) by sonic.net (8.6.10/8.6.10) with SMTP id PAA30327; Fri, 5 May 1995 15:54:07 -0700
Date: Fri, 5 May 1995 15:54:07 -0700
Message-Id: <199505052254.PAA30327@sonic.net>
X-Sender: ann@mail.sonic.net
X-Mailer: Windows Eudora Version 2.0.3
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
To: www-talk@www10.w3.org
From: ann@sonic.net (Ann Lynnworth)
Subject: CGI environment vars for browser ID, etc.
Cc: eshyjka@dc.isx.com
Content-Length: 2338

To answer Elisabeth's question about identifying the browser --

A CGI program has access to a great number of environment variables which
give you all sorts of information about what's happening at run time.  This
is not available to plain HTML; you need to have a CGI program.

But if you do, here are some of the benefits:

' [CGI]                <== The standard CGI variables
' CGI Version=         The version of CGI spoken by the server
' Request Protocol=    The server's info protocol (e.g. HTTP/1.0)
' Request Method=      The method specified in the request (e.g., "GET")
' Executable Path=     Physical pathname of the back-end (this program)
' Logical Path=        Extra path info in logical space
' Physical Path=       Extra path info in local physical space
' Query String=        String following the "?" in the request URL
' Content Type=        MIME content type of info supplied with request
' Content Length=      Length, bytes, of info supplied with request
' Server Software=     Version/revision of the info (HTTP) server
' Server Name=         Server's network hostname (or alias from config)
' Server Port=         Server's network port number
' Server Admin=        E-Mail address of server's admin. (config)
' Referer=             URL of referring document (HTTP/1.0 draft 12/94)
' From=                E-Mail of client user  (HTTP/1.0 draft 12/94)
' Remote Host=         Remote client's network hostname
' Remote Address=      Remote client's network address
' Authenticated Username=Username if present in request
' Authenticated Password=Password if present in request
' Authentication Method=Method used for authentication (e.g., "Basic")
' Authentication Realm=Name of realm for users/groups

If you're developing on a Windows NT platform, you might check out
http://website.ora.com for more information.  O'Reilly's server makes it
very easy to get at these facts.

Hope this helps,

-Ann
*******************************************************
Ann Lynnworth                             ann@sonic.net
Infobahn Construction Seminar -- http://www.sonic.net/~ann/htmlsmnr.html
TimeWarp Records -- http://www.vintage.com/mall/record
On-Line Classifieds -- http://www.sonic.net/~ann/forsale
Software Development Since 1983: Paradox, C, Delphi, CGI....
*******************************************************

eturn-Path: sanders@austin.bsdi.com 
Return-Path: <sanders@austin.bsdi.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB04093; Fri, 5 May 1995 19:03:06 +0500
Received: from austin.bsdi.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA12111; Fri, 5 May 1995 19:02:56 +0500
Received: from austin.bsdi.com (sanders#oE+ngj0Uyf1yvCzwQ90Eb4Vwbvk2ES75#@localhost [127.0.0.1]) by austin.bsdi.com (8.6.12/8.6.12) with ESMTP id SAA05798 for <www-talk@www10.w3.org>; Fri, 5 May 1995 18:02:50 -0500
Message-Id: <199505052302.SAA05798@austin.bsdi.com>
To: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Re: your mail 
In-Reply-To: Rupesh Kapoor's message of Fri, 05 May 1995 16:28:56 +0500.
References: <Pine.BSD/.3.91.950505131450.4474B-100000@altair.shaktiweb.com> 
From: Tony Sanders <sanders@earth.com>
Date: Fri, 05 May 1995 18:02:45 -0500
Sender: sanders@austin.bsdi.com
Content-Length: 6249

> On Thu, 4 May 1995 www-talk@www10.w3.org wrote:
> > I am new to this mailing list and am not sure if this
> > has been discussed. But here it goes.
> >     Is it possible to have 2 different names resolve to 
> >     the same server but point to two different home pages.
...
> > If yes, how do I do that. We use DNS and I can setup CNAME aliases
> > for my sun so that x.y.z and a.b.c both point to my sun platform. 
Nope -- CNAME won't do it.

Rupesh Kapoor writes:
> 	That's not how it works. You have to run two different httpd's (with 
> different root areas, of course). They would run on diff ports, say one 
> on 80 (default) and the other on 2000. URL for the first is
...
> The port number has to be there. Am I missing a point?
Yes, and even worse, you commited the sin of misinformation.


<TITLE>multi-domain hosts (the shocking truth -- long)</TITLE>

<P>I assume what you really want are multiple,
parallel services that appear to be different machines
but are really one machine -- and since I believe
this is of general interest, here is that story.

<P>The most common request for this is from people
that want to setup parallel web servers (e.g.,
http://www.number1.com/ and http://www.number2.com/
on the same machine serving totally independant
information).  This information is aimed mainly at
people using a un*x platform.

<P>First, you <B>must</B> be able to configure multiple
network interfaces on the machine with different IP
addresses.  Some (un*x) machines can do this with
the <I><A HREF="http://www.bsdi.com/bsdi-man/?ifconfig(8)">ifconfig</A></I>
command using the alias option, for example:<PRE>
        ifconfig XX# alias ##.##.##.##</PRE>

<P>On other machines you can use spare SLIP or PPP
interfaces as placeholders for the new IP address,
but one way or another you <B>must</B> have multiple IP
addresses on that machine or else you simply
<B>cannot</B> do it (you cannot do it simply by messing
with DNS).  There is nothing special
you need to do with DNS either, you just assign the new IP
address to the name you want to use.  The seperation
totally depends on the IP addresses, the names really
have nothing to do with (they are what the user sees
and nothing more -- the machines really don't care much about
them).

<P>Ok, now for some technical bits, (<I>please note
that some details will vary depending on the
implementation of TCP/IP</I>).  When an application
<A HREF="http://www.bsdi.com/bsdi-man/?bind(2)">binds</A></I>
to a <B>TCP</B> port it must specify (amongst
other things) the <B>port</B> to which it will
bind (e.g., gopher, ftp, http, except it uses the
port number) and the address (which is often specified
as <B>INADDR_ANY</B> which means that it will
accept packets for all local destinations, that is,
IP addresses on this machine).

<P>This specification is done in a <B>struct sockaddr_in</B>
which looks like: <PRE>
        struct sockaddr_in {
                u_char  sin_len;
                u_char  sin_family;
                u_short sin_port;
                struct  in_addr sin_addr;
                char    sin_zero[8];
        }; </PRE>

<P>And the C code to setup the socket goes something like this
(this is just pseudo-code, error checking should be done, etc):<PRE>
        int socket_fd;
        int on = 1;
        struct sockaddr_in sa_in;

        socket_fd = socket(AF_INET, SOCK_STREAM, IPPROTO_TCP);
        (void) setsockopt(socket_fd, SOL_SOCKET, SO_REUSEADDR,
                (char *)&on, sizeof on);

	/* ... */

        /* fill in data for bind */
	bzero((char *) &sa_in, sizeof sa_in);
        sa_in.sin_family = AF_INET;
        sa_in.sin_port = 80;                    /* shouldn't be hardcoded */
	/* At this point, most applications have something like the following:
         *     sa_in.sin_addr.s_addr = INADDR_ANY;
	 * and you need to change it to something like: */
	{
	    /* should use gethostbyname() in real code and get
	     * the argument from the command line  */
	    long int hostaddr1 = inet_addr("1.2.3.4");
	    sa_in.sin_addr.s_addr = hostaddr1;      /* change binding here */
	}
        bind(socket_fd, (struct sockaddr *)&sa_in, sizeof sa_in);
</PRE>
<P>The reason I've given you the technical details
above is that most existing internet applications
do not provide any way to specify the host to use
in the <B>s_addr</B> field, they just hardcode the
<B>INADDR_ANY</B>
and that does what most people want.  So in most
case you will have to go in and modify the code
(in the best possible case you would add an option
that allows you to specify the binding on the command
line or in a configuration file, if not, I think
many people could hardcode in a new binding given the
information above).  If that fails, call your local
system admin (but please don't call me).

<P>One really neat thing you can do with this is,
once you get it all working you can run a "default"
server bound to <B>INADDR_ANY</B> (remember, that is the
default address and will accept packets for any
address not bound to some other server) and have it
return a "sorry, temporarily out of service" message
to the client <B>and</B> email the local webmaster
that something has gone wrong! (<I>help me</I>)

<P>Since I am, among other things, the author of
the <A HREF="http://www.bsdi.com/server/doc/plexus.html">Plexus</A>
HTTP server (it's public domain code so don't worry,
this isn't a plug for money) I now must give a
shameless little plug for my server, which supports
this feature of binding to specific addresses in
the standard distribution.  And all you have to do
(after you get it installed of course) is to use
the <TT>-h</TT> flag to specify which address to
bind to:<PRE>
        plexus -h austin.bsdi.com</PRE>

<P>That's it!  Now you have a server bound to the
IP address for austin.bsdi.com on port 80.
You could then run another "default"
server (as mentioned above) just by running <I>plexus</I>
without the <TT>-h</TT> flag, and it will bind to
<B>INADDR_ANY</B> (tada).


In addition to the above quick hack, you could make the server a little
bit smarter and a single server could service multiple IP addrs
-- I leave this as an exercise for the reader.

If you have further questions please contact your local guru and not me.
eturn-Path: rst@ai.mit.edu 
Return-Path: <rst@ai.mit.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA05941; Fri, 5 May 1995 19:22:39 +0500
Received: from life.ai.mit.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA12258; Fri, 5 May 1995 19:22:36 +0500
Received: from volterra (volterra.ai.mit.edu) by life.ai.mit.edu (4.1/AI-4.10) for www-talk@www10.w3.org id AA02185; Fri, 5 May 95 19:22:32 EDT
From: rst@ai.mit.edu (Robert S. Thau)
Received: by volterra (4.1/AI-4.10) id AA02251; Fri, 5 May 95 19:22:30 EDT
Date: Fri, 5 May 95 19:22:30 EDT
Message-Id: <9505052322.AA02251@volterra>
To: rupesh@altair.shaktiweb.com
Cc: www-talk@www10.w3.org
In-Reply-To: <Pine.BSD/.3.91.950505131450.4474B-100000@altair.shaktiweb.com> (message from Rupesh Kapoor on Fri, 5 May 1995 16:30:54 +0500)
Subject: Re: your mail
Content-Length: 1673

   On Thu, 4 May 1995 Ramani Iyer wrote:

   > I am new to this mailing list and am not sure if this
   > has been discussed. But here it goes.
   >     Is it possible to have 2 different names resolve to 
   >     the same server but point to two different home pages.
   >      For example : x.y.z should point to page 1 in my sun and
   >                    address http://a.b.c should point to page 2
   >                    in my sun.

You can't do it with DNS.  Here's one approach:

   From: Rupesh Kapoor <rupesh@altair.shaktiweb.com>

   That's not how it works. You have to run two different httpd's (with 
   different root areas, of course). They would run on diff ports, say one 
   on 80 (default) and the other on 2000. URL for the first is

   http://xyz.com/

   and that for the second

   http://xyz.com:2000/
   or http://abcd.com:2000/

Alternatively, if your computer has multiple network interfaces (or
can be configured to pretend that it does), you can assign one IP
address to abcd.com, assign another to xyz.com, and run a server which
behaves differently depending on which IP address is used to contact
it.  The important thing is that the IP addresses must be distinct,
since an HTTP client doesn't try to tell the server what it thinks the
server's name is.

The server you use has to be aware of this arrangement as well, of
course.  Patches are available for the NCSA server (and probably for
CERN as well); Apache, an NCSA derivative, comes with this code
integrated in to the main body (and we've even gotten around to
documenting it).  See

  http://www.hyperreal.com/apache/docs/virtual-host.html

for details on how this works...

rst
eturn-Path: ptong@netcom.com 
Return-Path: <ptong@netcom.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB10523; Fri, 5 May 1995 20:53:40 +0500
Received: from netcom22.netcom.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA12916; Fri, 5 May 1995 20:53:37 +0500
Received: by netcom22.netcom.com (8.6.12/Netcom)
	id RAA10527; Fri, 5 May 1995 17:52:46 -0700
Date: Fri, 5 May 1995 17:52:46 -0700
From: ptong@netcom.com (P.T. Ong)
Message-Id: <199505060052.RAA10527@netcom22.netcom.com>
To: rst@ai.mit.edu, www-talk@www10.w3.org
Subject: Re: your mail
Cc: ptong@netcom.com
Content-Length: 1100

% From www-talk@www10.w3.org  Fri May  5 17:02:51 1995
% Reply-To: rst@ai.mit.edu
% 
% Alternatively, if your computer has multiple network interfaces (or
% can be configured to pretend that it does), you can assign one IP

	How can you configure a machine to pretend it has more than
	one network interface?  Do you know if there's a standard
	way to do this in UNIX?

% address to abcd.com, assign another to xyz.com, and run a server which
% behaves differently depending on which IP address is used to contact
% it.  The important thing is that the IP addresses must be distinct,
% since an HTTP client doesn't try to tell the server what it thinks the
% server's name is.
% 
% The server you use has to be aware of this arrangement as well, of
% course.  Patches are available for the NCSA server (and probably for
% CERN as well); Apache, an NCSA derivative, comes with this code
% integrated in to the main body (and we've even gotten around to
% documenting it).  See
% 
%   http://www.hyperreal.com/apache/docs/virtual-host.html
% 
% for details on how this works...
% 
% rst
% 

Thanks.
pt
eturn-Path: oes@stud.unit.no 
Return-Path: <oes@stud.unit.no>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA25640; Sat, 6 May 1995 09:10:48 +0500
Received: from karl.stud.unit.no by www10.w3.org (5.0/NSCS-1.0S) 
	id AA16518; Sat, 6 May 1995 09:10:46 +0500
Received: from signy14 by karl.stud.unit.no with SMTP id <AA21587>
  (5.65c/karl-1.00-beta-5 for <www-talk@www10.w3.org>); Sat, 6 May 1995 15:10:43 +0200
From: "\yvind S{ter" <oes@stud.unit.no>
Received: by signy14 ; Sat, 6 May 1995 13:10:42 GMT
Message-Id: <199505061310.AA02828@signy14>
Subject: re: HTML 2.0 standard doc's
To: www-talk@www10.w3.org
Date: Sat, 6 May 1995 15:10:42 +0200 (MET DST)
In-Reply-To: <199505061212.OAA25279@flipper.pvv.unit.no> from "www-talk@www10.w3.org" at May 6, 95 02:12:57 pm
X-Mailer: ELM [version 2.4 PL23]
Content-Type: text
Content-Length: 525       


> I am training people to code with html coding.  I want them to code to  
> html 2.0 standards - I cannot find the document that lists the standards  
> for html 2.0 coding - could you point me to that?
> Thanks,
> Sharon Cling


You may tri one of these:

       	http://www.yahoo.com/Computers/World_Wide_Web/HTML/
       	http://hopf.math.nwu.edu/html2.0/htmlspec281194_1.html
      	http://www.ncsa.uiuc.edu/SDG/Software/Mosaic/Docs/fill-out-forms/overview.html
	http://www.w3.org/hypertext/WWW/MarkUp/MarkUp.html

oes
eturn-Path: martin@mrrl.lut.ac.uk 
Return-Path: <martin@mrrl.lut.ac.uk>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AC07613; Sat, 6 May 1995 11:12:51 +0500
Received: from gizmo.lut.ac.uk by www10.w3.org (5.0/NSCS-1.0S) 
	id AA17152; Sat, 6 May 1995 11:12:47 +0500
Received: from localhost (martin@localhost) by gizmo.lut.ac.uk (8.6.12/8.6.9) with SMTP id QAA24601 for <www-talk@w3.org>; Sat, 6 May 1995 16:12:43 +0100
Message-Id: <199505061512.QAA24601@gizmo.lut.ac.uk>
X-Mailer: exmh version 1.6 4/21/95
To: www-talk@w3.org
X-Uri: <URL:http://www.mrrl.lut.ac.uk/~martin>
Subject: User authentication
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Sat, 06 May 1995 16:12:42 +0100
From: Martin Hamilton <martin@mrrl.lut.ac.uk>
Content-Length: 612

There's been a lot of grumbling about the ease of mail/news forgery 
via WWW browsers, but it would be trivial to add a simple user 
authentication mechanism based on say the POP (RFC 1725) or IMAP (RFC 
1730) protocols

Using the simple (cleartext) POP3 authentication, the entire authentication dialogue need only consist of the following ...

  +OK POP server starting
  user martin
  +OK Password required for martin.
  pass secret
  +OK martin has 3 message(s) (34153 octets).
  quit
  +OK Pop server says bye!

Of course there are fancier scenarios using Kerberos, S/Key ... :-)

Just a thought!

Martin


eturn-Path: frystyk@w3.org 
Return-Path: <frystyk@w3.org>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA25055; Sat, 6 May 1995 22:59:34 +0500
Received: from www20 (www20.w3.org) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA20485; Sat, 6 May 1995 22:59:32 +0500
Received: by www20 (5.0/NSCS-1.0S) 
	id AA20524; Sat, 6 May 1995 22:59:31 +0500
Date: Sat, 6 May 1995 22:59:31 +0500
From: frystyk@w3.org (Henrik Frystyk Nielsen)
Message-Id: <9505070259.AA20524@www20>
To: www-talk@www10.w3.org
Reply-To: Henrik Frystyk Nielsen <frystyk@w3.org>
X-Sun-Charset: US-ASCII
Content-Length: 3128


> > > "well I'd really prefer mpeg over quicktime" so it'd set mpeg to 1.0 and 
> > > quicktime to 0.9, avi to 0.8, etc...  this really could be quite a 
> > > powerful mechanism.
> > 
> > Things brings up the interesting split: the q parameter is intended to
> > express lossiness of presentation, not user preferences.  Overloading it
> > has some interesting implications (e.g. most people might prefer to use
> > JPEG over GIF, but really it depends on how the image originated, since
> > converting a GIF to a JPEG for transmission will make it worse, not better.)
> 
> According to 
> <URL:http://www.w3.org/hypertext/WWW/Protocols/HTTP1.0/HTTP1.0-ID_38.html#HEADING112>:
> 
> q
>      The quality factor chosen by the user agent (and configurable by the 
>      user) that represents the desirability of that media type. In this 
>      case, desirability is usually a measure of the clients ability to 
>      faithfully represent the contents of that media type to the user. 
>      The value is in the range [0,1], where the default value is 1. 
> 
> How does the browser determine "the clients ability to faithfully represent
> the contents of that media type to the user"?  The best the client could do
> would be 0 or 1.  I might have a much better WAV player than MPEG player, but
> I'd much prefer to get MPEG streams given that I know they pack higher
> quality per byte.  At some level, the user should have control over this. 

Having the client setting the quality factors to either '0' or '1' depending on
what it can find, and then have the users adjusting according to their personal
preferences is in my opinion a perfectly good way of doing it. In the gray zone
between '0' and '1' the quality factor acts pretty much like the accept language
header: The user has to take part in assigning values.

> In a general sense, overall "quality" shouldn't be mapped directly to 
> media types, I agree.  The only thing that really maps to quality is 
> bandwidth in most cases - so an overall "Accept-Quality" header might be 
> needed (implemented as a slider replacing the throbbing N at the top of 
> my browser :) that the server uses to determine what to send.

The HTTP specification is not very clear when it comes to content-negotiation.
There are several reasons for this: First of all because most URLs maps directly
to a specific file type which leaves little room for content-negotiation (because
the document only exists in one format); Second because nobody has been able to
come up with a good set of parameters for network performance that can be used
to select the desired media type (BTW: this is not only a problem for HTTP). 
Obviously, size is a reasonable parameter, but I don't think the only one required
in a flexible algorithm.

So maybe we simply have no choice but to let the quality factor be a mapping of user
preferences!


-- cheers --

Henrik Frystyk                                          frystyk@W3.org
World-Wide Web Consortium,                              Tel + 1 617 258 8143
MIT/LCS, NE43-356					Fax + 1 617 258 8682
77 Massachusetts Avenue
Cambridge MA 02154, USA


eturn-Path: jcaron@pressimage.fr 
Return-Path: <jcaron@pressimage.fr>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA04420; Sun, 7 May 1995 01:59:25 +0500
Received: from ns.pressimage.fr by www10.w3.org (5.0/NSCS-1.0S) 
	id AA21465; Sun, 7 May 1995 01:59:19 +0500
Received: from [194.2.222.120] (jc.pressimage.fr [194.2.222.120]) by ns.pressimage.fr (8.6.9/8.6.9) with SMTP id HAA26264 for <www-talk@mail.w3.org>; Sun, 7 May 1995 07:59:11 +0200
X-Sender: jc@mail.pressimage.fr
Message-Id: <v01510100abd20ff47280@[194.2.222.120]>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Sun, 7 May 1995 07:58:59 +0200
To: www-talk@www10.w3.org
From: jcaron@pressimage.fr (Jacques Caron)
Subject: Re: your mail
Content-Length: 3393

>   On Thu, 4 May 1995 Ramani Iyer wrote:
>
>   > I am new to this mailing list and am not sure if this
>   > has been discussed. But here it goes.
>   >     Is it possible to have 2 different names resolve to
>   >     the same server but point to two different home pages.
>   >      For example : x.y.z should point to page 1 in my sun and
>   >                    address http://a.b.c should point to page 2
>   >                    in my sun.
>
>You can't do it with DNS.  Here's one approach:
>
>   From: Rupesh Kapoor <rupesh@altair.shaktiweb.com>
>
>   That's not how it works. You have to run two different httpd's (with
>   different root areas, of course). They would run on diff ports, say one
>   on 80 (default) and the other on 2000. URL for the first is
>
>   http://xyz.com/
>
>   and that for the second
>
>   http://xyz.com:2000/
>   or http://abcd.com:2000/
>
>Alternatively, if your computer has multiple network interfaces (or
>can be configured to pretend that it does), you can assign one IP
>address to abcd.com, assign another to xyz.com, and run a server which
>behaves differently depending on which IP address is used to contact
>it.  The important thing is that the IP addresses must be distinct,
>since an HTTP client doesn't try to tell the server what it thinks the
>server's name is.
>
>The server you use has to be aware of this arrangement as well, of
>course.  Patches are available for the NCSA server (and probably for
>CERN as well); Apache, an NCSA derivative, comes with this code
>integrated in to the main body (and we've even gotten around to
>documenting it).  See
>
>  http://www.hyperreal.com/apache/docs/virtual-host.html
>
>for details on how this works...
>
>rst

Hey, this gets on my nerves. When will someone do something about it? It's so
easy to add a f...ing "Host: " or "Full-URI: " header that would enable us
to do that without such a hack. Multiple IPs per host _is_ a hack. It can
only be done on a small number of OSes, and it is a real waste of IP
addresses. I _thought_ we were running out of IP address space. Looks like
you guys just want to use it still more quickly!

I have to agree that the guy that came up with this was clever, as this
works with the installed base, but such a small modification of the
protocol would be _sooooo_ easy to implement in both clients and servers
that I think that every single client and server existing would be upgraded
in a matter of weeks, and that the installed base would in a few months be
80 to 90% converted.

One would still need a "choose which home page you want" page for the case
when the server does not know who was in fact selected, and that would work
for everybody till he switches to a newer client.

So, why, _why_, *why*, WHY? A single line! It's so easy!

The worst is, it's been discussed a number of times, everytime everyone
agrees and says, OK, that's a good idea, let's do it, and then nobody
moves. What should I do? Send every web browser and server author a
personal mail to ask him to do it?

Hope this one is the good one :-)

Jacques


+-------------------------+------------------------+
|Jacques Caron            | Pressimage Telematique |
|jcaron@pressimage.fr     | 5/7 rue Raspail        |
|Tel: +33 (1) 49 88 63 56 | 93108 Montreuil Cedex  |
|Fax: +33 (1) 49 88 63 64 | France                 |
+-------------------------+------------------------+


eturn-Path: ptitz@dux.ru 
Return-Path: <ptitz@dux.ru>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB24514; Sun, 7 May 1995 10:05:11 +0500
Received: from dux.ru (ns.dux.ru) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA23299; Sun, 7 May 1995 10:05:08 +0500
Received: (from ptitz@localhost) by dux.ru (8.6.10/8.6.6) id SAA11884 for www-talk@www10.w3.org; Sun, 7 May 1995 18:04:21 +0400
From: Dmitry Mishin <ptitz@dux.ru>
Message-Id: <199505071404.SAA11884@dux.ru>
Organization: DUX
Subject: Re: User authentication
To: www-talk@www10.w3.org
Date: Sun, 7 May 1995 18:04:20 +0400 (MSD)
In-Reply-To: <199505061512.QAA24601@gizmo.lut.ac.uk> from "Martin Hamilton" at May 6, 95 11:23:48 am
X-Mailer: ELM [version 2.4 PL23]
Mime-Version: 1.0
Content-Type: text/plain; charset=koi8-r
Content-Transfer-Encoding: 8bit
Content-Length: 888       

Martin Hamilton <martin@mrrl.lut.ac.uk> wrote:
> 
> There's been a lot of grumbling about the ease of mail/news forgery 
> via WWW browsers, but it would be trivial to add a simple user 
> authentication mechanism based on say the POP (RFC 1725) or IMAP (RFC 
> 1730) protocols
> 

What for is to add a bad implementation of another (not strongly 
required) protocol into complex browser.

Now we have a bad mailto implementation (required) and bad newsreader
implementation (not strongly required). Doubtfully some more bad 
implementations are needed.

For the one hand the browsers we see now are getting to grow into
monsters. They take loads of resources (MS-Windows based for example).
IMHO, much better is to have the browser made of modules or with posibility
fork a module, mosts of them can be replaced with (or added) specialised 
and more convenient to the user's choice.

D.
eturn-Path: swepett@kajen.malmo.se 
Return-Path: <swepett@kajen.malmo.se>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA25191; Sun, 7 May 1995 10:33:29 +0500
Received: from kajen.malmo.se ([194.16.18.1]) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA23469; Sun, 7 May 1995 10:33:17 +0500
Received: (from swepett@localhost) by kajen.malmo.se (8.6.9/8.6.9) id QAA24661; Sun, 7 May 1995 16:28:46 +0200
Date: Sun, 7 May 1995 16:28:44 +0100
From: Svante Pettersson <swepett@kajen.malmo.se>
Subject: 
To: Svante Pettersson <swepett@kajen.malmo.se>
Message-Id: <Pine.3.89.9505071652.A24627-0100000@julia.kajen.malmo.se>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 380

test

-------------------------------------------------------------------------------
Svante Pettersson			    "The world isn't flat.
Kajplats 305			             It's actually +6 dBa at 5.7 kHz."
swepett@kajen.malmo.se
webmaster@kajen.malmo.se				 
http://www.kajen.malmo.se/~swepett/				-FSCFRA 1994
-------------------------------------------------------------------------------
eturn-Path: nazgul@utopia.com 
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB25886; Sun, 7 May 1995 10:57:39 +0500
Received: from utopia.com (utopia.wing.net) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA23634; Sun, 7 May 1995 10:57:31 +0500
Received: from [204.57.39.6] by utopia.com (8.6.12/SONY-4.0MX)
	id KAA06422; Sun, 7 May 1995 10:55:45 -0400
Return-Path: <nazgul@utopia.com>
X-Sender: nazgul@mailhost.utopia.com
Message-Id: <v02110123abd28c318914@[204.57.39.6]>
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Date: Sun, 7 May 1995 10:57:32 -0400
To: www-html@www10.w3.org
From: nazgul@utopia.com (Kee Hinckley)
Subject: Re: 
Cc: www-talk@www10.w3.org
Content-Length: 1794

>designing WWW sites. I have been asked to explore the possibilites of
>deigning a WWW site that has an user fill-in form that the information is
>then automatically stored in a searchable database.
>
>I would like to know what questions I should ask service providers before
>we choose one.  Any hints you experienced users could give would be greatly
>appreciated.

I assume you want the searchable database to be on the server?  Most Web
Site providers we have talked to are happy to let you have forms which mail
you the results. However they are very hesitant about letting people write
arbitrary scripts that run on their machines. Just to complicate things,
even if they do let you write such scripts, they probably run their server
with a user of 'none', which means that any files you have your scripts
create will have to be writable by everyone on the machine - a not so
pleasant situation.

That said, it's not impossible. We have relationships with several
providers that allow us to write scripts - but they don't let all of their
customers do so. And the degree of security ranges from none to having to
pass our scripts through their vetting process.

I don't think you'll see the situation improve much until the server
companies start providing servers that run chroot'd as a specified a user.
Then you can have your own partition on the machine and the provider won't
have as many concerns.

(This topic may be better suited for www-talk, I've cc'd it there.)

Kee Hinckley      Utopia Inc. - Cyberspace Architects=81    617/721-6100
nazgul@utopia.com                               http://www.utopia.com/

I'm not sure which upsets me more: that people are so unwilling to accept
responsibility for their own actions, or that they are so eager to regulate
everyone else's.


eturn-Path: agulbra@troll.no 
Return-Path: <agulbra@troll.no>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA26024; Sun, 7 May 1995 11:06:41 +0500
Received: from Norway.EU.net (nic.eunet.no) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA23727; Sun, 7 May 1995 11:06:32 +0500
Received: by Norway.EU.net with UUCP id AA06939
  (5.65c/IDA-1.4.4/EUnet/NO for www10.w3.org!www-talk); Sun, 7 May 1995 17:05:54 +0200
Date: Sun, 7 May 1995 16:32:11 +0200 (MET DST)
From: Arnt Gulbrandsen <agulbra@troll.no>
To: Jacques Caron <jcaron@pressimage.fr>
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Re: your mail
In-Reply-To: <v01510100abd20ff47280@[194.2.222.120]>
Message-Id: <Pine.LNX.3.91.950507161326.402C-100000@pentagram.troll.no>
Return-Receipt-To: agulbra@troll.no
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 1427

> Hey, this gets on my nerves. When will someone do something about it? It's so
> easy to add a f...ing "Host: " or "Full-URI: " header that would enable us
> to do that without such a hack. Multiple IPs per host _is_ a hack. It can
> only be done on a small number of OSes, and it is a real waste of IP
> addresses. I _thought_ we were running out of IP address space. Looks like
> you guys just want to use it still more quickly!

Not a great waste, no.  The _great_ waste of IP addresses is the number of
blocks being allocated, one block per organization.  I work for a small
company which has been allocated almost a hundred times as many addresses
as we use; a fairly common situation and very wasteful.

Compared to that waste, everything else is just surface noise.  We're
about to set up www.troll.no, when we have we'll occupy 257 IP addresses,
one up from 256.  And that's the worst case.  If troll.no had 1024 IP
addresses we'd would use 1025; 0.1% extra.  When even the worst case is
less than 0.25% waste, who cares?

The internet draft draft-gulbrandsen-dns-rr-srvcs-00.txt may, with one
change (an additional "actual port" data field in the RR), be used to
solve this "problem".  But it would require all the clients to change,
every single one.

I'd appreciate comments on the draft; should I put back in "actual port"? 
It was there in the very first version, but went away for lack of
perceived use.

--Arnt
eturn-Path: martin@mrrl.lut.ac.uk 
Return-Path: <martin@mrrl.lut.ac.uk>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA28396; Sun, 7 May 1995 12:02:38 +0500
Received: from gizmo.lut.ac.uk by www10.w3.org (5.0/NSCS-1.0S) 
	id AA24088; Sun, 7 May 1995 12:02:33 +0500
Received: from localhost (martin@localhost) by gizmo.lut.ac.uk (8.6.12/8.6.9) with SMTP id RAA19935 for <www-talk@w3.org>; Sun, 7 May 1995 17:02:30 +0100
Message-Id: <199505071602.RAA19935@gizmo.lut.ac.uk>
X-Mailer: exmh version 1.6 4/21/95
To: www-talk@w3.org
Subject: Re: User authentication 
X-Uri: <URL:http://www.mrrl.lut.ac.uk/~martin>
In-Reply-To: Your message of "Sun, 07 May 1995 10:12:36 +0500."
             <199505071404.SAA11884@dux.ru> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Sun, 07 May 1995 17:02:30 +0100
From: Martin Hamilton <martin@mrrl.lut.ac.uk>
Content-Length: 539


Dmitry Mishin writes:

| Now we have a bad mailto implementation (required) and bad newsreader
| implementation (not strongly required). Doubtfully some more bad 
| implementations are needed.

Sure - but if you are going to implement (say) mailto support in your browser, ought you not to at least make a token(!) effort to find out whether the user is who they're claiming to be ?

Plus, the scenario I outlined hardly adds virtually no new code to the browser. Is this more or less worthwhile than <blink> ... </blink> ? :-)

Martin


eturn-Path: andersg@ludd.luth.se 
Return-Path: <andersg@ludd.luth.se>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA01384; Sun, 7 May 1995 12:24:54 +0500
Received: from zed.ludd.luth.se by www10.w3.org (5.0/NSCS-1.0S) 
	id AA24171; Sun, 7 May 1995 12:24:52 +0500
Received: from zero.ludd.luth.se (zero.ludd.luth.se [130.240.16.35]) by zed.ludd.luth.se (8.6.11/8.6.11) with ESMTP id SAA28200 for <www-talk@www10.w3.org>; Sun, 7 May 1995 18:24:49 +0200
From: andersg@ludd.luth.se
Received: (andersg@localhost) by zero.ludd.luth.se (8.6.11/8.6.11) id SAA10336 for www-talk@www10.w3.org; Sun, 7 May 1995 18:24:48 +0200
Message-Id: <199505071624.SAA10336@zero.ludd.luth.se>
Subject: Show movies in Netscape?
To: www-talk@www10.w3.org
Date: Sun, 7 May 1995 18:24:46 +0200 (EET DST)
Content-Type: text
Content-Length: 325       

Can anyone give me some hints or information about how I shall make a 
HTML-page wich plays a movie when you enter this side or when you press a 
button ?
I would like to show quicktime and mpeg movies!
My platform is UNIX sparc solaris2.4, and I'm using NCSA httpd1.3 web-server.
/Anders Gunnare
e-mail:andersg@ludd.luth.se
eturn-Path: rich@kitty.oester.com 
Return-Path: <rich@kitty.oester.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA10341; Sun, 7 May 1995 15:00:16 +0500
Received: from kitty.oester.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA25145; Sun, 7 May 1995 15:00:12 +0500
Received: from kitty.oester.com (localhost [127.0.0.1]) by kitty.oester.com (8.6.10/8.6.10) with ESMTP id MAA12723 for <www-talk@www10.w3.org>; Sun, 7 May 1995 12:00:11 -0700
Message-Id: <199505071900.MAA12723@kitty.oester.com>
From: "Gintaras Richard Gircys (GG148)" <rich@oester.com>
X-Phone: 408 722 3682
X-Orgs: Oesterreich & Assc. Inc.
X-Snail: 2014 Eureka Canyon Road, Corralitos, CA. 95076
To: www-talk@www10.w3.org
Subject: multiple home pages/ips
In-Reply-To: Message from Sun, 07 May 1995 11:36:26 +0500.
             <Pine.LNX.3.91.950507161326.402C-100000@pentagram.troll.no> 
Date: Sun, 07 May 1995 12:00:10 -0700
Sender: rich@kitty.oester.com
Content-Length: 1788

> > Hey, this gets on my nerves. When will someone do something about it? It's so
> > easy to add a f...ing "Host: " or "Full-URI: " header that would enable us

i agree that this is a nervy thing, and i don't like the way it's currently
done, but unless i am missing something, i don't see how adding ANY information
to a documentt will fix this.

> > to do that without such a hack. Multiple IPs per host _is_ a hack. It can
> > only be done on a small number of OSes, and it is a real waste of IP

yes, it's a hack, a a benefit of running a real os (where this ability is there
exactly for this reason - hopefully a temporary solution until...)

> The internet draft draft-gulbrandsen-dns-rr-srvcs-00.txt may, with one
> change (an additional "actual port" data field in the RR), be used to
> solve this "problem".  But it would require all the clients to change,
> every single one.
>
... the above can be implemented. doing this by dns is the correct way (analogous
to the correct use of MX records to do similar functional vis-a-vis email).
 
> I'd appreciate comments on the draft; should I put back in "actual port"? 
> It was there in the very first version, but went away for lack of
> perceived use.
> 
if this change to dns is being considered, that's good enought for me - year
ago you couldn't get any to even listen to such an idea.

as for the current hack being clever - it ain't; ip alias games has been a useful
unix admin tool for years - saying it's clever is like saying "ain't it clever
to put a bandage on that cut" - it's obvious.

as for the apache hint - unless something has changed, how to do this is not
documented, and the actual routing manipulations will probably be somewhat
unix vendor/setup specific; there's at least two ways this can be done.

rich

eturn-Path: srinivas@cs.iastate.edu 
Return-Path: <srinivas@cs.iastate.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB14305; Sun, 7 May 1995 15:52:09 +0500
Received: from cs.iastate.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA25417; Sun, 7 May 1995 15:52:07 +0500
Received: from shazam.cs.iastate.edu (shazam.cs.iastate.edu [129.186.3.2]) by cs.iastate.edu (8.6.10/8.6.9) with ESMTP id OAA25397 for <www-talk@www10.w3.org>; Sun, 7 May 1995 14:52:05 -0500
From: Srivatsa Srinivasan <srinivas@cs.iastate.edu>
Received: (srinivas@localhost) by shazam.cs.iastate.edu (8.6.10/8.6.9) id OAA20612 for www-talk@www10.w3.org; Sun, 7 May 1995 14:51:56 -0500
Message-Id: <199505071951.OAA20612@shazam.cs.iastate.edu>
Subject: HTTP Content-length question...
To: www-talk@www10.w3.org
Date: Sun, 7 May 95 14:51:55 CDT
X-Hpvue$Revision: 1.8 $
Mime-Version: 1.0
Content-Type: Message/rfc822
X-Vue-Mime-Level: 4
Mailer: Elm [revision: 70.85]
Content-Length: 1174

Hello WWWorld,

 I have a HTTP protocol question.

I have a simple HTTP client that requests an object from a HTTP server using
the GET method. Now, the client is written such that it mandatorily needs the
server to return the 'Content-length:' header in its response.

The problem I am facing is, this header is returned for most kind of objects
(image/gif, audio/au etc) but not for text/html. Is there any special form
of the GET request that will force the server to return the size of the object
(the 'Content-length:' header) irrespective of its MIME type ?

Thanks for your help!
Srivatsa Srinivasan

--
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
|                   Srivatsa Belur Srinivasan                       |
| 211, Atanasoff Hall,                   246,N. Hyland Avenue, #208 |
| Iowa State University,                 Ames, IA 50014, USA        |
| Email : srinivas@iastate.edu           Phone : (515) 292-5458     |
| WWW   : http://www.public.iastate.edu/~srinivas                   |
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
For A Deterministic Finite State Patrol On The Information Superhighway.
eturn-Path: frystyk@w3.org 
Return-Path: <frystyk@w3.org>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA17794; Sun, 7 May 1995 16:49:39 +0500
Received: from www20 (www20.w3.org) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA25758; Sun, 7 May 1995 16:49:38 +0500
Received: by www20 (5.0/NSCS-1.0S) 
	id AA22106; Sun, 7 May 1995 16:49:37 +0500
Date: Sun, 7 May 1995 16:49:37 +0500
From: frystyk@w3.org (Henrik Frystyk Nielsen)
Message-Id: <9505072049.AA22106@www20>
To: www-talk@www10.w3.org, srinivas@cs.iastate.edu
Subject: Re: HTTP Content-length question...
Reply-To: Henrik Frystyk Nielsen <frystyk@w3.org>
X-Sun-Charset: US-ASCII
Content-Length: 1140


> I have a simple HTTP client that requests an object from a HTTP server using
> the GET method. Now, the client is written such that it mandatorily needs the
> server to return the 'Content-length:' header in its response.
> 
> The problem I am facing is, this header is returned for most kind of objects
> (image/gif, audio/au etc) but not for text/html. Is there any special form
> of the GET request that will force the server to return the size of the object
> (the 'Content-length:' header) irrespective of its MIME type ?

If this is the case then it is a server bug! The Content-Length header
should be sent whenever possible in a full response, and it should certainly
not have anything to do with the media type in which the object is rendered.

I can imagine that the problems can be due to server side includes in the
documents or dynamic documents generated by scripts.

-- cheers --

Henrik Frystyk                                          frystyk@W3.org
World-Wide Web Consortium,                              Tel + 1 617 258 8143
MIT/LCS, NE43-356					Fax + 1 617 258 8682
77 Massachusetts Avenue
Cambridge MA 02154, USA


eturn-Path: frystyk@w3.org 
Return-Path: <frystyk@w3.org>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA18927; Sun, 7 May 1995 17:00:46 +0500
Received: from www20 (www20.w3.org) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA25851; Sun, 7 May 1995 17:00:41 +0500
Received: by www20 (5.0/NSCS-1.0S) 
	id AA22163; Sun, 7 May 1995 17:00:40 +0500
Date: Sun, 7 May 1995 17:00:40 +0500
From: frystyk@w3.org (Henrik Frystyk Nielsen)
Message-Id: <9505072100.AA22163@www20>
To: www-talk@www10.w3.org
Reply-To: Henrik Frystyk Nielsen <frystyk@w3.org>
X-Sun-Charset: US-ASCII
Content-Length: 1133


> Could anyone describe the pros and cons of various systems (operating 
> system, hardware/software, connections, etc.) for setting up a Web Server 
> for home or office.
> 
> Is a mac better than windows NT or Sparcstation?  For the office we have 
> direct access to the internet and we have one Sparcstation 20 and one Sparc 
> 5.  We may or may not be able to use those for internet apps.  Should be 
> purchase another Sparc or buy a mac...windows NT?

Please refer such questions to either the news lists or to the mailing lists
that concern server administration. Only by keeping down the level of
unrelated questions, comments, etc. can we keep a reasonable forum for
discussions.

For information regarding mail addresses and mailing lists please take
a look at the page

	http://www.w3.org/hypertext/WWW/Mailing/Mail/

This page will give you the information needed.

-- cheers --

Henrik Frystyk                                          frystyk@W3.org
World-Wide Web Consortium,                              Tel + 1 617 258 8143
MIT/LCS, NE43-356					Fax + 1 617 258 8682
77 Massachusetts Avenue
Cambridge MA 02154, USA
eturn-Path: akosut@nueva.pvt.k12.ca.us 
Return-Path: <akosut@nueva.pvt.k12.ca.us>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA21620; Sun, 7 May 1995 17:20:54 +0500
Received: from mail.barrnet.net by www10.w3.org (5.0/NSCS-1.0S) 
	id AA26030; Sun, 7 May 1995 17:20:52 +0500
Received: from ace.nueva.pvt.k12.ca.us (ace.nueva.pvt.k12.ca.us [198.31.42.1]) by mail.barrnet.net (8.6.10/MAIL-RELAY-LEN) with SMTP id OAA02852; Sun, 7 May 1995 14:18:06 -0700
Received: by ace.nueva.pvt.k12.ca.us
	(1.38.193.5/15.5+ECS 3.3+HPL1.1) id AA05103; Sun, 7 May 1995 14:20:49 -0700
Date: Sun, 7 May 1995 14:20:49 -0700 (PDT)
From: Alexei Kosut <akosut@nueva.pvt.k12.ca.us>
To: Srivatsa Srinivasan <srinivas@cs.iastate.edu>
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Re: HTTP Content-length question...
In-Reply-To: <199505071951.OAA20612@shazam.cs.iastate.edu>
Message-Id: <Pine.HPP.3.91.950507141858.5053A-100000@ace.nueva.pvt.k12.ca.us>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 1251

On Sun, 7 May 1995, Srivatsa Srinivasan wrote:

> The problem I am facing is, this header is returned for most kind of objects
> (image/gif, audio/au etc) but not for text/html. Is there any special form
> of the GET request that will force the server to return the size of the object
> (the 'Content-length:' header) irrespective of its MIME type ?

The most recent HTTP 1.0 draft I could find states:

   Although it is not required, applications are strongly encouraged to
   use this field to indicate the size of the Entity-Body to be
   transferred, regardless of the media type of the entity.

The first five words here are the important ones... the server doesn't 
have to include Content-Length, and sometimes it doesn't. It's certainly 
nice, but you can't assume that this header will exist when writing an 
HTTP client. Because it won't always. But it is odd that your server 
returns the header for images, but not text... sounds like a bug.

--
Alexei Kosut          Live, Londo and Prosper: /\/\/\\____-_____--  __.__..
akosut@nueva.pvt.k12.ca.us             |-|-----|:|:|:: ..| |...|  ||=/  \
Lefler on IRC                          |-|-----======____| |---|  |-=\__/
<URL:http://www.nueva.pvt.k12.ca.us/~akosut/>  \/\/\/     -     --

eturn-Path: frystyk@w3.org 
Return-Path: <frystyk@w3.org>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA02835; Sun, 7 May 1995 19:24:51 +0500
Received: from www20 (www20.w3.org) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA26782; Sun, 7 May 1995 19:24:50 +0500
Received: by www20 (5.0/NSCS-1.0S) 
	id AA22825; Sun, 7 May 1995 19:24:49 +0500
Date: Sun, 7 May 1995 19:24:49 +0500
From: frystyk@w3.org (Henrik Frystyk Nielsen)
Message-Id: <9505072324.AA22825@www20>
To: www-talk@www10.w3.org
Reply-To: Henrik Frystyk Nielsen <frystyk@w3.org>
X-Sun-Charset: US-ASCII
Content-Length: 576


> It took more than a year for FORMS to make it into any spec, in fact
> they still are not in any official spec, so it's no suprise that
> the Refresh header isn't in the HTTP spec yet.

The refresh header is planned for HTTP/1.1 - actually, as far as I remember,
it was in one of the drafts but was taken out again?


-- cheers --

Henrik Frystyk                                          frystyk@W3.org
World-Wide Web Consortium,                              Tel + 1 617 258 8143
MIT/LCS, NE43-356					Fax + 1 617 258 8682
77 Massachusetts Avenue
Cambridge MA 02154, USA


eturn-Path: cwerner@hsdemo.merit.edu 
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB16879; Sun, 7 May 1995 22:13:33 +0500
Received: from hostserver.merit.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA27892; Sun, 7 May 1995 22:13:32 +0500
Return-Path: <cwerner@hsdemo.merit.edu>
Received: from [198.111.2.19] by hostserver.merit.edu (8.6.12/hostsrvr-1.1)
	id WAA19195; Sun, 7 May 1995 22:13:27 -0400
Message-Id: <199505080213.WAA19195@hostserver.merit.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Sun, 7 May 1995 22:18:58 -0500
To: www-talk@www10.w3.org
From: cwerner@hsdemo.merit.edu (Christopher L. Werner)
Subject: Modular Browsers (was:User authentication)
Content-Length: 1334


Dmitry Mishin writes:
>For the one hand the browsers we see now are getting to grow into
>monsters. They take loads of resources (MS-Windows based for example).
>IMHO, much better is to have the browser made of modules or with posibility
>fork a module, mosts of them can be replaced with (or added) specialised
>and more convenient to the user's choice.

Gee, sounds like Sun marketing is on the right track.  Modular browsers,
what a concept! So we all wait for WebBrowser to be ported to Windows?
Hot Java here we come...

Personally, I find the feature creep is bring us ever so close to SGML
that we should have jumped on the DoD bandwagon years ago. Better
authentication will only work if the encryption methods are available
internationally and legislation like the new California key-server laws
don't drive the price of registration to a magnetude greater than the
application cost ($30).  That they require more overhead is only a side
effect.

-----------------------------------------------------------------
Opinions expressed are my own and not those of Robert Bosch Corp.
-----------------------------------------------------------------
Christopher L. Werner         |  Robert Bosch Corporation
System Engineer               |  38000 Hills Tech Dr.
(810)553-1389                 |  Farmington Hills, MI 48331-3417


eturn-Path: mcclurea@nag.cs.colorado.edu 
Return-Path: <mcclurea@nag.cs.colorado.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA18944; Sun, 7 May 1995 22:43:40 +0500
Received: from nag.cs.colorado.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA28178; Sun, 7 May 1995 22:43:38 +0500
Received: (from mcclurea@localhost) by nag.cs.colorado.edu (8.6.10/8.6.10) id UAA19294; Sun, 7 May 1995 20:43:29 -0600
From: "Adam T. McClure" <mcclurea@nag.cs.colorado.edu>
Message-Id: <199505080243.UAA19294@nag.cs.colorado.edu>
Subject: Re: Modular Browsers (was:User authentication)
To: cwerner@hsdemo.merit.edu
Date: Sun, 7 May 1995 20:43:29 -0600 (MDT)
Cc: www-talk@www10.w3.org
In-Reply-To: <199505080213.WAA19195@hostserver.merit.edu> from "Christopher L. Werner" at May 7, 95 10:22:25 pm
X-Mailer: ELM [version 2.4 PL23]
Content-Type: text
Content-Length: 849       

Christopher L. Werner
> 
> >For the one hand the browsers we see now are getting to grow into
> >monsters. They take loads of resources (MS-Windows based for example).
> >IMHO, much better is to have the browser made of modules or with posibility
> >fork a module, mosts of them can be replaced with (or added) specialised
> >and more convenient to the user's choice.
> 
Yeah, how bout an OpenDoc browser Netscape?
Is that a no-brainer or a novel concept?

-- 
_/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/
Adam T. McClure    	Integrated Teaching & Learning project
mcclurea@colorado.edu	University of Colorado-Boulder

    <a href="http://itldemo.colorado.edu/rootitl.html></a>

You know what I like about standards?  There's so many to choose from.
_/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/
eturn-Path: sunok@shiva.hsr.re.kr 
Return-Path: <sunok@shiva.hsr.re.kr>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AD20938; Sun, 7 May 1995 23:11:17 +0500
Received: from garam.kreonet.re.kr by www10.w3.org (5.0/NSCS-1.0S) 
	id AA28459; Sun, 7 May 1995 23:11:09 +0500
Received: from heis1.hsr.re.kr (heis1.hsr.re.kr [134.75.214.2]) by garam.kreonet.re.kr (8.6.9H1/8.6.9) with ESMTP id MAA10665 for <www-talk@www10.w3.org>; Mon, 8 May 1995 12:13:33 +0900
From: sunok@shiva.hsr.re.kr
Received: from shiva.hsr.re.kr (shiva.hsr.re.kr [134.75.214.124]) by heis1.hsr.re.kr (8.6.11/8.6.9) with ESMTP id MAA04722 for <www-talk@www10.w3.org>; Mon, 8 May 1995 12:13:02 +0900
Received: from  (angel.hsr.re.kr [134.75.214.138]) by shiva.hsr.re.kr (8.6.10H1/8.6.9) with SMTP id KAA05378 for <www-talk@www10.w3.org>; Thu, 4 May 1995 10:53:28 +0900
Date: Thu, 4 May 1995 10:53:28 +0900
Message-Id: <199505040153.KAA05378@shiva.hsr.re.kr>
To: www-talk@www10.w3.org
Subject: help! HTTP method trying
X-Mailer: AIR Mail 3.X (SPRY, Inc.)
Content-Length: 652

It is the first try for me to ask a question.
I am studying HTTP now.
after I logined by "telnet 'hostname' 80 ",
I try to PUT, GET, POST methods test.
for example, I did like
"GET http://shiva.hsr.re.kr/~~/file.html HTTP/1.0".
but at that time, I received 404 error message.
I became to know that the server is the NCSA/1.3.
I want to the response from the server that I write.
How can I do at this time?
if you know, reply to me. 
the faster, the better. 

regards, 
  Sunny.

ps::sorry for my clumsy english. ;)


   ============================
        CHANG SUN OK
     sunok@shiva.hsr.re.kr
        Tel. 398-4732
   ============================

eturn-Path: bruno@www10.w3.org 
Return-Path: <bruno@www10.w3.org>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB00378; Mon, 8 May 1995 00:52:37 +0500
Received: from www19 (www19.w3.org) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA29338; Mon, 8 May 1995 00:52:35 +0500
Received: by www19 (5.0/NSCS-1.0S) 
	id AA00374; Mon, 8 May 1995 00:52:34 +0500
Date: Mon, 8 May 1995 00:52:34 +0500
Message-Id: <9505080452.AA00374@www19>
To: www-talk@www10.w3.org, sscling@ingr.com
Subject: Re: HTML 2.0 DOCUMENT / specifications
Reply-To: BFG@w3.org
From: bfg@w3.org
X-Sun-Charset: US-ASCII
Content-Length: 661


Sharon Cling wrote:
> I am training people to code with html coding.  I want them to code to  
> html 2.0 standards - I cannot find the document that lists the standards  
> for html 2.0 coding - could you point me to that?



	http://www.w3.org/hypertext/WWW/MarkUp/MarkUp.html

lets you access the latest specifications as of May 6, 1995 :

	http://www.w3.org/hypertext/WWW/MarkUp/html-spec/html-spec_toc.html 

Bruno Girschweiler
B F G

----------------------------------------------------------------------------
   email  :   bfg@w3.org               76373.3506@compuserve.com
----------------------------------------------------------------------------

eturn-Path: F.O.Estensen@newcastle.ac.uk 
Return-Path: <F.O.Estensen@newcastle.ac.uk>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA20590; Mon, 8 May 1995 09:11:23 +0500
Received: from cheviot.ncl.ac.uk by www10.w3.org (5.0/NSCS-1.0S) 
	id AA03094; Mon, 8 May 1995 09:11:11 +0500
Received: from burnmoor.ncl.ac.uk by cheviot.ncl.ac.uk id <OAA24839@cheviot.ncl.ac.uk>
  (8.6.10/ for ncl.ac.uk) with SMTP; Mon, 8 May 1995 14:11:03 +0100
Received: from eata.ncl.ac.uk (eata.ncl.ac.uk [128.240.2.18]) by burnmoor.ncl.ac.uk (8.6.12/8.6.10-cf revision 2 for Solaris 2.x) with ESMTP id OAA00703; Mon, 8 May 1995 14:11:03 +0100
Received: (n314974@localhost) by eata.ncl.ac.uk (8.6.11/8.6.10-cf revision 1 for SunOS 4.1.x) id OAA11972; Mon, 8 May 1995 14:11:01 +0100
Date: Mon, 8 May 1995 14:11:01 +0100 (BST)
From: Frank Olav Estensen <F.O.Estensen@newcastle.ac.uk>
To: www-talk@www10.w3.org
Subject: File names
Message-Id: <Pine.SUN.3.91-941213.950508140848.11828A-100000@eata.ncl.ac.uk>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 525

How can I find the file name of a document with the URL:

http://foo.co.uk/

Is there a methode to get the filename, or do I just have to assume 
index.html?

Mr Frank Olav Estensen
Student, Computing Science part III, University of Newcastle
WWW   :   http://www.newcasle.ac.uk/~n314974
          http://www.nordnett.no/~estensen
E-mail:				Snail-mail:
F.O.Estensen@newcastle.ac.uk	105 Stratford Road
F.O.Estensen@nordnett.no        Heaton
Phone:				Newcastle upon Tyne
+44 91 224 2355  		NE6 5AS
+47 75 56 09 48			England

eturn-Path: jimg@dcz.gso.uri.edu 
Return-Path: <jimg@dcz.gso.uri.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA27418; Mon, 8 May 1995 09:58:45 +0500
Received: from gsosun1 (gsosun1.gso.uri.edu) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA03678; Mon, 8 May 1995 09:58:42 +0500
Received: from dcz.gso.uri.EDU by gsosun1 (4.1/SMI-4.1)
	id AA19707; Mon, 8 May 95 09:56:06 EDT
Received: by dcz.gso.uri.EDU (4.1/SMI-4.1)
	id AA09944; Mon, 8 May 95 09:58:39 EDT
Date: Mon, 8 May 95 09:58:39 EDT
From: jimg@dcz.gso.uri.edu (James Gallagher)
Message-Id: <9505081358.AA09944@dcz.gso.uri.EDU>
To: www-talk@www10.w3.org
Subject: Re: your mail
In-Reply-To: <Pine.BSD/.3.91.950505131450.4474B-100000@altair.shaktiweb.com>
References: <Pine.BSD/.3.91.950505131450.4474B-100000@altair.shaktiweb.com>
Content-Length: 827

>>Rupesh Kapoor <rupesh@altair.shaktiweb.com> writes:

 > On Thu, 4 May 1995 www-talk@www10.w3.org wrote:
 >> I am new to this mailing list and am not sure if this has been
 >> discussed. But here it goes.  Is it possible to have 2 different names
 >> resolve to the same server but point to two different home pages.  For

...

 >> 
 >> - Ramani Iyer
 >> 

 > 	That's not how it works. You have to run two different httpd's (with

...

 > -- Rupesh

But check out http://www.thesphere.com/~dlp/TwoServers/ for some more ideas.


James Gallagher ----------------------------------------------------------
The University of Rhode Island 	|	email: jimg@dcz.gso.uri.edu
South Ferry Road		|	Phone: 401.792.6939
Narragansett, RI 02882		|	FAX: 401.792.6728
--------------------------------------------------------------------------

eturn-Path: DANTONIO@process.com 
Return-Path: <DANTONIO@process.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA27734; Mon, 8 May 1995 10:01:15 +0500
Received: from sirius.process.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA03719; Mon, 8 May 1995 10:01:05 +0500
Date:     Mon, 8 May 1995 10:01 -0400
From: DANTONIO@process.com (Momentary Language, Sexual Situations)
Message-Id: <009900B318A87A40.06CA@PROCESS.COM>
To: www-talk@www10.w3.org
Subject:  RE: FW: Pros and Cons of various OSs for WWW Servers...
X-Vms-To: smtp%"www-talk@www10.w3.org"
Content-Length: 707

> From: staffan@ep.se (Staffan Berger, ElektroPost)
>>cbrenton@digprod.com wrote:

>>If you have a server that is
>>expecting heavy traffic however (50,000+ hits per day) there's no
>>substitution for a UNIX platform.

>And what about Microsoft? They are running their WWW server under Windows
>NT and I'm sure that they have more than 50 000 hits per day. We have

>around 8000 and still have capacity for ten times more on our Windows NT 
>server.

Our server WWW server for Windows NT (Purveyor) routinely took 250,000 hits
in a 24 hour period (measured during finalt testing right before release)
so Unix certainly isn't the only choice.

DDA

P.S. for more info on Purveyor, check out www.process.com

eturn-Path: rst@ai.mit.edu 
Return-Path: <rst@ai.mit.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA00419; Mon, 8 May 1995 10:27:25 +0500
Received: from life.ai.mit.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA04044; Mon, 8 May 1995 10:27:21 +0500
Received: from volterra (volterra.ai.mit.edu) by life.ai.mit.edu (4.1/AI-4.10) for www-talk@www10.w3.org id AA02166; Mon, 8 May 95 10:27:19 EDT
From: rst@ai.mit.edu (Robert S. Thau)
Received: by volterra (4.1/AI-4.10) id AA02848; Mon, 8 May 95 10:27:17 EDT
Date: Mon, 8 May 95 10:27:17 EDT
Message-Id: <9505081427.AA02848@volterra>
To: srinivas@cs.iastate.edu
Cc: www-talk@www10.w3.org
In-Reply-To: <199505071951.OAA20612@shazam.cs.iastate.edu> (message from Srivatsa Srinivasan on Sun, 7 May 1995 15:58:18 +0500)
Subject: Re: HTTP Content-length question...
Content-Length: 1308

   From: Srivatsa Srinivasan <srinivas@cs.iastate.edu>

   Hello WWWorld,

    I have a HTTP protocol question.

   I have a simple HTTP client that requests an object from a HTTP
   server using the GET method. Now, the client is written such that
   it mandatorily needs the server to return the 'Content-length:'
   header in its response.

   The problem I am facing is, this header is returned for most kind
   of objects (image/gif, audio/au etc) but not for text/html. Is
   there any special form of the GET request that will force the
   server to return the size of the object (the 'Content-length:'
   header) irrespective of its MIME type ?

As some people have mentioned, the answer is no --- the server doesn't
know the content-length of dynamically generated objects (i.e., those
resulting from processing of server-side includes or invocation of a
CGI script), so it *can't* report content-length for those items.  If
the server is NCSA-type, and the webmaster has declared all *.html
files to be server-side-included, the effect will be as you report ---
no HTML files will have content-length reported.

(In principle, the server could do a prepass over the file to see if
any inclusion directives are present, and report content-length if
there aren't any; in practice, it doesn't).

rst

eturn-Path: klute@heike.nads.de 
Return-Path: <klute@heike.nads.de>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA02508; Mon, 8 May 1995 10:47:13 +0500
Received: from LCS.MIT.EDU (mintaka.lcs.mit.edu) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA04234; Mon, 8 May 1995 10:47:05 +0500
Received: from heike.nads.de by MINTAKA.LCS.MIT.EDU id aa21538;
          8 May 95 10:46 EDT
Received: from heike (localhost [127.0.0.1]) by heike.nads.de (8.6.12/8.6.12) with SMTP id QAA09552; Mon, 8 May 1995 16:36:01 +0200
Message-Id: <199505081436.QAA09552@heike.nads.de>
To: F.O.Estensen@newcastle.ac.uk
Cc: Multiple recipients of list <www-talk@www10.w3.org>, klute@heike.nads.de
Subject: Re: File names 
Reply-To: Rainer Klute <klute@nads.de>
Organization: NADS GmbH, Germany
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
In-Reply-To: Your message of Mon, 08 May 1995 09:43:18 +0500.
             <Pine.SUN.3.91-941213.950508140848.11828A-100000@eata.ncl.ac.uk> 
Date: Mon, 08 May 1995 16:36:01 +0200
From: Rainer Klute <klute@nads.de>
Content-Length: 406

>How can I find the file name of a document with the URL:
>
>http://foo.co.uk/

You can't. Not shouldn't you. And perhaps it is not even a file.

Best regards
Rainer Klute

  Dipl.-Inform. Rainer Klute        NADS - Advertising on nets
  NADS GmbH
  Emil-Figge-Str. 80                Tel.: +49 231 9742570
D-44227 Dortmund                    Fax:  +49 231 9742571

            <http://www.nads.de/~klute/>
eturn-Path: dmk@allegra.att.com 
Return-Path: <dmk@allegra.att.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA02783; Mon, 8 May 1995 10:51:04 +0500
Received: from research.att.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA04274; Mon, 8 May 1995 10:50:46 +0500
Message-Id: <9505081450.AA04274@www10.w3.org>
Received: by research.att.com; Mon May  8 09:56 EDT 1995
Received: from aleatory.tempo.att.com by allegra.tempo.att.com; id AA20087; Mon, 8 May 95 09:56:39 EDT
Date: Mon, 8 May 95 09:56:31 EDT
From: dmk@allegra.att.com (Dave Kristol)
To: jcaron@pressimage.fr
Subject: Re: your mail
Cc: www-talk@www10.w3.org
Content-Length: 2499

jcaron@pressimage.fr (Jacques Caron) said:
  > Hey, this gets on my nerves. When will someone do something about it? It's so
  > easy to add a f...ing "Host: " or "Full-URI: " header that would enable us
  > to do that without such a hack. Multiple IPs per host _is_ a hack. It can
  > only be done on a small number of OSes, and it is a real waste of IP
  > addresses. I _thought_ we were running out of IP address space. Looks like
  > you guys just want to use it still more quickly!
  > 
  > I have to agree that the guy that came up with this was clever, as this
  > works with the installed base, but such a small modification of the
  > protocol would be _sooooo_ easy to implement in both clients and servers
  > that I think that every single client and server existing would be upgraded
  > in a matter of weeks, and that the installed base would in a few months be
  > 80 to 90% converted.
  > 
  > One would still need a "choose which home page you want" page for the case
  > when the server does not know who was in fact selected, and that would work
  > for everybody till he switches to a newer client.
  > 
  > So, why, _why_, *why*, WHY? A single line! It's so easy!
  > 
  > The worst is, it's been discussed a number of times, everytime everyone
  > agrees and says, OK, that's a good idea, let's do it, and then nobody
  > moves. What should I do? Send every web browser and server author a
  > personal mail to ask him to do it?

Yes, it has been discussed, and it is a good idea.  The problem is
deployment.  Suppose you're running a server for companies X and Y.
The server gets a request via an old client program, and there's no
Host: or Full-URI:  header.  What should the server do?

Here are some possibilities, all bad:
1) Always return the page for company X [Y].  Obviously this surprises
the customers of company Y [X].
2) Return a page that says "Follow this link to company X and this link
to company Y."  But company X may not like to be mentioned in the same
sentence as company Y.
3) Return an error.

I suppose one approach is to require conformance by browsers by date X.
Servers are not an issue -- a server only cares if it wants to offer a
service that depends on the new headers, and it might plan to do so,
say, at X plus six months.  There's still the issue of how to deal with
old browsers.  (Not everyone can or will get a fresh, shiny new browser
every couple of months.)  A server could deny them access, but that's
not very friendly.

Dave Kristol
eturn-Path: jared@math.hmc.edu 
Return-Path: <jared@math.hmc.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA17510; Mon, 8 May 1995 12:49:42 +0500
Received: from aslan.math.hmc.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA05503; Mon, 8 May 1995 12:49:38 +0500
Received: (from jared@localhost) by aslan.math.hmc.edu (8.6.8.1/8.6.7) id JAA03728; Mon, 8 May 1995 09:49:36 -0700
Date: Mon, 8 May 1995 09:49:36 -0700
Message-Id: <199505081649.JAA03728@aslan.math.hmc.edu>
From: Jared Rhine <Jared_Rhine@hmc.edu>
To: rst@ai.mit.edu
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Content-lengths of dynamic objects
References: <9505081427.AA02848@volterra>
X-Attribution: JRhine
X-Uri: <URL:http://www.hmc.edu/~jared/home>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Length: 1613

[Citation date: Mon, 8 May 1995 10:56:06 +0500]
RST == Robert S Thau <rst@ai.mit.edu>

  RST> [...] The server doesn't know the content-length of dynamically
  RST> generated objects (i.e., those resulting from processing of
  RST> server-side includes or invocation of a CGI script), so it *can't*
  RST> report content-length for those items.

I'd like to pick a minor nit here; while your analysis of Srivatsa's problem
is likely correct (server-side includes turned on for html files), I would
point out that the phrase "dynamic object" applies to more than just
includes and CGI scripts.  Reasonable implementations of dynamic objects
should not have this problem.  My web is 100% dynamic, but always (I hope)
reports correct content-lengths (as well as Title and Links, as well); it is
simply an implementation issue.

For the curious, I've done this by having all objects implement two standard
methods: 'compute_lastmod' and 'render'.  When I want to render an object, I
first instantiate the object, and invoke 'compute_lastmod' on it.  I compare
the return value with the timestamp of an on-disk copy of the last
rendering.  If they are the same, I simply serve the file off disk; if not,
I invoke method 'render', take the output of that and dump it in the disk
file, touch the file appropriately, and then have the server serve the file
off disk.  Dynamic objects with local render caching, with the added bonus
of correct Content-Lengths.

-- 
Jared_Rhine@hmc.edu / HMC / <URL:http://www.hmc.edu/~jared/home>

"A hundred thousand lemmings can't be wrong."
        -- attributed to Larry Sheldon, Jr.
eturn-Path: TROTH@ua1vm.ua.edu 
Return-Path: <TROTH@ua1vm.ua.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA01870; Mon, 8 May 1995 14:41:30 +0500
Received: from UA1VM.UA.EDU by www10.w3.org (5.0/NSCS-1.0S) 
	id AA06756; Mon, 8 May 1995 14:39:17 +0500
Message-Id: <9505081839.AA06756@www10.w3.org>
Received: from UA1VM.UA.EDU by UA1VM.UA.EDU (IBM VM SMTP V2R2)
   with BSMTP id 6087; Mon, 08 May 95 13:39:27 CDT
Received: from UA1VM.UA.EDU (NJE origin TROTH@UA1VM) by UA1VM.UA.EDU (LMail
 V1.2a/1.8a) with BSMTP id 4636; Mon, 8 May 1995 13:39:27 -0500
Mime-Version: 1.0
Content-Type: text/plain
X-Mail-User-Agent: MAILBOOK/90.01.01
Date:         Mon, 08 May 95 13:38:58 CDT
From: Rick Troth <TROTH@ua1vm.ua.edu>
Subject:      Re: your mail
To: Multiple recipients of list <www-talk@www10.w3.org>
In-Reply-To:  Message of Mon, 8 May 1995 11:27:56 +0500 from
 <dmk@allegra.att.com>
Content-Length: 958

>Here are some possibilities, all bad:
>1) Always return the page for company X [Y].  Obviously this surprises
>the customers of company Y [X].
>2) Return a page that says "Follow this link to company X and this link
>to company Y."  But company X may not like to be mentioned in the same
>sentence as company Y.
>3) Return an error.

        Dave illustrates the problem(s) pretty well.

        I don't understand why we're so hung up on

                http://X.mythical.host.com/
                http://Y.mythical.host.com/

        Why can't we go directly to  "non home"  or  sub-pages?
Why can't we do things like this:

                http://service.provider.net/home_page_X
                http://service.provider.net/home_page_Y

        Perhaps I've forgotten why,  if the original poster indicated.

        Ramani?   What was the problem with this approach?

--
Rick Troth <troth@ua1vm.ua.edu>, Houston, Texas, USA
http://ua1vm.ua.edu/~troth/
eturn-Path: rupesh@altair.shaktiweb.com 
Return-Path: <rupesh@altair.shaktiweb.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA04699; Mon, 8 May 1995 15:18:33 +0500
Received: from altair.shaktiweb.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA07056; Mon, 8 May 1995 15:16:19 +0500
Received: (from rupesh@localhost) by altair.shaktiweb.com (8.6.9/8.6.9) id MAA02360; Mon, 8 May 1995 12:15:20 -0700
Date: Mon, 8 May 1995 12:15:20 -0700 (PDT)
From: Rupesh Kapoor <rupesh@altair.shaktiweb.com>
To: www-talk@www10.w3.org
Subject: Parsing query string in perl
In-Reply-To: <1362.9505031021@afs.mcc.ac.uk>
Message-Id: <Pine.BSD/.3.91.950508121328.2352A-100000@altair.shaktiweb.com>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 594

Hi,
	Can anyone give me a pointer to perl code that parses a CGI query 
string, similar to "unescape_url" etc in url.c?

Thanx
                             _\\|//_ 
                             ( O-O )
--------------------------o00--(_)--00o------------------------------
Rupesh Kapoor		   	   email: rupesh@shaktiweb.com
PARSEC Communications            details: finger / HTTP

MOTD:  I haven't lost my mind, it's backed up on some tape somewhere.
--------------------------o00--(_)--00o------------------------------
                            _( O-O )_
                              //|\\

eturn-Path: TROTH%ua1vm.ua.edu@www10.w3.org 
Return-Path: <TROTH%ua1vm.ua.edu@www10.w3.org>
Received: from mintaka.lcs.mit.edu by www19 (5.0/NSCS-1.0S) 
	id AC02164; Mon, 8 May 1995 18:05:23 +0500
Received: from www10.w3.org by MINTAKA.LCS.MIT.EDU id aa25346;
          8 May 95 16:44 EDT
Received: from UA1VM.UA.EDU by www10.w3.org (5.0/NSCS-1.0S) 
	id AA07072; Mon, 8 May 1995 15:17:57 +0500
Message-Id: <9505081917.AA07072@www10.w3.org>
Received: from UA1VM.UA.EDU by UA1VM.UA.EDU (IBM VM SMTP V2R2)
   with BSMTP id 6027; Mon, 08 May 95 14:18:04 CDT
Received: from UA1VM.UA.EDU (NJE origin TROTH@UA1VM) by UA1VM.UA.EDU (LMail
 V1.2a/1.8a) with BSMTP id 5752; Mon, 8 May 1995 14:18:04 -0500
Mime-Version: 1.0
Content-Type: text/plain
X-Mail-User-Agent: MAILBOOK/90.01.01
Date:         Mon, 08 May 95 13:42:06 CDT
From: Rick Troth <TROTH@ua1vm.ua.edu>
Subject:      Re: HTTP Content-length question...
To: Henrik Frystyk Nielsen <frystyk@w3.org>,
        Multiple recipients of list <www-talk@www10.w3.org>
In-Reply-To:  Message of Sun, 7 May 1995 16:54:07 +0500 from <frystyk@w3.org>
Content-Length: 0

>> The problem I am facing is, this header is returned for most kind of
>> objects (image/gif, audio/au etc) but not for text/html.    ...

>If this is the case then it is a server bug! The Content-Length header
>should be sent whenever possible in a full response, and it should certainly
>not have anything to do with the media type in which the object is rendered.

        NO,  it's not as much of a bug as you might at first suspect.
Objects which are text/html should really be converted from LF line
delimiters to CR/LF line delimiters.   That takes more work than just a
quick stat(),  which is sufficient for image/gif and audio/au  (both of
which are sent out as-is in binary).   It is unfortunate that a number
of servers are  "lazy"  and don't  "canonicalize"  text/html correctly.

>I can imagine that the problems can be due to server side includes in the
>documents or dynamic documents generated by scripts.

        Well, yes.   That too.

>Henrik Frystyk                                          frystyk@W3.org
>World-Wide Web Consortium,                              Tel + 1 617 258 8143
>MIT/LCS, NE43-356					Fax + 1 617 258 8682
>77 Massachusetts Avenue
>Cambridge MA 02154, USA
>
>

--
Rick Troth <troth@ua1vm.ua.edu>, Houston, Texas, USA
http://ua1vm.ua.edu/~troth/
eturn-Path: TROTH@ua1vm.ua.edu 
Return-Path: <TROTH@ua1vm.ua.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA03748; Mon, 8 May 1995 18:12:44 +0500
Received: from UA1VM.UA.EDU by www10.w3.org (5.0/NSCS-1.0S) 
	id AA08541; Mon, 8 May 1995 18:10:17 +0500
Message-Id: <9505082210.AA08541@www10.w3.org>
Received: from UA1VM.UA.EDU by UA1VM.UA.EDU (IBM VM SMTP V2R2)
   with BSMTP id 6185; Mon, 08 May 95 17:10:27 CDT
Received: from UA1VM.UA.EDU (NJE origin TROTH@UA1VM) by UA1VM.UA.EDU (LMail
 V1.2a/1.8a) with BSMTP id 2042; Mon, 8 May 1995 17:10:27 -0500
Mime-Version: 1.0
Content-Type: text/plain
X-Mail-User-Agent: MAILBOOK/90.01.01
Date:         Mon, 08 May 95 17:00:37 CDT
From: Rick Troth <TROTH@ua1vm.ua.edu>
Subject:      Re: User authentication
To: Multiple recipients of list <www-talk@www10.w3.org>
Content-Length: 1641

        I've been considering  "agent authentication"  for a different
protocol  (other than HTTP or SMTP or NNTP).   It seems  (and someone
suggested this to me)  that we need a general purpose user authentication
mechanism,  something that several protocols could consult.   IDENT is
available  [I may get flamed for even mentioning it]  but isn't  "real
authentication".   Perhaps something stronger?   But what about the
range of administrative domains?   Why (HOW) would you trust  foo.com
to confirm that I am  troth@foo.com  when you don't trust  foo.com
for anything else?   (no shared trust;  no mutual trust)

        I can see a client that tells a server  "I'm acting on behalf of
so-and-so;  check me out for yourself"  where the server then takes a
challenge/response pair and runs that against the  client host.
I don't think this is any stronger than IDENT.

        IDENT is pretty good  (but not 100%)  FOR LOGGING.   If you
check a socket and the  client host  says,  "it's  troth  on this end",
then you can defer the real authentication to  the client host.
That is,  if you don't trust the  client host  at all,  fine.
But if somehow you can trust the  client host,  then you can
(within reason)  trust the IDENT info from it.   (forgery here is
possible but somewhat more difficult than with raw SMTP or NTTP)

        I don't see any way to do  "real authentication"  without
using public key electronic signatures,  and I question whether or not
we need something that strong to eliminate news and mail forgery.

        Thoughts?

--
Rick Troth <troth@ua1vm.ua.edu>, Houston, Texas, USA
http://ua1vm.ua.edu/~troth/
eturn-Path: frystyk%w3.org@www10.w3.org 
Return-Path: 96
Received: from mintaka.lcs.mit.edu by www19 (5.0/NSCS-1.0S) 
	id AH11126; Mon, 8 May 1995 20:29:20 +0500
Received: from www10.w3.org by MINTAKA.LCS.MIT.EDU id aa24963;
          8 May 95 16:11 EDT
Received: from www20 (www20.w3.org) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA07508; Mon, 8 May 1995 16:02:12 +0500
Received: by www20 (5.0/NSCS-1.0S) 
	id AA24306; Mon, 8 May 1995 16:00:40 +0500
Date: Mon, 8 May 1995 16:00:40 +0500
From: frystyk@w3.org (Henrik Frystyk Nielsen)
Message-Id: <9505082000.AA24306@www20>
To: www-talk@w3.org, TROTH@ua1vm.ua.edu
Subject: Re: HTTP Content-length question...
Reply-To: Henrik Frystyk Nielsen <frystyk@w3.org>
X-Sun-Charset: US-ASCII
Content-Length: 1168


> >If this is the case then it is a server bug! The Content-Length header
> >should be sent whenever possible in a full response, and it should certainly
> >not have anything to do with the media type in which the object is rendered.
> 
>         NO,  it's not as much of a bug as you might at first suspect.
> Objects which are text/html should really be converted from LF line
> delimiters to CR/LF line delimiters.   That takes more work than just a
> quick stat(),  which is sufficient for image/gif and audio/au  (both of
> which are sent out as-is in binary).   It is unfortunate that a number
> of servers are  "lazy"  and don't  "canonicalize"  text/html correctly.

This is prefectly OK, as HTTP/1.0 defines its own `canonicalization' of
type `text' and `application' where CR, LF, or any octet sequence defined
by a character set all represent the equivalent of CRLF, though only in the
entity-body!


-- cheers --

Henrik Frystyk                                          frystyk@W3.org
World-Wide Web Consortium,                              Tel + 1 617 258 8143
MIT/LCS, NE43-356					Fax + 1 617 258 8682
77 Massachusetts Avenue
Cambridge MA 02154, USA


eturn-Path: othello!alvin@uunorth.north.net 
Return-Path: <othello!alvin@uunorth.north.net>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA18193; Mon, 8 May 1995 21:36:08 +0500
Received: from uunorth.uunorth.north.net (uunorth.north.net) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA10671; Mon, 8 May 1995 21:36:01 +0500
Received: from othello by uunorth.uunorth.north.net with uucp
	(Smail3.1.29.1 #3) id m0s8f4i-0006M5C; Mon, 8 May 95 21:31 CDT
Received: by eyepoint.com id <111>; Mon, 8 May 1995 21:31:38 -0400
Subject: Re: your mail
From: Alvin Starr <alvin@eyepoint.com>
To: uunorth!www10.w3.org!www-talk
Date: 	Mon, 8 May 1995 21:31:37 -0400
In-Reply-To: <m0s8E13-0006LnC@uunorth.uunorth.north.net>; from "www10.w3.org!www-talk" at May 7, 95 5:37 pm
X-Mailer: ELM [version 2.3 PL11]
Message-Id: <95May8.213138edt.111@eyepoint.com>
Content-Length: 1347

> Could anyone describe the pros and cons of various systems (operating 
> system, hardware/software, connections, etc.) for setting up a Web Server 
> for home or office.
> 
> Is a mac better than windows NT or Sparcstation?  For the office we have 
> direct access to the internet and we have one Sparcstation 20 and one Sparc 
> 5.  We may or may not be able to use those for internet apps.  Should be 
> purchase another Sparc or buy a mac...windows NT?
> 
> For the home what would be best?
> 
If you are not going to make heavy use of the web server it may be a good
idea to use Linux on a 386 clone PC. The price for linux is right, about
$20-$50 for a CD. The performance is acceptable, Linux will run quite happly
on an 8M system. 

If you have a reason to want to minimize your support issues then it may be
worth your while to go with a sparc solution since you allready have 2.

Windows NT will work but the cost/performace of NT is about 1/2 of that
of linux. Also the off the shelf software for NT is nowhere as complete as
that available for Linux/Solaris.

I have installed Internet services on the 3 above OS's and am baseing my 
comments on experiance.

Mac I have no idea about

-- 
Alvin Starr                   ||   voice: (905)513-6717
Eyepoint Inc.                 ||   fax:   (905)513-6718
alvin@eyepoint.com            ||
eturn-Path: bsobilof@inet.ed.gov 
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA29456; Mon, 8 May 1995 23:35:55 +0500
Received: from inet.ed.gov by www10.w3.org (5.0/NSCS-1.0S) 
	id AA11575; Mon, 8 May 1995 23:35:54 +0500
Return-Path: <bsobilof@inet.ed.gov>
Message-Id: <v02120c05abd48dd36447@[168.143.7.200]>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Mon, 8 May 1995 23:37:13 -0400
To: www-talk@www10.w3.org
From: bsobilof@inet.ed.gov (Blake Sobiloff)
Subject: Re: your mail
Content-Length: 1612

At 1:36 PM 5/8/95, Alvin Starr wrote:
>Mac I have no idea about

I'll chime in on this one. :-) I've setup and maintained NCSA, Cern, WN,
and Netscape httpd systems running under SunOS, Solaris, Linux (Slackware),
and MacHTTP under the MacOS. By far, the easiest server to get up and
running was the MacOS with MacHTTP (now called WebStar). MacHTTP can do
just about everything the other httpds can do, but it is a piece of cake to
manage (uses forms just like the commercial Netscape daemon) and its a heck
of a lot cheaper than a UNIX box.

The UNIX-variants were/are more involved and really require someone with
good UNIX knowledge to be able to run the systems. It's harder to find (and
costs more to keep) someone with UNIX skills than Mac skills. And frankly,
until you're dishing out well over a million hits a week, it really doesn't
pay for most folks to futz with the UNIX-based systems -- especially if
they don't already have in-house UNIX experience. If there's a feature with
a particular UNIX daemon that the Mac version doesn't have, and you're
willing to pay the premium, go ahead and go UNIX. But if you just want to
get some pages up cheaply, the Mac is hard to beat (and I'm not even going
to bring in arguments about RAICs (redundant arrays of inexpensive
computers) :-).

--
Blake Sobiloff                             <bsobilof@inet.ed.gov>
Internet Systems Analyst/Webmaster     (speaking only for myself)
Decision Systems Technologies, Inc.           Voice: 301/441-3377
Greenbelt, MD  20770  USA                       Fax: 301/441-4571
                  http://inet.ed.gov/~bsobilof/


eturn-Path: masinter@parc.xerox.com 
Return-Path: <masinter@parc.xerox.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA03840; Tue, 9 May 1995 00:41:35 +0500
Received: from alpha.xerox.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA12023; Tue, 9 May 1995 00:41:33 +0500
Received: from golden.parc.xerox.com ([13.1.100.139]) by alpha.xerox.com with SMTP id <14493(4)>; Mon, 8 May 1995 21:41:20 PDT
Received: by golden.parc.xerox.com id <2761>; Mon, 8 May 1995 21:41:12 -0700
To: Jared_Rhine@hmc.edu
Cc: www-talk@www10.w3.org
In-Reply-To: Jared Rhine's message of Mon, 8 May 1995 01:23:08 -0700 <199505081649.JAA03728@aslan.math.hmc.edu>
Subject: Re: Content-lengths of dynamic objects
From: Larry Masinter <masinter@parc.xerox.com>
Sender: Larry Masinter <masinter@parc.xerox.com>
Fake-Sender: masinter@parc.xerox.com
Message-Id: <95May8.214112pdt.2761@golden.parc.xerox.com>
Date: Mon, 8 May 1995 21:41:10 PDT
Content-Length: 573

Caching is a bad idea for something that changes length every time
it's called. (Oh, quote-of-the-day, for example, or just the output
from a request form.) Waiting to cache something that will never be
needed again doesn't optimize anything, and adds unnecessary latency
to the transaction while the server generates all of the data.

I've noticed web sites that give the appearance of rapid response from
search requests by returning immediately the header (complete with
large inline usually-cached-by-the-client images) before actually
completing the search results.


eturn-Path: mvanheyn@pellet.spry.com 
Return-Path: <mvanheyn@pellet.spry.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AC04332; Tue, 9 May 1995 00:48:08 +0500
Received: from homer.spry.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA12052; Tue, 9 May 1995 00:48:02 +0500
Received: from pellet.spry.com (pellet [198.185.1.189]) by homer.spry.com (8.6.9/8.6.9) with ESMTP id VAA16554; Mon, 8 May 1995 21:48:34 -0700
From: marcvh@spry.com (Marc VanHeyningen)
To: TROTH@ua1vm.ua.edu
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Re: HTTP Content-length question... 
In-Reply-To: Your message of "Mon, 08 May 1995 20:36:13 +0500."
             <9505081917.AA07072@www10.w3.org> 
Date: Mon, 08 May 1995 21:47:51 -0700
Message-Id: <13339.799994871@pellet.spry.com>
Sender: mvanheyn@pellet.spry.com
Content-Length: 1400

Thus wrote: Rick Troth
>        NO,  it's not as much of a bug as you might at first suspect.
>Objects which are text/html should really be converted from LF line
>delimiters to CR/LF line delimiters.   That takes more work than just a
>quick stat(),  which is sufficient for image/gif and audio/au  (both of
>which are sent out as-is in binary).   It is unfortunate that a number
>of servers are  "lazy"  and don't  "canonicalize"  text/html correctly.

Well, it is unfortunate that more serious thought wan't given to
to the problems involved with this, perhaps.  Canonicalization of textual
line breaks, however, is no longer required under HTTP in general; instead,
all HTTP software is required to be able to recognize various different
common line endings when receiving bodies.  The result is not fantastic
or elegant but it's workable.

(Not that it matters too much with HTML, since aside from in <PRE> all
whitespace is created equal.  But, since HTML and HTTP are orthogonal,
it's still an issue.)

>>I can imagine that the problems can be due to server side includes in the
>>documents or dynamic documents generated by scripts.

Of course; there are limits on how much buffering one wants to do.
I'll use this moment to revisit my personal pet idea, which is that the
headers and status line and other metadata really should go after the
body of the response, not before it. :-)

- Marc
eturn-Path: narnett@verity.com 
Return-Path: <narnett@verity.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AC07150; Tue, 9 May 1995 01:19:53 +0500
Received: from verity.com (unknown-143-5.verity.com) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA12289; Tue, 9 May 1995 01:19:51 +0500
Received: from [192.187.143.12] (portanick.verity.com) by verity.com (4.1/SMI-4.1_Verity-Main-950202)
	id AA13861; Mon, 8 May 95 22:19:03 PDT
X-Sender: narnett@hawaii.verity.com
Message-Id: <abd4a234010210044429@[192.187.143.12]>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Mon, 8 May 1995 22:19:29 -0700
To: www-talk@www10.w3.org, www-html@www10.w3.org, newshound-list@best.com,
        free-market@ar.com, machttp-talk@academ.com, /CN=robots/@nexor.co.uk,
        web4lib@library.berkeley.edu, www-lib@www10.w3.org,
        owner-www-security@ns2.rutgers.edu
From: narnett@verity.com (Nick Arnett)
Subject: Verity Internet Virtual Library update; Verity/Netscape announcement
Content-Length: 1097

As some of you may know, though we haven't advertised it for a while, the
lists to which I'm sending this message, as well as many other Web-related
mailing lists, are browsable at [http://vlibmail.verity.com/www.html] (also
known as asearch.mccmedia.com).  We've been working on making the
text-to-hypertext better and more reliable, while also adding increased
search and retrieval capabilities, with much more to come.

Each list has a search page that searches only that list; the entire
collection plus about 100 Web-related sites are searchable at
[http://www.verity.com/http://www.verity.com/vlibsearch.html].  (If you get
this message in the next few hours, you'll find that its index is being
rebuilt; we're getting ready to make a major announcement with Netscape
tomorrow.

See <a href="http://vlibmail.verity.com/PR/950508ns.html">Verity and
Netscape Team Up to Bring Topic Agent Technology to the Internet</a>.  We
also have an announcement with Microsoft, with regard to Exchange.

Nick Arnett
World Wide Web Product Marketing Manager
Verity Inc.
narnett@verity.com
(415) 960-7600


eturn-Path: jeremie@netins.net 
Return-Path: <jeremie@netins.net>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB07223; Tue, 9 May 1995 01:25:12 +0500
Received: from worf.infonet.net by www10.w3.org (5.0/NSCS-1.0S) 
	id AA12302; Tue, 9 May 1995 01:25:10 +0500
Received: from s2173.netins.net (s2173.netins.net [167.142.102.173]) by worf.infonet.net with SMTP id AAA04832 for <www-talk@www10.w3.org>; Tue, 9 May 1995 00:25:06 -0500
Message-Id: <199505090525.AAA04832@worf.infonet.net>
X-Sender: jeremie@netins.net
X-Mailer: Windows Eudora Version 1.4.4
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Tue, 09 May 1995 00:24:37 -0400
To: www-talk@www10.w3.org
From: jeremie@netins.net (Jeremie Miller)
Subject: Reasonable new feature?
Content-Length: 1768

I have done a good deal of HTML page development for various different
reasons, and have found this idea to be useful in many cases.  I propose an
addition to HTML 3.0 that would allow small windows to pop up when the mouse
is overtop of some keywords, containing some information.  Example:

<body>
This is a new feature of
<window show="HyperText Markup Language">HTML</window>.
</body>

OR

<body>
This is a new feature of
<a html="HyperText Markup Language" window>HTML</a>.
</body>

Rendered

This is a new feature of HTML
                             \_________________________
                             |HyperText Markup Language|
                              ^^^^^^^^^^^^^^^^^^^^^^^^^

There are many times when this would be useful to present information.  From
my experience in programming, it does not seem a difficult task to add such
a feature in a browser.  The tag is simple and aids the general puprpose of
HTML.  I would appresiate any feedback and discussion on it's possibilities
or functionality.

Jeremie Miller
jeremie@netins.net
------------0 -----------Jeremie Miller------------------------* -* -* -------
-----* ----------------jeremie@netins.net-------------------0 ----------* ----
---* ------* -------------------------------------------------------------* --
--* ----* -* --* * ----** ------* * ------** ---** --** ---* -------* * ----* 
---* * ---* -* ---* --* -* ---* ---* ---* -* -* -* -* * ---* ------* ---* ---*
---------* --* ---* --* --* --* ---* --* --* * --* * -* --* * -----* ---* ---*
--------* -----* -----* --* ----* -----* ---* ----* ---* -* --* -----* -----* 
-* ----* --* * ---* * ------* * ---* * ------------------* ------* * ---* -* -
---* * -----------------------------------------------------------------------

eturn-Path: klute@heike.nads.de 
Return-Path: <klute@heike.nads.de>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AC16430; Tue, 9 May 1995 02:56:59 +0500
Received: from heike.nads.de by www10.w3.org (5.0/NSCS-1.0S) 
	id AA13002; Tue, 9 May 1995 02:56:57 +0500
Received: from heike (localhost [127.0.0.1]) by heike.nads.de (8.6.12/8.6.12) with SMTP id IAA11381; Tue, 9 May 1995 08:56:07 +0200
Message-Id: <199505090656.IAA11381@heike.nads.de>
To: masinter@parc.xerox.com
Cc: Multiple recipients of list <www-talk@www10.w3.org>, klute@heike.nads.de
Subject: Re: Content-lengths of dynamic objects 
Reply-To: Rainer Klute <klute@nads.de>
Organization: NADS GmbH, Germany
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
In-Reply-To: Your message of Tue, 09 May 1995 01:01:42 +0500.
             <95May8.214112pdt.2761@golden.parc.xerox.com> 
Date: Tue, 09 May 1995 08:56:06 +0200
From: Rainer Klute <klute@nads.de>
Content-Length: 886

>Caching is a bad idea for something that changes length every time
>it's called. (Oh, quote-of-the-day, for example, or just the output
>from a request form.)

That's what the Expires header is good for, isn't it? See
<http://www.nads.de/NADS/Ueberblick/heute> for an example. This URL
gives you an overview of events in Düsseldorf and about for the
very day. It expires at the end of the day (local time in Germany).

(In fact, it is just a 302 Redirect message pointing to the real
document. To make the CERN httpd cache redirections, see my patch
under
<http://www.nads.de/NADS/Public/WWWTechnik/WWWDaemon-patch.01.gz>.)

Best regards
Rainer Klute

  Dipl.-Inform. Rainer Klute        NADS - Advertising on nets
  NADS GmbH
  Emil-Figge-Str. 80                Tel.: +49 231 9742570
D-44227 Dortmund                    Fax:  +49 231 9742571

            <http://www.nads.de/~klute/>
eturn-Path: mcclurea@nag.cs.colorado.edu 
Return-Path: <mcclurea@nag.cs.colorado.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA17644; Tue, 9 May 1995 03:13:18 +0500
Received: from nag.cs.colorado.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA13131; Tue, 9 May 1995 03:13:16 +0500
Received: (from mcclurea@localhost) by nag.cs.colorado.edu (8.6.10/8.6.10) id BAA20417; Tue, 9 May 1995 01:13:11 -0600
From: "Adam T. McClure" <mcclurea@nag.cs.colorado.edu>
Message-Id: <199505090713.BAA20417@nag.cs.colorado.edu>
Subject: How bout types of colored links?
To: jeremie@netins.net
Date: Tue, 9 May 1995 01:13:11 -0600 (MDT)
Cc: www-talk@www10.w3.org
In-Reply-To: <199505090525.AAA04832@worf.infonet.net> from "Jeremie Miller" at May 9, 95 02:27:19 am
X-Mailer: ELM [version 2.4 PL23]
Content-Type: text
Content-Length: 1382      

I think Jeremy Miller's idea of having popup menus is an excellent
one, and wanted to ask this group if an idea I had is worth considering.

One of the things I ran into in creating a teching tool recently was
that I couldn't denote a particular thread.  I wanted to target the
paper for multiple audiences and provide support for less technically-
inclined audiences without compromising the flow of the paper.  It
seemed logical to me to define a named type of link that has a particular
color associated with it.  All the references to definitions could be
in blue, the links to other elements of the discussion in red, and 
purple would jump off my server to an off-site page.  All visited
links could be tan if I felt like it.  There's currently a method in
Netscape 1.1 to define different colors for visited and new links,
as well as redefining the standard text color, but there's no way
to have your own labelled links.  

Am I way behind the times, or is this a legitimate item?

-- 
_/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/
Adam T. McClure    	Integrated Teaching & Learning project
mcclurea@colorado.edu	University of Colorado-Boulder

    <a href="http://itldemo.colorado.edu/rootitl.html></a>

You know what I like about standards?  There's so many to choose from.
_/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/ _/
eturn-Path: jared@math.hmc.edu 
Return-Path: <jared@math.hmc.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AC19214; Tue, 9 May 1995 03:31:06 +0500
Received: from aslan.math.hmc.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA13246; Tue, 9 May 1995 03:31:00 +0500
Received: (from jared@localhost) by aslan.math.hmc.edu (8.6.8.1/8.6.7) id AAA09553; Tue, 9 May 1995 00:30:55 -0700
Date: Tue, 9 May 1995 00:30:55 -0700
Message-Id: <199505090730.AAA09553@aslan.math.hmc.edu>
From: Jared Rhine <Jared_Rhine@hmc.edu>
To: Larry Masinter <masinter@parc.xerox.com>
Cc: www-talk@www10.w3.org
Subject: Re: Content-lengths of dynamic objects
References: <199505081649.JAA03728@aslan.math.hmc.edu>
  <95May8.214112pdt.2761@golden.parc.xerox.com>
X-Attribution: JRhine
X-Uri: <URL:http://www.hmc.edu/~jared/home>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Length: 986

[Citation date: Mon, 8 May 1995 21:41:10 PDT]
LM == Larry Masinter <masinter@parc.xerox.com>

  LM> Caching is a bad idea for something that changes length every time
  LM> it's called. (Oh, quote-of-the-day, for example, or just the output
  LM> from a request form.) Waiting to cache something that will never be
  LM> needed again doesn't optimize anything, and adds unnecessary latency
  LM> to the transaction while the server generates all of the data.

Granted, but I'm not entirely sure of the connection you're trying to make.
For the most part, my dynamically generated objects don't change length or
content and it's ok to cache them.  For the ones that do change every time,
I add an Expires header to the response to let know caching mechanisms now
not to cache it.  Unfortunately, my tests indicate that Netscape doesn't
handle the Expires header properly :(

-- 
Jared_Rhine@hmc.edu / HMC / <URL:http://www.hmc.edu/~jared/home>

"To live is to war with trolls." -- Ibsen
eturn-Path: klute@heike.nads.de 
Return-Path: <klute@heike.nads.de>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA00340; Tue, 9 May 1995 05:15:18 +0500
Received: from heike.nads.de by www10.w3.org (5.0/NSCS-1.0S) 
	id AA14178; Tue, 9 May 1995 05:15:09 +0500
Received: from heike (localhost [127.0.0.1]) by heike.nads.de (8.6.12/8.6.12) with SMTP id KAA11662; Tue, 9 May 1995 10:30:57 +0200
Message-Id: <199505090830.KAA11662@heike.nads.de>
To: Jared_Rhine@hmc.edu
Cc: Multiple recipients of list <www-talk@www10.w3.org>, klute@heike.nads.de
Subject: Re: Content-lengths of dynamic objects 
Reply-To: Rainer Klute <klute@nads.de>
Organization: NADS GmbH, Germany
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
In-Reply-To: Your message of Tue, 09 May 1995 03:59:52 +0500.
             <199505090730.AAA09553@aslan.math.hmc.edu> 
Date: Tue, 09 May 1995 10:30:57 +0200
From: Rainer Klute <klute@nads.de>
Content-Length: 499

>Unfortunately, my tests indicate that Netscape doesn't
>handle the Expires header properly :(

Not to defend Netscape against anything, but is there a client at
all that handles Expires properly? Not even the CERN proxy does.

Mit freundlichen Grüßen
Rainer Klute

  Dipl.-Inform. Rainer Klute        NADS - Advertising on nets
  NADS GmbH
  Emil-Figge-Str. 80                Tel.: +49 231 9742570
D-44227 Dortmund                    Fax:  +49 231 9742571

            <http://www.nads.de/~klute/>
eturn-Path: bert@let.rug.nl 
Return-Path: <bert@let.rug.nl>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA09663; Tue, 9 May 1995 06:22:38 +0500
Received: from freya.let.rug.nl by www10.w3.org (5.0/NSCS-1.0S) 
	id AA14936; Tue, 9 May 1995 06:22:34 +0500
Received: from grid.let.rug.nl by freya.let.rug.nl with ESMTP
	(1.37.109.15/16.2) id AA212924947; Tue, 9 May 1995 12:22:27 +0200
Message-Id: <199505091022.AA212924947@freya.let.rug.nl>
Received: by grid.let.rug.nl
	(1.37.109.16/16.2) id AA079314946; Tue, 9 May 1995 12:22:26 +0200
From: Bert Bos <bert@let.rug.nl>
Subject: Re: Reasonable new feature?
To: jeremie@netins.net
Date: Tue, 9 May 1995 12:22:25 +0200 (METDST)
Cc: www-talk@www10.w3.org
In-Reply-To: <199505090525.AAA04832@worf.infonet.net> from "Jeremie Miller" at May 9, 95 02:17:18 am
X-Mailer: ELM [version 2.4 PL24]
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Content-Length: 1354      

Jeremie Miller <jeremie@netins.net> writes:

 |I have done a good deal of HTML page development for various different
 |reasons, and have found this idea to be useful in many cases.  I propose an
 |addition to HTML 3.0 that would allow small windows to pop up when the mouse
 |is overtop of some keywords, containing some information.  Example:
 |
 |<body>
 |This is a new feature of
 |<window show="HyperText Markup Language">HTML</window>.
 |</body>
 |
 |OR
 |
 |<body>
 |This is a new feature of
 |<a html="HyperText Markup Language" window>HTML</a>.
 |</body>
 |
 |Rendered
 |
 |This is a new feature of HTML
 |                             \_________________________
 |                             |HyperText Markup Language|
 |                              ^^^^^^^^^^^^^^^^^^^^^^^^^

You're too late, it has already been proposed:-) It now seems likely
that there will be a footnote tag (<FN>) in HTML 3.0, with a suggested
rendering like what you describe. See
<http://www.hpl.hp.co.uk/people/dsr/html3/CoverPage.html> for the
currently proposed syntax. The exact syntax may still change, but the
basic idea is there.


Bert
-- 
                          Bert Bos                      Alfa-informatica
                 <bert@let.rug.nl>           Rijksuniversiteit Groningen
    <http://www.let.rug.nl/~bert/>     Postbus 716, NL-9700 AS GRONINGEN
eturn-Path: randy@zyzzyva.com 
Return-Path: <randy@zyzzyva.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA22977; Tue, 9 May 1995 08:04:10 +0500
Received: from sierra.zyzzyva.com (ppp0.zyzzyva.com) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA15912; Tue, 9 May 1995 08:03:56 +0500
Received: from zyzzyva.com (randy@localhost.zyzzyva.com [127.0.0.1]) by sierra.zyzzyva.com (8.6.11/8.6.11) with ESMTP id GAA24807; Tue, 9 May 1995 06:55:27 -0500
Message-Id: <199505091155.GAA24807@sierra.zyzzyva.com>
To: rich@oester.com
Cc: www-talk@w3.org
Subject: Re: multiple home pages/ips 
In-Reply-To: rich's message of Sun, 07 May 1995 15:10:18 +0500.
         <199505071900.MAA12723@kitty.oester.com> 
X-Uri: http://www.inetnebr.com/zyzzyva/
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Tue, 09 May 1995 06:55:18 -0500
From: Randy Terbush <randy@zyzzyva.com>
Content-Length: 368

> 
> as for the apache hint - unless something has changed, how to do this is not
> documented, and the actual routing manipulations will probably be somewhat
> unix vendor/setup specific; there's at least two ways this can be done.
> 
> rich


Check out <URL: http://www.apache.org/apache/docs/virtual-host>

or

<URL: http://hyperreal.com/apache/docs/virtual-host>

eturn-Path: sarr@citi.umich.edu 
Return-Path: <sarr@citi.umich.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB07200; Tue, 9 May 1995 09:59:34 +0500
Received: from citi.umich.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA17072; Tue, 9 May 1995 09:57:16 +0500
Message-Id: <9505091357.AA17072@www10.w3.org>
Received: from citi.umich.edu by citi.umich.edu for mcclurea@nag.cs.colorado.edu www-talk@www10.w3.org with SMTP; Tue, 09 May 95 09:55:58 -0400
From: Sarr Blumson <sarr@citi.umich.edu>
To: mcclurea@nag.cs.colorado.edu
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Date: Tue, 09 May 1995 09:55:56 -0400
Subject: Re: How bout types of colored links? 
X-Mailer: exmh version 1.5.3 12/28/94
In-Reply-To: Your message of "Tue, 09 May 1995 03:24:30 +0500."
             <199505090713.BAA20417@nag.cs.colorado.edu> 
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Length: 1958


Well, here I go again:

In message <199505090713.BAA20417@nag.cs.colorado.edu>you write:
>One of the things I ran into in creating a teching tool recently was
>that I couldn't denote a particular thread.  I wanted to target the
>paper for multiple audiences and provide support for less technically-
>inclined audiences without compromising the flow of the paper.  It
>seemed logical to me to define a named type of link that has a particular
>color associated with it.  All the references to definitions could be
>in blue, the links to other elements of the discussion in red, and 
>purple would jump off my server to an off-site page.  All visited
>links could be tan if I felt like it.  There's currently a method in
>Netscape 1.1 to define different colors for visited and new links,
>as well as redefining the standard text color, but there's no way
>to have your own labelled links.  

No, No, No!  HTML should describe structure, not rendering.  The 
ability to describe different classes of links is an interesting idea, 
but how those classes are represented is a browser issue.

Something like 10 percent of males in the United States are at least 
pertially color blind; many of the rest of us have vision difficulties 
that are sensitive to the contrast between background and text colors.  
What you are proposing will make your links invisible to many people, 
and because they're described as "green links" rather than "definition 
links" you've made it difficult to configure our browsers so we can.

It is sad that the folks at Netscape have gone off in the direction 
they have, and even sadder that the market place is making them the 
definition of the standard rather than the standards.  But it's still 
the wrong thing.

--------
Sarr Blumson                     sarr@umich.edu
voice: +1 313 764 0253           FAX: +1 313 763 4434
CITI, University of Michigan     http://www.umich.edu/~sarr/
519 W William, Ann Arbor, MI 48103-4943

eturn-Path: habib@world.std.com 
Return-Path: <habib@world.std.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB11958; Tue, 9 May 1995 10:23:20 +0500
Received: from europe.std.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA17436; Tue, 9 May 1995 10:21:03 +0500
Received: from world.std.com by europe.std.com (8.6.12/Spike-8-1.0)
	id KAA03353; Tue, 9 May 1995 10:20:38 -0400
Received: by world.std.com (5.65c/Spike-2.0)
	id AA07716; Tue, 9 May 1995 10:20:37 -0400
Date: Tue, 9 May 1995 10:20:37 +0059 (EDT)
From: Steve H Rose <habib@world.std.com>
Subject: Re: File names
To: Multiple recipients of list <www-talk@www10.w3.org>
In-Reply-To: <Pine.SUN.3.91-941213.950508140848.11828A-100000@eata.ncl.ac.uk>
Message-Id: <Pine.3.89.9505091051.G18201-0100000@world.std.com>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 887

This is server specific.  Some servers, at least, are set up to 
automatically load other file names when they are not specified in your 
URL (e.g. welcome.html).  What browser are you using? (please reply to me 
directly, and I'll see if I can help)

Yours,

Steve Habib Rose

On Mon, 8 May 1995, Frank Olav Estensen wrote:

> How can I find the file name of a document with the URL:
> 
> http://foo.co.uk/
> 
> Is there a methode to get the filename, or do I just have to assume 
> index.html?
> 
> Mr Frank Olav Estensen
> Student, Computing Science part III, University of Newcastle
> WWW   :   http://www.newcasle.ac.uk/~n314974
>           http://www.nordnett.no/~estensen
> E-mail:				Snail-mail:
> F.O.Estensen@newcastle.ac.uk	105 Stratford Road
> F.O.Estensen@nordnett.no        Heaton
> Phone:				Newcastle upon Tyne
> +44 91 224 2355  		NE6 5AS
> +47 75 56 09 48			England
> 
eturn-Path: jkillian@access.digex.net 
Return-Path: <jkillian@access.digex.net>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB17879; Tue, 9 May 1995 11:00:19 +0500
Received: from access5.digex.net by www10.w3.org (5.0/NSCS-1.0S) 
	id AA17954; Tue, 9 May 1995 10:57:44 +0500
Received: from bac06998.slip (bac06998.slip.digex.net) by access5.digex.net with SMTP id AA17911
  (5.67b8/IDA-1.5 for <www-talk@www10.w3.org>); Tue, 9 May 1995 10:56:32 -0400
Message-Id: <199505091456.AA17911@access5.digex.net>
X-Sender: jkillian@access.digex.net
X-Mailer: Windows Eudora Version 1.4.4
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Tue, 09 May 1995 07:56:12 -0700
To: "Ron Wolf" <ron.e.wolf@tsphere.com>, "Dianna Goble" <dmgoble@netcom.com>,
        ken@best.com, "Greg Lindberg" <theg@tripnspin.com>,
        "Multiple recipients of list" <www-talk@www10.w3.org>,
        "Charles Pontious" <charlie@mdli.com>,
        "Andrew Weiss" <aweiss@aimnet.com>,
        "Preston Gordon" <admin@tsphere.com>,
        "George Kellar" <george@tsphere.com>,
        "Charlie Schoenhoeft" <chaz@tsphere.com>
From: jkillian@access.digex.net (Jim Killian)
Subject: Re: Dynamic HTML documents with
Content-Length: 2523

I believe Netscape intends to use this (or something similar to it) for
simple animations techniques as well.

>>> Jim

At 10:54 AM 5/8/95 -0800, Ron Wolf wrote:
>Mail*Link(r) SMTP               Dynamic HTML documents with client pull
>
>Interesting feature here.
>___________Ron
>Ulf Kronman writes:
> > 
> > I had to take a look at the HTML source to find out what this was, and I
> > found a new HTML directive looking something like this:
> > 
> > <META HTTP-EQUIV=REFRESH CONTENT="20; URL=http://[file name].au">
> > 
> > I presume this tells Netscape to download the audio file 20 seconds after
> > the page has been loaded.
> > 
> > Is this nuicance going to be a part of the proposed HTML v 3.0 or is it a
> > Netscape specific extension?
>
>The META tag is HTML 2. Standard stuff. Used for several indexing
>applications, as far as I know.
>
>Sounds like they've chosen to interpret "http-equiv=refresh" to mean
>that the client should go retrieve some stuff after a little
>while. Hmmm... I thought http-equiv was intended for the server to
>spit out HTTP headers when this document is served. This might be
>different from the spirit of the spec, but it's not inconsistent,
>technically.
>
>In any case: the HTML spec is hardly relavent. This is an HTTP-related
>innovation.
>
> > This is AWFULL - this means that the time when *you* choose what you want
> > so see on the web is gone! TV-style advertising is entering the Web!
>
>So don't visit providers that use this feature.
>
>Netscape has enabled a certain class of applications, but ultimately,
>you still choose whether to participate in them or not.
>
>As an information provider, you choose whether you want to rely on such
>experimental features or not too.
>
>Dan
>
>------------------ RFC822 Header Follows ------------------
>Received: by tsphere.com with SMTP;28 Apr 1995 18:09:53 -0800
>Received: from  (localhost) by www19 (5.0/NSCS-1.0S) 
>	id AA06983; Fri, 28 Apr 1995 20:23:05 +0500
>Date: Fri, 28 Apr 1995 20:23:05 +0500
>Message-Id: <9504282104.AA18211@www16>
>Errors-To: procmaster@www19.w3.org
>Reply-To: connolly@w3.org
>Originator: www-talk@mail.w3.org
>Sender: www-talk@www10.w3.org
>Precedence: bulk
>From: connolly@w3.org (Dan Connolly)
>To: Multiple recipients of list <www-talk@www10.w3.org>
>Subject: Dynamic HTML documents with client pull
>X-Listprocessor-Version: 6.0c -- ListProcessor by Anastasios Kotsikonas
>X-Comment: To sign off, send mail to listproc@mail.w3.org with body DEL
>WWW-TALK
>content-length: 1419
>
>
>
>
>

eturn-Path: brian@organic.com 
Return-Path: <brian@organic.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA23549; Tue, 9 May 1995 11:41:27 +0500
Received: from eat.organic.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA18383; Tue, 9 May 1995 11:38:40 +0500
Received: (from brian@localhost) by eat.organic.com (8.6.11/8.6.9) id IAA19318; Tue, 9 May 1995 08:38:58 -0700
Date: Tue, 9 May 1995 08:38:57 -0700 (PDT)
From: Brian Behlendorf <brian@organic.com>
Subject: Re: How bout types of colored links?
To: "Adam T. McClure" <mcclurea@nag.cs.colorado.edu>
Cc: Multiple recipients of list <www-talk@www10.w3.org>
In-Reply-To: <199505090713.BAA20417@nag.cs.colorado.edu>
Message-Id: <Pine.3.89.9505090812.f24326-0100000@eat.organic.com>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 1391

On Tue, 9 May 1995, Adam T. McClure wrote:
> One of the things I ran into in creating a teching tool recently was
> that I couldn't denote a particular thread.  I wanted to target the
> paper for multiple audiences and provide support for less technically-
> inclined audiences without compromising the flow of the paper.  It
> seemed logical to me to define a named type of link that has a particular
> color associated with it.  All the references to definitions could be
> in blue, the links to other elements of the discussion in red, and 
> purple would jump off my server to an off-site page.  All visited
> links could be tan if I felt like it.  There's currently a method in
> Netscape 1.1 to define different colors for visited and new links,
> as well as redefining the standard text color, but there's no way
> to have your own labelled links.  

This feels like a style sheet issue - you have something like

a.class1 = #F00
a.class2 = #0F0

and then

<a href="file1.html" class="class1">This class of link</a>
<a href="file2.html" class="class2">That other class of link</a>

Is this what you want?  Style sheet designers should note the need for 
different colors for visited and not-yet-visited links of course.

	Brian

--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--
brian@organic.com  brian@hyperreal.com  http://www.[hyperreal,organic].com/

eturn-Path: glv@utdallas.edu 
Return-Path: <glv@utdallas.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA27329; Tue, 9 May 1995 12:12:57 +0500
Received: from utdallas.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA18770; Tue, 9 May 1995 12:10:45 +0500
Received: from localname (oblivion.utdallas.edu [129.110.10.23]) by utdallas.edu (8.6.11/8.6.11) with ESMTP id LAA24747; Tue, 9 May 1995 11:07:17 -0500
Message-Id: <199505091607.LAA24747@utdallas.edu>
From: Glenn Vanderburg <glv@utdallas.edu>
To: Rick Troth <troth@ua1vm.ua.edu>
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Re: your mail 
In-Reply-To: Your message of "Mon, 08 May 1995 19:24:54 +0500 (+0500)"
             <9505081839.AA06756@www10.w3.org> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Tue, 09 May 1995 11:06:21 -0500
Sender: glv@utdallas.edu
Content-Length: 1200

Rick Troth writes:
> 
>         I don't understand why we're so hung up on
> 
>                 http://X.mythical.host.com/
>                 http://Y.mythical.host.com/
> 
>         Why can't we go directly to  "non home"  or  sub-pages?
> Why can't we do things like this:
> 
>                 http://service.provider.net/home_page_X
>                 http://service.provider.net/home_page_Y

One reason is that the first form can often be guessed.  I suspect that most
of us have had an experience like this: "Gee, I sure would like to know what
company X is up to these days.  I wonder if they have a Web page?  I suppose
'http://www.X.com/' is worth a try ..."  It's surprising how often this works.

And, to a degree, this is a good thing.  Those of us who want to discourage
broadcast-style advertising on the net see the advantage: if it's easy for
people to find information on a company when they want to, then the passive
style of advertising is more effective.

The key problem here is that we have locators which are being used as names.
People or organizations who are establishing Web presences are justified in
thinking of the URL of their home pages as their Web "names".  

---glv

eturn-Path: ppetit@wct.fr 
Return-Path: <ppetit@wct.fr>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA04263; Tue, 9 May 1995 13:05:34 +0500
Received: from ares.wct.fr (zeus.wct.fr) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA19518; Tue, 9 May 1995 13:02:51 +0500
Received: (from ppetit@localhost) by ares.wct.fr (8.6.9/8.6.9) id TAA11575 for www-talk@www10.w3.org; Tue, 9 May 1995 19:13:24 GMT
From: Patrick Petit <ppetit@wct.fr>
Message-Id: <199505091913.TAA11575@ares.wct.fr>
Subject: Netscape Navigator License
To: www-talk@www10.w3.org
Date: Tue, 9 May 1995 19:13:24 +0000 (GMT)
X-Mailer: ELM [version 2.4 PL23]
Content-Type: text
Content-Length: 518       

Hi,

I'm not sure it's the right place to send that query, but here it goes....


Well, I've been trying to get an answer from moreinfo@netscape.com
without success. Maybe someone here has the answer.

Does Netscape support Navigator's binary license for Linux and
Solaris x86 platforms?

Cheers,

----------------------------------------------------------------
Patrick Petit				Voice: (33) 1 42 36 90 47
World Channel Technologies		Email: ppetit@wct.fr
8, rue Saint Marc 75002 Paris		Fax: (33) 1 45 08 51 96
France

eturn-Path: nazgul@utopia.com 
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA00556; Tue, 9 May 1995 16:36:58 +0500
Received: from utopia.com (utopia.wing.net) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA21769; Tue, 9 May 1995 16:34:43 +0500
Received: from [204.57.39.6] by utopia.com (8.6.12/SONY-4.0MX)
	id QAA11530; Tue, 9 May 1995 16:32:48 -0400
Return-Path: <nazgul@utopia.com>
X-Sender: nazgul@mailhost.utopia.com
Message-Id: <v02110111abd57f5a6e6a@[204.57.39.6]>
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Date: Tue, 9 May 1995 16:34:36 -0400
To: www-talk@www10.w3.org
From: nazgul@utopia.com (Kee Hinckley)
Subject: NCSA Server (1.3) and ScriptAlias problems
Content-Length: 2054

The issue is how to deal with a ScriptAlias that is serving up multiple
HTML files, e.g.:

        ScriptAlias /bgc/ /xxx/bgc/access/
        ScriptAlias /bgc /xxx/bgc/access/

In particular, the problem is that the script can't tell the difference
between being invoked with or without the trailing slash, and the
difference is quite critical. I suspect that there should be a difference
in PATH_INFO, but there isn't. Is that a bug, or is there another better
way of dealing with this?

1. If you don't have the version with the trailing slash, the (non-trivial
number of) people who hit the site without it will get a file not-found.
This isn't the case when the server is handling access, but it is the case
when you are using a ScriptAlias.

2. If you point them both at the same script, the script can't tell whether
it was invoked with or without the slash (I believe there ought to be a
difference in PATH_INFO, but there isn't).

3. The script can do one of two things when faced with a directory name
with no index.html. It can't do a redirect to the index.html file (which
takes up a fair amount of time, and of course additional server resources),
or it can just server up the file.

4. If it serves up the file, the client will get all the relative links
wrong. It's URL will look like "http://www.xxx.bar/foo", it will assume
that "foo" is a file and make all links relative to the link instead of
"foo/". You can try and fix this with a <base href=3D"correcturl"> in the
index.html file, but that doesn't work for all browsers.

5. My solution therefore, is to have a separate script, bound to the
non-slash version of the name, which does nothing but redirect the user to
the correct URL. Ugly, but there it is.

Comments?

Kee Hinckley      Utopia Inc. - Cyberspace Architects=81    617/721-6100
nazgul@utopia.com                               http://www.utopia.com/

I'm not sure which upsets me more: that people are so unwilling to accept
responsibility for their own actions, or that they are so eager to regulate
everyone else's.


eturn-Path: nazgul@utopia.com 
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB00615; Tue, 9 May 1995 16:37:21 +0500
Received: from utopia.com (utopia.wing.net) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA21770; Tue, 9 May 1995 16:35:04 +0500
Received: from [204.57.39.6] by utopia.com (8.6.12/SONY-4.0MX)
	id QAA11536; Tue, 9 May 1995 16:32:56 -0400
Return-Path: <nazgul@utopia.com>
X-Sender: nazgul@mailhost.utopia.com
Message-Id: <v02110114abd580d2c6d2@[204.57.39.6]>
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Date: Tue, 9 May 1995 16:34:50 -0400
To: Jared_Rhine@hmc.edu, Multiple recipients of list <www-talk@www10.w3.org>
From: nazgul@utopia.com (Kee Hinckley)
Subject: Re: Content-lengths of dynamic objects
Content-Length: 896

At 1:14 PM 5/8/95, Jared Rhine wrote:
>methods: 'compute_lastmod' and 'render'.  When I want to render an object, =
I
>first instantiate the object, and invoke 'compute_lastmod' on it.  I compar=
e
>the return value with the timestamp of an on-disk copy of the last
>rendering.  If they are the same, I simply serve the file off disk; if not,


This has the disadvantage of leaving the user to wait until your render is
done. Otherwise for large options you could be presenting the object while
rendering, thus decreasing the overall display time and increasing user
satisfaction.

Kee Hinckley      Utopia Inc. - Cyberspace Architects=81    617/721-6100
nazgul@utopia.com                               http://www.utopia.com/

I'm not sure which upsets me more: that people are so unwilling to accept
responsibility for their own actions, or that they are so eager to regulate
everyone else's.


eturn-Path: dceccald@elaine.crcg.edu 
Return-Path: <dceccald@elaine.crcg.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA08051; Tue, 9 May 1995 17:42:28 +0500
Received: from elaine.crcg.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA22536; Tue, 9 May 1995 17:40:15 +0500
Received: from currawong.crcg.edu by elaine.crcg.edu (4.1/SMI-4.1)
	id AA18107; Tue, 9 May 95 17:40:52 EST
Received: by currawong.crcg.edu (5.x/SMI-SVR4)
	id AA02685; Tue, 9 May 1995 18:42:05 -0400
Date: Tue, 9 May 1995 18:42:05 -0400
From: dceccald@elaine.crcg.edu (Danyel Ceccaldi)
Message-Id: <9505092242.AA02685@currawong.crcg.edu>
To: www-talk@w3.org
Subject: Re: Modular Browsers (was:User authentication)
X-Sun-Charset: US-ASCII
Content-Length: 676


> From: "Adam T. McClure" <mcclurea@nag.cs.colorado.edu>
> Subject: Re: Modular Browsers (was:User authentication)
> Yeah, how bout an OpenDoc browser Netscape?
> Is that a no-brainer or a novel concept?

Perhaps there is more.
What about an 'realistic' platform independant interface.
Reality (for the moment) is:

OS/2    - OpenDoc
MacOS   - OpenDoc
Windows - OLE 2.0
UNIX    - i don't remember a name, but there also a OpenDoc-like interface
...

A few issues for such a modular Browser:

- (Inter)action on client side
- Style sheets
- Layout control
- Proxy-control (resolving host-moving-problems)
- Downloading (and auto-installing) of small patches
...


By
  Danyel
eturn-Path: jared@math.hmc.edu 
Return-Path: <jared@math.hmc.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA24246; Tue, 9 May 1995 20:06:44 +0500
Received: from aslan.math.hmc.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA23684; Tue, 9 May 1995 20:04:24 +0500
Received: (from jared@localhost) by aslan.math.hmc.edu (8.6.8.1/8.6.7) id RAA17419; Tue, 9 May 1995 17:04:04 -0700
Date: Tue, 9 May 1995 17:04:04 -0700
Message-Id: <199505100004.RAA17419@aslan.math.hmc.edu>
From: Jared Rhine <Jared_Rhine@hmc.edu>
To: nazgul@utopia.com
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Re: Content-lengths of dynamic objects
References: <v02110114abd580d2c6d2@[204.57.39.6]>
X-Attribution: JRhine
X-Uri: <URL:http://www.hmc.edu/~jared/home>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Length: 2353

[Citation date: Tue, 9 May 1995 17:11:40 +0500]
KH == Kee Hinckley <nazgul@utopia.com>

  JRhine> When I want to render an object, I first instantiate the object,
  JRhine> and invoke 'compute_lastmod' on it.  I compare the return value
  JRhine> with the timestamp of an on-disk copy of the last rendering.  If
  JRhine> they are the same, I simply serve the file off disk...

  KH> This has the disadvantage of leaving the user to wait until your
  KH> render is done. Otherwise for large options you could be presenting
  KH> the object while rendering, thus decreasing the overall display time
  KH> and increasing user satisfaction.

True, but I'm not sure what your objection is.  In the vast majority of
cases, object rendering time is inconsequential compared to network lag and
client rendering time.  Being able to incrementally render client side isn't
that big of a win; the time difference between streaming to the client as
the object is rendering and waiting until it is done before sending is no
more than a few tenths of a second at the greatest and does not
substantially affect perceived performance.  In any case, this model is
unavoidable because in many cases, dynamic objects do not follow a linear
rendering flow; that is to say, I can't properly render the top until I've
processed all the data for the object.

There is a possible exception for heavily loaded servers that take on the
order of seconds to render objects; in this case, there would be a
noticeable difference is response time.  In that case, since you're working
with dynamic objects, just distribute the buggers to reduce server load.
Clean, simple, and cheap.

Hmmm, I just considered objects that inherently take a long time to render
regardless of server load (databases and such).  My objects aren't like that
(yet), so I haven't had to deal with it.  In that case, you would indeed
want some mechanism to present some feedback to the user early; in this
case, we will obviously need some kind of boundary or packetized data
scheme; no getting around that, I suspect, if we want accurate entity-body
delimiting.

-- 
Jared_Rhine@hmc.edu / HMC / <URL:http://www.hmc.edu/~jared/home>

"I regret to say that we of the F.B.I. are powerless to act in cases of
 oral-genital intimacy, unless it has in some way obstructed interstate
 commerce." -- J Edgar Hoover
eturn-Path: ren@dstc.edu.au 
Return-Path: <ren@dstc.edu.au>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA00552; Tue, 9 May 1995 21:26:10 +0500
Received: from trapdoor.dstc.edu.au by www10.w3.org (5.0/NSCS-1.0S) 
	id AA24430; Tue, 9 May 1995 21:26:06 +0500
Received: from foxtail.dstc.edu.au (foxtail.dstc.edu.au [130.102.176.38]) by trapdoor.dstc.edu.au (8.6.10/8.6.9) with ESMTP id LAA26210; Wed, 10 May 1995 11:26:01 +1000
Received: (from ren@localhost) by foxtail.dstc.edu.au (8.6.10/8.6.10) id LAA13023; Wed, 10 May 1995 11:26:00 +1000
Date: Wed, 10 May 1995 11:25:59 +1000 (EST)
From: Renato Iannella <ren@dstc.edu.au>
To: Brian Behlendorf <brian@organic.com>
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Re: How bout types of colored links?
In-Reply-To: <Pine.3.89.9505090812.f24326-0100000@eat.organic.com>
Message-Id: <Pine.OSF.3.91.950510112217.28764B-100000@foxtail.dstc.edu.au>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 954

On Tue, 9 May 1995, Adam T. McClure wrote:
> There's currently a method in
> Netscape 1.1 to define different colors for visited and new links,
> as well as redefining the standard text color, but there's no way
> to have your own labelled links.  

It is not a good idea to play with link colours.
Link colours should be a user option.
For example, I have set my visited links to Blue
and new links to Red - If you swap these around
with html code you will create some problems.

Use a small graphic icon to depict differnet paths.


Cheers... Renato
______________________________________________________________________
Dr Renato Iannella                      Research Data Network CRC
Distributed Systems Technology Centre   TEL:     +61 7 365 4310
The University of Queensland            FAX:     +61 7 365 4311
QLD, 4072, AUSTRALIA                    EMAIL:   ren@dstc.edu.au
________________________________________WWW: <http://www.dstc.edu.au/>

eturn-Path: bob@mano.soest.hawaii.edu 
Return-Path: <bob@mano.soest.hawaii.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA00756; Tue, 9 May 1995 21:28:57 +0500
Received: from mano.soest.hawaii.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA24448; Tue, 9 May 1995 21:28:50 +0500
Received: (from bob@localhost) by mano.soest.hawaii.edu (8.6.12/8.6.6) id PAA26401; Tue, 9 May 1995 15:27:31 -1000
Date: Tue, 9 May 1995 15:27:31 -1000
From: Bob Cunningham <bob@mano.soest.hawaii.edu>
Message-Id: <199505100127.PAA26401@mano.soest.hawaii.edu>
To: dnew@sgf.fv.com, www-talk@www10.w3.org
Subject: Re: common log format...unresolved hosts?
Content-Length: 362

>I was guessing these were people with dial-up slip 
>accounts with dynamic IP assignments.

Well the local ISPs in my area dynamically assign IP numbers,
but nonetheless register their entires series on DNS.  But
probably lots of other ISPs don't.

Mainly, though, I suspect that there's simply a lot of reverse
DNS table entries missing all over the world...

eturn-Path: cnw3@di.uminho.pt 
Return-Path: <cnw3@di.uminho.pt>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA01591; Tue, 9 May 1995 21:38:16 +0500
Received: from dxmint.cern.ch by www10.w3.org (5.0/NSCS-1.0S) 
	id AA24483; Tue, 9 May 1995 21:37:52 +0500
Received: from wep.uminho.pt by dxmint.cern.ch
	id AA14574; Wed, 10 May 1995 03:37:35 +0200
Received: from s700.di.uminho.pt by wep.uminho.pt with SMTP (PP) 
          id <15204-0@wep.uminho.pt>; Wed, 10 May 1995 03:34:39 +0200
Received: by di.uminho.pt (4.1/SMI-4.1) id AA18639;
          Wed, 10 May 95 03:34:37 +0200
Date: Wed, 10 May 95 03:34:37 +0200
From: cnw3@di.uminho.pt (Conferencia Nacional WWW)
Message-Id: <9505100134.AA18639@di.uminho.pt>
To: www-talk@w3.org
Subject: [WWW National Conference] Registration Form and Accomodations
Content-Length: 1095


-------------------------------------------------------------------------------

                            WWW National Conference
                        Internet Multimedia Information

                               July 6-8, 1995
			  Departamento de Informatica
                             Universidade do Minho
                               Braga, Portugal 

-------------------------------------------------------------------------------


 You can find the Registration Form (text, PostScript and html on-line
"form" format) and accomodation information (html format) under the
following URL:

	   http://www.di.uminho.pt/IMI/RF/index.eng.html

For more informations please contact:

     CNW3
     Departamento de Informatica
     Universidade do Minho, Campus de Gualtar
     4710 Braga

     URL: http://www.di.uminho.pt/cnw3.html
     Phone: +351 (53) 604470
     Fax: +351 (53) 612954
     email: cnw3@di.uminho.pt



Best regards,

  The Local Organization Committee,
      
      cnw3@di.uminho.pt


P.S.: Payment must be received by May 19 to receive "early-bird" prices.


eturn-Path: jtilton%willamette.edu@www10.w3.org 
Return-Path: x-face
Received: from mintaka.lcs.mit.edu by www19 (5.0/NSCS-1.0S) 
	id AJ02394; Tue, 9 May 1995 21:43:25 +0500
Received: from www10.w3.org by MINTAKA.LCS.MIT.EDU id aa12792;
          9 May 95 20:54 EDT
Received: from earth.willamette.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA24242; Tue, 9 May 1995 20:50:23 +0500
Received: from jupiter (jtilton@jupiter.willamette.edu [158.104.1.1]) 
	by earth.willamette.edu (8.6.12/8.6.4) with SMTP 
	id RAA26714 for <www-talk@w3.org>; Tue, 9 May 1995 17:50:17 -0700
Date: Tue, 9 May 1995 17:50:17 -0700 (PDT)
From: "James (Eric) Tilton" <jtilton@willamette.edu>
To: www-talk@w3.org
Subject: Versions of HTML spec
Message-Id: <Pine.SUN.3.91.950509174922.27867B-100000@jupiter>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 1509

Hi -- we're trying to describe, in a few paragraphs, the development and 
significance of the various HTML specifications. I'd like to get comments 
on these paragraphs to see whether they are, in fact, an accurate summary:

"The first release of HTML, HTML Level 0, is the original HTML 
specification. Although the base-level HTML reference, some of the tags 
present in Level 0 have since become depreciated.

"HTML Level 1 includes the Level 0 tags, with some additional features, 
such as images, primarily introduced by the Mosaic development team.

"With the introduction of it Netscape Navigator Web browser, Netscape 
introduced many tags which were particular to its browser, such as CENTER 
and BLINK. Netscape has since joined the ranks of the 
standards committees and have labeled the tags it introduces not within 
the official HTML specification as "experimental".

"HTML Level 2 contained some revisions to HTML to make it an 
SGML-compliant language, and added support for forms.

"HTML Level 3 includes work to include support for non-Western European 
writing systems, table, figure, and scientific notation support, 
client-side image maps and other interactive applications, and style sheets."

My questions are -- is HTML Level 2 synonymous with HTML 2.0, and so on? 
Or are the conformance levels different from the version numbers, and if 
so, how? And, in general, is the above accurate?

Any comments greatly appreciated. Thanks!

						-et

 --------End of Unsent Message--------

eturn-Path: nazgul@utopia.com 
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA06727; Tue, 9 May 1995 22:19:59 +0500
Received: from utopia.com (utopia.wing.net) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA24776; Tue, 9 May 1995 22:19:46 +0500
Received: from [204.57.39.6] by utopia.com (8.6.12/SONY-4.0MX)
	id WAA12004; Tue, 9 May 1995 22:17:54 -0400
Return-Path: <nazgul@utopia.com>
X-Sender: nazgul@mailhost.utopia.com
Message-Id: <v02110116abd5d15ab28c@[204.57.39.6]>
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Date: Tue, 9 May 1995 22:19:42 -0400
To: Jared_Rhine@hmc.edu, Multiple recipients of list <www-talk@www10.w3.org>
From: nazgul@utopia.com (Kee Hinckley)
Subject: Re: Content-lengths of dynamic objects
Content-Length: 1251

At 8:09 PM 5/9/95, Jared Rhine wrote:
>  KH> This has the disadvantage of leaving the user to wait until your
>  KH> render is done. Otherwise for large options you could be presenting

>Hmmm, I just considered objects that inherently take a long time to render
>regardless of server load (databases and such).  My objects aren't like tha=
t
>(yet), so I haven't had to deal with it.  In that case, you would indeed
>want some mechanism to present some feedback to the user early; in this
>case, we will obviously need some kind of boundary or packetized data
>scheme; no getting around that, I suspect, if we want accurate entity-body
>delimiting.

I think a packet oriented approach would help. I know at one time there was
talk of doing inline objects that way and interspersing the packets so that
you could do what Netscape does without opening N-connections to the
server. Obviously it seriously complicates the protocols though.

Kee Hinckley      Utopia Inc. - Cyberspace Architects=81    617/721-6100
nazgul@utopia.com                               http://www.utopia.com/

I'm not sure which upsets me more: that people are so unwilling to accept
responsibility for their own actions, or that they are so eager to regulate
everyone else's.


eturn-Path: jared@math.hmc.edu 
Return-Path: <jared@math.hmc.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB02183; Wed, 10 May 1995 02:49:52 +0500
Received: from aslan.math.hmc.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA26958; Wed, 10 May 1995 02:49:43 +0500
Received: (from jared@localhost) by aslan.math.hmc.edu (8.6.8.1/8.6.7) id XAA20170; Tue, 9 May 1995 23:49:32 -0700
Date: Tue, 9 May 1995 23:49:32 -0700
Message-Id: <199505100649.XAA20170@aslan.math.hmc.edu>
From: Jared Rhine <Jared_Rhine@hmc.edu>
To: nazgul@utopia.com
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Re: Content-lengths of dynamic objects
References: <v02110116abd5d15ab28c@[204.57.39.6]>
X-Attribution: JRhine
X-Uri: <URL:http://www.hmc.edu/~jared/home>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Length: 1456

[Citation date: Tue, 9 May 1995 22:58:03 +0500]
KH == Kee Hinckley <nazgul@utopia.com>

  KH> This has the disadvantage of leaving the user to wait until your
  KH> render is done.

  JRhine> ...we will obviously need some kind of boundary or packetized data
  JRhine> scheme; no getting around that, I suspect, if we want accurate
  JRhine> entity-body delimiting.

  KH> I think a packet oriented approach would help. I know at one time
  KH> there was talk of doing inline objects that way and interspersing the
  KH> packets so that you could do what Netscape does without opening
  KH> N-connections to the server. Obviously it seriously complicates the
  KH> protocols though.

That is a different issue and definitely best left to HTTP NG.  The original
issue was how to guarantee the client that they have received the full
entity-body, specifically how to do this for dynamic objects.  I suggested
that one way to handle this was to put the entity-body on disk and then use
content-length.  If you don't want to do that, I believe there are two
primary proposals: unique boundaries and packet encoding.  I don't think
packet encoding was ever discussed in the context of multiplexing a
connection, as you discuss.

-- 
Jared_Rhine@hmc.edu / HMC / <URL:http://www.hmc.edu/~jared/home>

"To hear many religious people talk, one would think God created the
 torso, head, legs and arms, but the devil slapped on the genitals."
        -- Don Schrader
eturn-Path: jared@math.hmc.edu 
Return-Path: <jared@math.hmc.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA03541; Wed, 10 May 1995 02:57:07 +0500
Received: from aslan.math.hmc.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA27059; Wed, 10 May 1995 02:57:01 +0500
Received: (from jared@localhost) by aslan.math.hmc.edu (8.6.8.1/8.6.7) id XAA20191; Tue, 9 May 1995 23:56:54 -0700
Date: Tue, 9 May 1995 23:56:54 -0700
Message-Id: <199505100656.XAA20191@aslan.math.hmc.edu>
From: Jared Rhine <Jared_Rhine@hmc.edu>
To: Rainer Klute <klute@nads.de>
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Expires non-compliance
References: <199505090730.AAA09553@aslan.math.hmc.edu>
  <199505090830.KAA11662@heike.nads.de>
X-Attribution: JRhine
X-Uri: <URL:http://www.hmc.edu/~jared/home>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Length: 1079

[Citation date: Tue, 09 May 1995 10:30:57 +0200]
RK == Rainer Klute <klute@nads.de>

  JRhine> Unfortunately, my tests indicate that Netscape doesn't handle the
  JRhine> Expires header properly :(

  RK> Not to defend Netscape against anything, but is there a client at all
  RK> that handles Expires properly? Not even the CERN proxy does.

I believe Emacs-w3 does, but I'd have to run some more exhaustive tests,
since when I found out Netscape's behavior, I had to resort to other means
besides Expires.  I would note that section 7.1.8 of
draft-ietf-http-v10-spec-00 states:

   Caching clients (including proxies) must not cache this copy of the
   resource beyond the date given, unless its status has been updated by a
   later check of the origin server.

The key word is "must"; caching clients that do not handle the Expires
header are non-compliant HTTP applications (well, granted that the standard
is only a draft, but it's all we've got).

-- 
Jared_Rhine@hmc.edu / HMC / <URL:http://www.hmc.edu/~jared/home>

"One cannot mark the point without marking the path."
eturn-Path: jared@math.hmc.edu 
Return-Path: <jared@math.hmc.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA05070; Wed, 10 May 1995 03:12:43 +0500
Received: from aslan.math.hmc.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA27342; Wed, 10 May 1995 03:12:36 +0500
Received: (from jared@localhost) by aslan.math.hmc.edu (8.6.8.1/8.6.7) id AAA20303; Wed, 10 May 1995 00:12:32 -0700
Date: Wed, 10 May 1995 00:12:32 -0700
Message-Id: <199505100712.AAA20303@aslan.math.hmc.edu>
From: Jared Rhine <Jared_Rhine@hmc.edu>
To: pitkow@cc.gatech.edu
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Re: Session tracking
References: <199504212253.SAA18137@hapeville.cc.gatech.edu>
X-Attribution: JRhine
X-Uri: <URL:http://www.hmc.edu/~jared/home>
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Length: 950

[Citation date: Fri, 21 Apr 1995 19:31:52 +0500]
JP == James Pitkow <pitkow@cc.gatech.edu>

[The topic was session-tracking proposals, esp the proposed "Session-ID"
header.]

  JP> 1) identify sessions (esp. from firewalled domains)
  JP> [...]
  JP> If the 'From:' field is widely used, you'd have what you wanted(1).

No, I don't believe so, because one of the stated requirements is that we be
able to identify sessions without compromising the identity of the user.

If this is the case, I fail to see how current HTTP practices allow for
session tracking as you claim:

  JP> My point is that the protocol does not need to be changed to handle
  JP> session(1) &/or user identification(2) - WWW browsers/interfaces need
  JP> to pass the right information.

Elaborations?

-- 
Jared_Rhine@hmc.edu / HMC / <URL:http://www.hmc.edu/~jared/home>

"Remember: Once you pull the pin, Mr. Grenade is no longer your friend."
        -- Zippy the Pinhead
eturn-Path: njw@sarc.city.ac.uk 
Return-Path: <njw@sarc.city.ac.uk>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA17289; Wed, 10 May 1995 04:44:37 +0500
Received: from cripplecock.sarc.city.ac.uk by www10.w3.org (5.0/NSCS-1.0S) 
	id AA28234; Wed, 10 May 1995 04:44:30 +0500
Received: (from daemon@localhost) by cripplecock.sarc.city.ac.uk (8.6.9/8.6.9) id JAA18157 for www-talk@www10.w3.org; Wed, 10 May 1995 09:44:00 +0100
Received: from Messages.8.5.N.CUILIB.3.45.SNAP.NOT.LINKED.cripplecock.sarc.city.ac.uk.i386.bsd
          via MS.5.6.cripplecock.sarc.city.ac.uk.i386_bsd;
          Wed, 10 May 1995 09:44:00 +0100 (WET)
Message-Id: <8jg7nEcQ0k00J62Vst@sarc.city.ac.uk>
Date: Wed, 10 May 1995 09:44:00 +0100 (WET)
From: Nick Williams <njw@sarc.city.ac.uk>
To: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Re: How bout types of colored links?
In-Reply-To: <199505090713.BAA20417@nag.cs.colorado.edu>
References: <199505090713.BAA20417@nag.cs.colorado.edu>
Content-Length: 987

Excerpts from www.talk: 9-May-95 How bout types of colored l.. "Adam T.
McClure"@nag.cs (1382)

> It seemed logical to me to define a named type of link that has a particular
> color associated with it.  All the references to definitions could be
> in blue, the links to other elements of the discussion in red, and 
> purple would jump off my server to an off-site page. 

Surely the nicest way to do this would be to enhance the stylesheet spec
(if it doesn't provide for it already) so that links of specific
relation types can be modified seperately to the general class of links.
 E.g. all links should look blue, but links.Annotation should be
paleblue, blah, blah.

You might even be able to go so far as to say, I'm only interested in
links of class "Subdocument", so only highlight those and make all
others non-lighted (as 'twere).

Web: http://web.cs.city.ac.uk/finger/njw
E-mail: njw@sarc.city.ac.uk (MIME and ATK)
Work Telephone: +44 171 477 8551
Work Fax: +44 171 477 8587
eturn-Path: martin@mrrl.lut.ac.uk 
Return-Path: <martin@mrrl.lut.ac.uk>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA01768; Wed, 10 May 1995 09:46:01 +0500
Received: from gizmo.lut.ac.uk by www10.w3.org (5.0/NSCS-1.0S) 
	id AA01083; Wed, 10 May 1995 09:43:27 +0500
Received: from localhost (martin@localhost) by gizmo.lut.ac.uk (8.6.12/8.6.9) with SMTP id OAA07307 for <www-talk@w3.org>; Wed, 10 May 1995 14:43:02 +0100
Message-Id: <199505101343.OAA07307@gizmo.lut.ac.uk>
X-Mailer: exmh version 1.6 4/21/95
To: www-talk@w3.org
Subject: Re: User authentication 
X-Uri: <URL:http://www.mrrl.lut.ac.uk/~martin>
In-Reply-To: Your message of "Mon, 08 May 1995 20:54:32 +0500."
             <9505082210.AA08541@www10.w3.org> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Wed, 10 May 1995 14:42:52 +0100
From: Martin Hamilton <martin@mrrl.lut.ac.uk>
Content-Length: 1171


Rick Troth writes:

| I don't see any way to do  "real authentication"  without
| using public key electronic signatures,  and I question whether or not
| we need something that strong to eliminate news and mail forgery.

I suspect SMTP and NNTP are still an authentication free zone for the vast majority of Internet users.  Best shot is to have a browser that can be configured to refuse to send messages unless _it_ is happy with your credentials ?

So, the authentication mechanism just has to be strong enough that you can satisfy the browser, rather than the whole world.  In which case, all we really need is some variation on

  o domain name/port number to contact
  o user name
  o "password"

There are lots of simple authentication scenarios which fit into this model, e.g. POP, IMAP, and even FTP.  No export restrictions here!

Servers are trivial to implement, and (hey!) you might have one already ? :-)

Client support should be trivial to add to the browser - after all, you already have primitive operations to open and close TCP connections & read and write data

QED?

Martin

PS For something more sophisticated, see the APOP command in RFC 1725


eturn-Path: derek@pipex.net 
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AD12472; Wed, 10 May 1995 10:57:57 +0500
Received: from mailhost.pipex.net by www10.w3.org (5.0/NSCS-1.0S) 
	id AA01911; Wed, 10 May 1995 10:55:28 +0500
Received: from khazi.pipex.net by pipe.pipex.net with SMTP (PP);
          Wed, 10 May 1995 15:55:07 +0100
Received: by khazi.pipex.net (8.6.9/Pipex Simple MT experimental) id PAA00769;
          Wed, 10 May 1995 15:55:04 +0100
Return-Path: <derek@pipex.net>
From: derek <derek@pipex.net>
Date: Wed, 10 May 1995 15:55:04 +0100
Message-Id: <199505101455.PAA00769@khazi.pipex.net>
To: www-talk@www10.w3.org
Subject: libwwwperl
Content-Length: 619

Does anyone have experience using libwww-perl 0.40 on Solaris 2.3 with Perl 5?

We are having some problems with it and I could do with any insights anyone may
have.

TIA

Derek
---
Derek Harding                                                   derek@pipex.net
Webmaster and Product & Software Developer                  +44 (0) 1223 250422
UnipalmPIPEX                                 http://www.pipex.net/people/derek/
-------------------------------------------------------------------------------
He was currently wondering vaguely who Moey and Chandon were.
        -- (Terry Pratchett & Neil Gaiman, Good Omens)
eturn-Path: drtr1@cus.cam.ac.uk 
Return-Path: <drtr1@cus.cam.ac.uk>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB23960; Wed, 10 May 1995 12:39:29 +0500
Received: from bootes.cus.cam.ac.uk by www10.w3.org (5.0/NSCS-1.0S) 
	id AA03209; Wed, 10 May 1995 12:37:01 +0500
Received: from grus.cus.cam.ac.uk [131.111.8.3] (ident = root) 
	by bootes.cus.cam.ac.uk with smtp 
	(Smail-3.1.29.0 #36) id m0s9EiO-000C44C; Wed, 10 May 95 17:34 BST
Received: by grus.cus.cam.ac.uk 
	(Smail-3.1.29.0 #36) id m0s9EiI-0007aFC; Wed, 10 May 95 17:34 BST
Message-Id: <m0s9EiI-0007aFC@grus.cus.cam.ac.uk>
Date: Wed, 10 May 95 17:34 BST
Sender: drtr1@cus.cam.ac.uk (David Robinson)
To: frystyk@w3.org
Subject: Re:  Strange problem with GMT on Solaris
Cc: www-talk@www10.w3.org
From: drtr1@cam.ac.uk (David Robinson)
Content-Length: 733

>I have a strange problem on Solaris:
Which version?

>The gmtime() function on Solaris seems to return the wrong value. It is one
>hour ahead of GMT which means that all date header values in a HTTP request
>or response is one hour ahead of GMT. This is for example the case on our
>www.w3.org servers which certainly confuses a lot of caches!
>
>It is simple to test:
>
>        time_t calendar = time(NULL);
>        struct tm *gmt = gmtime(&calendar);
>        strftime(buf, 40, "%a, %d %b %Y %H:%M:%S GMT", gmt);
>
>Has anybody else heard of this problem?

I have seen a similar problem under Solaris 2.3; gmtime would include the
hour for daylight saving time.

That problem does not exist under Solaris 2.4.

 David Robinson.
eturn-Path: @mutual.co.uk:HEPWORTH@mutual.co.uk 
Return-Path: <@mutual.co.uk:HEPWORTH@mutual.co.uk>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB07899; Wed, 10 May 1995 14:41:58 +0500
Received: from eros.britain.eu.net by www10.w3.org (5.0/NSCS-1.0S) 
	id AA04536; Wed, 10 May 1995 14:29:45 +0500
Received: from mutual.co.uk by eros.britain.eu.net with UUCP 
          id <sg.28651-1@eros.britain.eu.net>; Wed, 10 May 1995 19:29:19 +0100
Original-Received: from 
                   cc:Mail by mutual.co.uk id AA800146317 Wed, 10 May 95 
                   15:51:57
Pp-Warning: Illegal Received field on preceding line
Date: Wed, 10 May 95 15:51:57
From: HEPWORTH@mutual.co.uk
Message-Id: <9504108001.AA800146317@mutual.co.uk>
To: www-talk@www10.w3.org
Content-Length: 0



     I have a problem and if anybody could help I would be most pleased. If 
     this is not the correct list for this kind of help, please forgive my 
     ignorance.
     
     Appended arguments to a URL (GET method) do not appear to be passed to 
     the cgi-test.pl script that ships with the cgi-lib.zip distribution 
     package for WWW/perl/HTML. (httpd & Mosaic for Linux, perl4)
     
     To recreate :
     
     1) Generate some HTML from a gateway script. 2) Include a URL that 
     looks like
     
     <A HREF="HTTP://hostname/cgi-lib/test-cgi.pl?123+myfile.tmpl">click 
     here</A>.
     3) Click on the URL and wait for a return from the cgi-test script. 
     The script returns a blank page - the arguments are being passed but 
     are not being displayed (no prompt for an argument appears). 
     
     It was my impression that this script would be the best way to test 
     that the URLs I am generating in my gateway script are valid. Any 
     arguments I pass to them by appending should be listed in the returned 
     page. This is either not the case or something else is happening 
     between the click on the URL and the returned cgi-test page.
     
     I am certain that my directory aliases and rights etc are correctly 
     set up for my standalone httpd server. The access_log shows that the 
     script is being reached and dummy text HTML inserted in various parts 
     of the cgi-test script show that it is being axecuted and that it 
     recognises that arguments have been appended. A version problem 
     perhaps ?
     
     
     Thanks Ultan
     ocarroll@mutual.co.uk
eturn-Path: pitkow@cc.gatech.edu 
Return-Path: <pitkow@cc.gatech.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AC08634; Wed, 10 May 1995 14:48:27 +0500
Received: from burdell.cc.gatech.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA04715; Wed, 10 May 1995 14:48:23 +0500
Received: from hapeville.cc.gatech.edu (pitkow@hapeville.cc.gatech.edu [130.207.119.215]) by burdell.cc.gatech.edu (8.6.12/8.6.9) with ESMTP id OAA24907 for <www-talk@www10.w3.org>; Wed, 10 May 1995 14:48:21 -0400
Received: (from pitkow@localhost) by hapeville.cc.gatech.edu (8.6.10/8.6.9) id OAA15272; Wed, 10 May 1995 14:47:41 -0400
From: pitkow@cc.gatech.edu (James Pitkow)
Message-Id: <199505101847.OAA15272@hapeville.cc.gatech.edu>
Subject: Re: Session tracking
To: Multiple@cc.gatech.edu, recipients@cc.gatech.edu, of@cc.gatech.edu,
        list@cc.gatech.edu, <www-talk@www10.w3.org>
Date: Wed, 10 May 1995 14:47:40 -0400 (EDT)
In-Reply-To: <199505100712.AAA20303@aslan.math.hmc.edu> from "Jared Rhine" at May 10, 95 00:12:32 am
X-Mailer: ELM [version 2.4 PL23]
Content-Type: text
Content-Length: 3758      


Jared Rhine writes on Wed, 10 May 1995 00:12:32 -0700:

> [The topic was session-tracking proposals, esp the proposed "Session-ID"
> header.]
>
>  JP> 1) identify sessions (esp. from firewalled domains)
>  JP> [...]
>  JP> If the 'From:' field is widely used, you'd have what you wanted(1).
>
> No, I don't believe so, because one of the stated requirements is that we be
> able to identify sessions without compromising the identity of the user.
> 
> If this is the case, I fail to see how current HTTP practices allow for
> session tracking as you claim:

>  JP> My point is that the protocol does not need to be changed to handle
>  JP> session(1) &/or user identification(2) - WWW browsers/interfaces need
>  JP> to pass the right information.
>
> Elaborations?

  Sure:

  Inspection of the http specifications indicates the From: field be in 
Internet Mail Format and is supposed to give the name of the requesting user.  
Additionally, it recommends, but not require, that the address be valid.
Provided that the loci of control over contents of the From: field rests
on the client with the user, one can easily see where this field can be
used to support Chom's notion of multiple aliases and provide session tracking. 

  For example, a user upon start up of their WWW Browser for the first time
is asked to configure the From: field as part of the preferences which are 
currently configurable.  The user has the option of providing a real/valid 
address, or a bogus identification number.  This information accompanies 
requests and can be used by servers to customize information and track users 
across sessions.   Multi-user platforms that are not access controlled (e.g. 
Macs and PCs) can still be supported by providing browser level support for 
multiple aliases across multiple users.

  Now let's say that the user is not happy with the customization of the 
information/services provided by a server - it is the user who can reset 
the profiling and not be subjected to rigid server-side user models.  
Another scenario would be where the client keeps a table of aliases used for 
interactions with servers and allows the user to switch roles/aliases.  Here, 
the client could be used to generate unique client side ids.  And 
broadcasting of real identification is managed by the user.

  The downside with both being you may get name/id space collisions from 
clients behind the same gateway/firewall.  Following the 80/20 rule, I'd
argue that the majority of traffic will be correctly handled and that 
session id tracking is not a situation where 100/0 is required.  Where 
better assurance is needed to ensure the same user across each and every 
session, other mechanisms within http can be used.

  Of course there is the issue of session id management. Generous estimates 
(20 million users with 60,000 WWW servers) indicate that 1.2 x 10^12 possible
ids would need to be maintained by either the client or the server if every 
relation requires a different id.  In practice though, a user may have say 
10 aliases or so. Therefore each client would need to manage 600,000 ids for 
all WWW servers whereas each WWW server would need to maintain 1.3 x 10^13 ids.
Note that if you argue that each client will not interact with all servers,
this constant applies to both equations and therefore cancels, so the difference
in magnitudes remains the same.  An assumption being made here is that there 
will be always be more users than servers, which seems reasonable. So it seems 
from this approach that the latter solution (server-side) generation of 
sessions ids does not scale as well as client-side id/alias management, 
regardless of the loci of control approach argued above.

I'm away, so take your time in replying.

Jim.
eturn-Path: pitkow@cc.gatech.edu 
Return-Path: <pitkow@cc.gatech.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB11850; Wed, 10 May 1995 15:14:48 +0500
Received: from burdell.cc.gatech.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA05000; Wed, 10 May 1995 15:14:40 +0500
Received: from hapeville.cc.gatech.edu (pitkow@hapeville.cc.gatech.edu [130.207.119.215]) by burdell.cc.gatech.edu (8.6.12/8.6.9) with ESMTP id PAA26471 for <www-talk@www10.w3.org>; Wed, 10 May 1995 15:14:39 -0400
Received: (from pitkow@localhost) by hapeville.cc.gatech.edu (8.6.10/8.6.9) id PAA15465 for www-talk@www10.w3.org; Wed, 10 May 1995 15:14:24 -0400
From: pitkow@cc.gatech.edu (James Pitkow)
Message-Id: <199505101914.PAA15465@hapeville.cc.gatech.edu>
Subject: Typo
To: www-talk@www10.w3.org
Date: Wed, 10 May 1995 15:14:24 -0400 (EDT)
X-Mailer: ELM [version 2.4 PL23]
Content-Type: text
Content-Length: 305       


The perils of cut and paste - my last message stated incorrectly:

>10 aliases or so. Therefore each client would need to manage 600,000 ids for 
>all WWW servers whereas each WWW server would need to maintain 1.3 x 10^13 ids.

where the last number should be 200,000 million not 1.3 x 10^13 ids.

Jim.

eturn-Path: john@math.nwu.edu 
Return-Path: <john@math.nwu.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA28314; Wed, 10 May 1995 17:48:50 +0500
Received: from dxmint.cern.ch by www10.w3.org (5.0/NSCS-1.0S) 
	id AA06550; Wed, 10 May 1995 17:48:47 +0500
Received: from hopf.math.nwu.edu by dxmint.cern.ch
	id AA08407; Wed, 10 May 1995 23:48:27 +0200
Received: (from john@localhost) by hopf.math.nwu.edu (8.6.10/8.6.9) id QAA23233; Wed, 10 May 1995 16:49:45 -0500
Date: Wed, 10 May 1995 16:49:45 -0500
From: John Franks <john@math.nwu.edu>
Message-Id: <199505102149.QAA23233@hopf.math.nwu.edu>
To: www-talk@w3.org
Reply-To: john@math.nwu.edu
Subject: Re: Session tracking
Content-Length: 2293


In article <199505101847.OAA15272@hapeville.cc.gatech.edu>, you write:
> 
> Jared Rhine writes on Wed, 10 May 1995 00:12:32 -0700:
> 
> > [The topic was session-tracking proposals, esp the proposed "Session-ID"
> > header.]
> >
> >  JP> 1) identify sessions (esp. from firewalled domains)
> >  JP> [...]
> >  JP> If the 'From:' field is widely used, you'd have what you wanted(1).
> >
> > No, I don't believe so, because one of the stated requirements is that we be
> > able to identify sessions without compromising the identity of the user.
> > 
> > If this is the case, I fail to see how current HTTP practices allow for
> > session tracking as you claim:
> 
> >  JP> My point is that the protocol does not need to be changed to handle
> >  JP> session(1) &/or user identification(2) - WWW browsers/interfaces need
> >  JP> to pass the right information.
> >


There is a very good session-id/session-tracking/shopping basket proposal
from Netscape.  And the best thing about it is it already implemented in
recent Netscape browsers.  I must say it is great to have an HTTP proposal
that is actually implemented rather than just talked about.  Thanks Netscape.

Basically the server uses a header like

Set-Cookie: name=opaque; expires=date; path=xxx; domain=xxx

The client caches this until the end of session if there is no "epires"
field and until the expiration date otherwise (across sessions).  The
client returns one or more 

Cookie: name=opaque

headers when it makes a request in the domain/path realm.  This works
very well for lots of things.  In particular the MCI shopping area
uses this for their shopping baskets.  Details of this proposal were
posted a while back in www-talk.  Until then I had been wondering how
the shopping baskets worked.

The posting by Lou Montulli proposed that this be added to HTTP 1.1
and I certainly hope it will.  It is a good proposal and is just too
useful to lose by getting bogged down in politics!

There is support for this in the Netsite servers (presumably) and the
latest version of WN has support.  I think that NCSA and CERN servers
should pass the Cookie: header to a CGI environment variable so an
nph-  CGI script should work with them.  Try it out.

-- 

John Franks 	Dept of Math. Northwestern University
		john@math.nwu.edu

eturn-Path: pcrecycl@montrealnet.ca 
Return-Path: <pcrecycl@montrealnet.ca>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA00652; Wed, 10 May 1995 23:09:09 +0500
Received: from Amita.MontrealNet.ca ([204.19.33.10]) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA09231; Wed, 10 May 1995 23:09:02 +0500
Received: from pcrecycl ([204.19.33.160]) by Amita.MontrealNet.ca (8.6.9/8.6.9) with SMTP id XAA01644; Wed, 10 May 1995 23:04:16 -0400
Date: Wed, 10 May 1995 23:04:16 -0400
Message-Id: <199505110304.XAA01644@Amita.MontrealNet.ca>
X-Sender: pcrecycl@amita.montrealnet.ca
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
To: ppetit@wct.fr, Multiple recipients of list <www-talk@www10.w3.org>
From: pcrecycl@montrealnet.ca (Eric Paquin)
Subject: Re: Netscape Navigator License
X-Mailer: <PC Eudora Version 1.4>
Content-Length: 1321

At 01:20 PM 9/5/95 +0500, Patrick Petit wrote:
>Hi,
>
>I'm not sure it's the right place to send that query, but here it goes....
>
>
>Well, I've been trying to get an answer from moreinfo@netscape.com
>without success. Maybe someone here has the answer.
>
>Does Netscape support Navigator's binary license for Linux and
>Solaris x86 platforms?
>
>Cheers,
>
>----------------------------------------------------------------
>Patrick Petit				Voice: (33) 1 42 36 90 47
>World Channel Technologies		Email: ppetit@wct.fr
>8, rue Saint Marc 75002 Paris		Fax: (33) 1 45 08 51 96
>France
>
>
>Thanks you for your E-Mail
Passez une Belle Journ=E9e
                         Fido...
                                . . . . . . . .
                                  ... .. .. .=20
                                   ..  ..  .
                                     . . .          7oUP
                                    *    *
                                   ***I I *** =20
                                  ***     ***
                                     * . *
                                     *   *=20
                                     * O *
                                      * *
                                      * *
                                    --------
                                  --        --

eturn-Path: agulbra@troll.no 
Return-Path: <agulbra@troll.no>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA05225; Thu, 11 May 1995 00:04:05 +0500
Received: from Norway.EU.net (nic.eunet.no) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA09720; Thu, 11 May 1995 00:03:59 +0500
Received: by Norway.EU.net with UUCP id AA26674
  (5.65c/IDA-1.4.4/EUnet/NO for www10.w3.org!www-talk); Thu, 11 May 1995 06:01:16 +0200
Date: Thu, 11 May 1995 03:21:01 +0200 (MET DST)
From: Arnt Gulbrandsen <agulbra@troll.no>
To: Jacques Caron <jcaron@pressimage.fr>
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Re: your mail
In-Reply-To: <v01510100abd4164f60dd@[194.2.222.120]>
Message-Id: <Pine.LNX.3.91.950511025542.184D-100000@pentagram.troll.no>
Return-Receipt-To: agulbra@troll.no
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 825

<deletia, regarding the possible evils of multiple-name dns server hosts>

> When you'll see some providers setting up a few hundred IP addresses each
> just for a single web server, you'll be sorry :-)

I won't be sorry, no, because I'll remember that if even one-half percent
of those customers chose that web solution _instead_ of a real IP-level
solution (and an automatic 256-address block allocation), they net effect
is an IP-address saving.

Regarding your WWW-specific RR, I don't think it's such a good idea.  It's
a convenient band-aid for a specific problem and doesn't help the next guy
very much. Since it uses ASCII strings it may even be hurtful due to
increased reply size.  DNS, IMHO, cannot be extended with one new RR per
new service on the net, and even starting on that path (MX) was a mistake.

--Arnt
eturn-Path: drtr1@cus.cam.ac.uk 
Return-Path: <drtr1@cus.cam.ac.uk>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AC14466; Thu, 11 May 1995 06:05:29 +0500
Received: from bootes.cus.cam.ac.uk by www10.w3.org (5.0/NSCS-1.0S) 
	id AA13334; Thu, 11 May 1995 06:04:37 +0500
Received: from grus.cus.cam.ac.uk [131.111.8.3] (ident = root) 
	by bootes.cus.cam.ac.uk with smtp 
	(Smail-3.1.29.0 #36) id m0s9V3X-000C0AC; Thu, 11 May 95 11:01 BST
Received: by grus.cus.cam.ac.uk 
	(Smail-3.1.29.0 #36) id m0s9V3W-0007aFC; Thu, 11 May 95 11:01 BST
Message-Id: <m0s9V3W-0007aFC@grus.cus.cam.ac.uk>
Date: Thu, 11 May 95 11:01 BST
Sender: drtr1@cus.cam.ac.uk (David Robinson)
To: nazgul@utopia.com
Subject: Re:  NCSA Server (1.3) and ScriptAlias problems
Cc: www-talk@www10.w3.org
From: drtr1@cam.ac.uk (David Robinson)
Content-Length: 3191

>The issue is how to deal with a ScriptAlias that is serving up multiple
>HTML files, e.g.:
>
>        ScriptAlias /bgc/ /xxx/bgc/access/
>        ScriptAlias /bgc /xxx/bgc/access/
>
>In particular, the problem is that the script can't tell the difference
>between being invoked with or without the trailing slash, and the
>difference is quite critical. I suspect that there should be a difference
>in PATH_INFO, but there isn't. Is that a bug, or is there another better
>way of dealing with this?

I think you are using ScriptAlias in a way the authors of the NCSA server
did not envisage, and hence are encountering a bug.

A config option like
   ScriptAlias /foo/ /path/bar/
was intended for aliasing _directories of scripts_, not individual scripts.
Thus a typical usage would be for an access to /foo/script.pl to cause the
server to execute /path/bar/script.pl as a CGI script.

Use of ScriptAlias as you describe, e.g.
 ScriptAlias /foo/ /path/script.pl/
might work, but cannot be relied on with the NCSA server.

This has been fixed in the Apache server (which is derived from the NCSA 1.3
code):
 ScriptAlias /bgc /xxx/bgx/access
would do exactly what you want.
Accesses to /bgc would invoke the script with a NULL PATH_INFO
Accesses to /bgc/ would invoke the script with a PATH_INFO of "/" etc.

>1. If you don't have the version with the trailing slash, the (non-trivial
>number of) people who hit the site without it will get a file not-found.
>This isn't the case when the server is handling access, but it is the case
>when you are using a ScriptAlias.
Bug.

>2. If you point them both at the same script, the script can't tell whether
>it was invoked with or without the slash (I believe there ought to be a
>difference in PATH_INFO, but there isn't).
You don't need two aliases, but just one.

>3. The script can do one of two things when faced with a directory name
>with no index.html. It can't do a redirect to the index.html file (which
>takes up a fair amount of time, and of course additional server resources),
>or it can just server up the file.
I won't assert what you _should_ do, but common practice is to return
the file without doing a redirect.

>4. If it serves up the file, the client will get all the relative links
>wrong. It's URL will look like "http://www.xxx.bar/foo", it will assume
>that "foo" is a file and make all links relative to the link instead of
>"foo/". You can try and fix this with a <base href=3D"correcturl"> in the
>index.html file, but that doesn't work for all browsers.
Common practice would be to redirect to the URL with the trailing slash,
if one is missing. However, I think you will find that returning a
BASE tag will work for the vast majority of your users. (N.B. but not
necessarily browsers.)

>5. My solution therefore, is to have a separate script, bound to the
>non-slash version of the name, which does nothing but redirect the user to
>the correct URL. Ugly, but there it is.

For apache, you would only need one script, which could check for a NULL
PATH_INFO and do a redirect in that case.

 David Robinson.

See <URL:http://www.apache.org/apache/> or
<URL:http://www.hyperreal.com/apache/> for details of apache.
eturn-Path: koen@win.tue.nl 
Return-Path: <koen@win.tue.nl>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AC18879; Thu, 11 May 1995 06:45:12 +0500
Received: from wswiop05.win.tue.nl by www10.w3.org (5.0/NSCS-1.0S) 
	id AA13696; Thu, 11 May 1995 06:45:01 +0500
Received: by wswiop05.win.tue.nl (8.6.10/1.45)
    id MAA01635; Thu, 11 May 1995 12:44:54 +0200
From: koen@win.tue.nl (Koen Holtman)
Message-Id: <199505111044.MAA01635@wswiop05.win.tue.nl>
Subject: Client handling of Expires: (was: Content-lengths of dynamic objects)
To: www-talk@www10.w3.org
Date: Thu, 11 May 1995 12:44:53 +0200 (MET DST)
Cc: koen@win.tue.nl (Koen Holtman)
In-Reply-To: <199505090830.KAA11662@heike.nads.de> from "Rainer Klute" at May 9, 95 05:24:25 am
X-Mailer: ELM [version 2.4 PL23]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 8bit
Content-Length: 1128      

Rainer Klute:
>
>>Unfortunately, my tests indicate that Netscape doesn't
>>handle the Expires header properly :(
>
>Not to defend Netscape against anything, but is there a client at
>all that handles Expires properly? Not even the CERN proxy does.

NCSA Mosaic (at least the version I am using) handles Expires: properly, by
not having an internal cache at all.

On a related note, there seem to be differing views on what it means for a
client to handle Expires: properly.  Some issues:

- if a document on screen expires, should the client automatically reload
  by contacting the server?
  In my opinion, it should not.

- if a document in the client history list expires, should the client
  reload it by contacting the server when it is selected from the history
  list?
  In my opinion, it should not (what is the meaning of the word `history'?).

- if the reload button is pressed on a document that is NOT expired, should
  the client fetch a new copy from the server nevertheless? 
  The answer `no' would make html authoring using a plain text editor and
  a WWW client for previewing difficult.

>Rainer Klute

Koen.
eturn-Path: heaney@cambridge.scr.slb.com 
Return-Path: <heaney@cambridge.scr.slb.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB07838; Thu, 11 May 1995 09:18:39 +0500
Received: from cambridge.scr.slb.com (platinum.cambridge.scr.slb.com) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA15214; Thu, 11 May 1995 09:18:30 +0500
Received: from snow.cambridge.scr.slb.com by cambridge.scr.slb.com (4.1/SMI-4.1/JC-platinum)
	id AA22854; Thu, 11 May 95 14:15:07 BST
Date: Thu, 11 May 95 14:15:06 BST
From: heaney@cambridge.scr.slb.com (Steve Heaney)
Message-Id: <9505111315.AA22854@cambridge.scr.slb.com>
To: www-talk@www10.w3.org
Subject: Remote publishing via PUT/POST
Content-Length: 1736


I'm trying to get a feel for the current state of play regarding remote 
publishing via the Web.  I would appreciate if anyone can fill in any 
gaps.  So far I've got the following:

 - PUT and POST have been in HTTP/1.0 for a long time although largely 
   unnused.  (The latest Internet draft (12-May-95) [1] however refers 
   to "current implementations of the POST and PUT methods" (7.2) - where 
   are they?).

 - The current implementations I am aware of are the "intelligent 
   publishing environment" from Georgia Inst. of Technology [2] and 
   the Symposia editor/browser [3] from Grif and INRIA.

 - a method for file upload does not yet seem to be agreed upon although 
   there is a draft proposal [5] from Nebel, Masinter at Xerox.
   (Several questions over the past few months on www-talk on this).

 - the CERN server supports PUT/POST via external scripts (this is used 
   by Symposia).  Symposia documentation implies that more complete 
   support in the CERN/W3O server is on its way [4].

And some questions:

 - are there any browsers supporting file upload and PUT (aside from 
   Symposia)?

 - are there any servers with generic PUT and POST support other than 
   CERN httpd?

 - when can we hope to see broad support from clients/servers?

 - is forms-based file upload a prerequesite for remote publishing using 
   current Web browsers?

Thanks in advance,

Steve


[1] http://www.w3.org/hypertext/WWW/Protocols/HTTP1.0/draft-ietf-http-spec.html
[2] Pitkow, Jones
    ftp://ftp.cc.gatech.edu/pub/software/publishing/ipe.tar
[3] http://symposia.inria.fr/
[4] http://symposia.inria.fr/symposia/userdoc/put/writable-server.html
[5] ftp://ds.internic.net/internet-drafts/draft-ietf-html-fileupload-02.txt
eturn-Path: @www10.w3.org:klute@heike.nads.de 
Return-Path: <@www10.w3.org:klute@heike.nads.de>
Received: from LCS.MIT.EDU (mintaka.lcs.mit.edu) by www19 (5.0/NSCS-1.0S) 
	id AB00251; Thu, 11 May 1995 10:00:01 +0500
Received: from www10.w3.org by MINTAKA.LCS.MIT.EDU id aa01481;
          11 May 95 9:56 EDT
Received: from heike.nads.de by www10.w3.org (5.0/NSCS-1.0S) 
	id AA15612; Thu, 11 May 1995 09:56:29 +0500
Received: from heike (localhost [127.0.0.1]) by heike.nads.de (8.6.12/8.6.12) with SMTP id PAA20398; Thu, 11 May 1995 15:55:32 +0200
Message-Id: <199505111355.PAA20398@heike.nads.de>
To: koen@win.tue.nl
Cc: Multiple recipients of list <www-talk@www10.w3.org>, klute@heike.nads.de
Subject: Re: Client handling of Expires: (was: Content-lengths of dynamic objects) 
Reply-To: Rainer Klute <klute@nads.de>
Organization: NADS GmbH, Germany
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
In-Reply-To: Your message of Thu, 11 May 1995 06:56:55 +0500.
             <199505111044.MAA01635@wswiop05.win.tue.nl> 
Date: Thu, 11 May 1995 15:55:31 +0200
From: Rainer Klute <klute@nads.de>
Content-Length: 595

koen@win.tue.nl (Koen Holtman) in
<199505111044.MAA01635@wswiop05.win.tue.nl>:

>NCSA Mosaic (at least the version I am using) handles Expires: properly, by
>not having an internal cache at all.

That's not sufficient. Mosaic still caches inline images in
memory and does to pay attention to and Expires header coming with
the image.

Best regards
Rainer Klute

  Dipl.-Inform. Rainer Klute        NADS - Advertising on nets
  NADS GmbH
  Emil-Figge-Str. 80                Tel.: +49 231 9742570
D-44227 Dortmund                    Fax:  +49 231 9742571

            <http://www.nads.de/~klute/>
eturn-Path: nazgul@utopia.com 
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA00423; Thu, 11 May 1995 10:04:27 +0500
Received: from utopia.com (utopia.wing.net) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA15715; Thu, 11 May 1995 10:04:23 +0500
Received: from [204.57.39.6] by utopia.com (8.6.12/SONY-4.0MX)
	id JAA16331; Thu, 11 May 1995 09:51:30 -0400
Return-Path: <nazgul@utopia.com>
X-Sender: nazgul@mailhost.utopia.com
Message-Id: <v02110121abd7bcd32b47@[204.57.39.6]>
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Date: Thu, 11 May 1995 09:53:26 -0400
To: drtr1@cam.ac.uk (David Robinson)
From: nazgul@utopia.com (Kee Hinckley)
Subject: Re:  NCSA Server (1.3) and ScriptAlias problems
Cc: www-talk@www10.w3.org
Content-Length: 1953

At 11:01 AM 5/11/95, David Robinson wrote:
>>The issue is how to deal with a ScriptAlias that is serving up multiple
>>HTML files, e.g.:
>>
>>        ScriptAlias /bgc/ /xxx/bgc/access/
>>        ScriptAlias /bgc /xxx/bgc/access/
>>
>>In particular, the problem is that the script can't tell the difference
>>between being invoked with or without the trailing slash, and the
>I think you are using ScriptAlias in a way the authors of the NCSA server
>did not envisage, and hence are encountering a bug.

I'm not sure how they could fail to envision that usage and have it work at
all, but okay.

>This has been fixed in the Apache server (which is derived from the NCSA 1.=
3
>code):
> ScriptAlias /bgc /xxx/bgx/access
>would do exactly what you want.

Great. BTW, I recently reported to NCSA that they don't check the number of
aliases and if the number overflows the default (20 total Aliases and
ScriptAliases) they overwrite memory randomly. Do you know if that is fixed
in Apache?

>>3. The script can do one of two things when faced with a directory name
>>with no index.html. It can't do a redirect to the index.html file (which
>>takes up a fair amount of time, and of course additional server resources)=
,
>>or it can just server up the file.
>I won't assert what you _should_ do, but common practice is to return
>the file without doing a redirect.

Right, and certainly would if I could. Apache soulds like an excellent
solution, but it does require getting our site providers to switch to it.

>BASE tag will work for the vast majority of your users. (N.B. but not
>necessarily browsers.)

I don't see the distinction there.


Kee Hinckley      Utopia Inc. - Cyberspace Architects=81    617/721-6100
nazgul@utopia.com                               http://www.utopia.com/

I'm not sure which upsets me more: that people are so unwilling to accept
responsibility for their own actions, or that they are so eager to regulate
everyone else's.


eturn-Path: marsj@ida.liu.se 
Return-Path: <marsj@ida.liu.se>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA03633; Sat, 13 May 1995 15:19:21 +0500
Received: from ida.liu.se (portofix.ida.liu.se) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA00786; Sat, 13 May 1995 15:19:11 +0500
Received: from obel6 by ida.liu.se (5.65b/ida.minimaster-V1.0b6d5)
	id AA11903; Sat, 13 May 95 20:59:06 +0200
From: Martin Sjolin <marsj@ida.liu.se>
Received: by obel6 (4.1/ida.slave-V1.0b6d6)
	id AA21276; Sat, 13 May 95 20:59:06 +0200
Date: Sat, 13 May 95 20:59:06 +0200
Message-Id: <9505131859.AA21276@obel6>
To: www-talk@www10.w3.org
Subject: SUMMARY: benchmark program for HTTP server's
X-Attribution: msj
X-Mailer: VM Version 5.72 (beta) with GNU Emacs 18.58.4 of Tue Jun  9 1992 on panix (berkeley-unix)
Organization: Department of Computer Science, LiTH, Sweden
X-Disclaimer: My opinions do not necessarily represent those of my employer
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Length: 1796

I wrote:

 > 1. take a base URL
 > 2. retrieves all URL in the base document, but
 > 3. do not goes outside the server (e.g. restrict the set of
 >    allowed URL),
 > 4. minimum time between HEADs/GETs,
 > 5. runs under unix (preferable SunOS 4.1 - i have ported software
 >    to hp-ux/solaris 2.x/dec osf/4.3bsd/aix/ultrix/sgi/linux)

I better clarify (4) - i would like to retreive all URL from
a site, but according to (4), have minimum time between two
GETs as to avoid overloading the server. 

Answers to the query:

A). http://www.inria.fr/koala/abaird/oscheme/oscheme.html with
    the  "www-list" scripts (from Anselm.Baird_Smith@inria.fr)

B). http://www.ics.uci.edu/WebSoft/MOMspider/ (MOMspider)
    (from joshuap@sdsc.edu (Joshua Polterock))

C). http://iamwww.unige.ch/~scg/Src/Scripts/  with the
    explore script (diana@seldon.terminus.com (Cookie Monster))

D). Simon Spero <ses@tipper.oit.unc.edu> have a set of programs 
    for benmarking.

E). rst@ai.mit.edu (Robert S. Thau) has written a logfile replay program,
    runs SunOS, which reports the main latency for every 100 transactions,
    and which handle multiple outstanding requests. Found at
    ftp://ftp.ai.mit.edu/pub/users/rst/monkey.c

F). www2dot from einpost@win.tue.nl (Reinier Post), it might no
    fill the (4) requirement. Contact Reiner Post. Based on libwww2.

BTW, I probably try to use (C). For those interested, i'm running
a gateway (CGI based), which generates HTML pages on the fly. 
I'm interested the above to profile the gateway (written in C).

thanks to all who answered,

msj
--
Martin Sj\"olin | http://www.ida.liu.se/labs/iislab/people/marsj
Department of Computer Science, LiTH, S-581 83 Link\"oping, SWEDEN 
phone : +46 13 28 24 10 | fax : +46 13 28 26 66 | e-mail: marsj@ida.liu.se 
eturn-Path: martin@mrrl.lut.ac.uk 
Return-Path: <martin@mrrl.lut.ac.uk>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB06206; Sat, 13 May 1995 15:35:37 +0500
Received: from gizmo.lut.ac.uk by www10.w3.org (5.0/NSCS-1.0S) 
	id AA00812; Sat, 13 May 1995 15:33:58 +0500
Received: from localhost (martin@localhost) by gizmo.lut.ac.uk (8.6.12/8.6.9) with SMTP id SAA22023 for <www-talk@w3.org>; Sat, 13 May 1995 18:22:04 +0100
Message-Id: <199505131722.SAA22023@gizmo.lut.ac.uk>
To: www-talk@w3.org
X-Uri: <URL:http://www.mrrl.lut.ac.uk/~martin>
Subject: problem with ftp.w3.org
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Sat, 13 May 1995 18:22:04 +0100
From: Martin Hamilton <martin@mrrl.lut.ac.uk>
Content-Length: 142

Anyone else having difficulties with this machine ... ?

  > dir

  Can't create data socket (128.141.201.214,20): Interrupted system call.


eturn-Path: batie%ibeam.jf.intel.com@www10.w3.org 
Return-Path: <batie%ibeam.jf.intel.com@www10.w3.org>
Received: from mintaka.lcs.mit.edu by www19 (5.0/NSCS-1.0S) 
	id AC13451; Sat, 13 May 1995 17:12:00 +0500
Received: from www10.w3.org by MINTAKA.LCS.MIT.EDU id aa07023;
          11 May 95 17:28 EDT
Received: from ibeam.intel.com (ibeam.jf.intel.com) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA19650; Thu, 11 May 1995 17:28:28 +0500
Received: from aahz.jf.intel.com by ibeam.intel.com with smtp
	(Smail3.1.28.1 #6) id m0s9flX-0003fEC; Thu, 11 May 95 14:27 PDT
Received: by aahz.jf.intel.com (Smail3.1.28.1 #13)
	id m0s9fmV-000RjyC; Thu, 11 May 95 14:28 PDT
Message-Id: <m0s9fmV-000RjyC@aahz.jf.intel.com>
From: batie@ibeam.jf.intel.com (Alan Batie)
Subject: Cern Proxy restrictions
To: www-talk@www10.w3.org
Date: Thu, 11 May 1995 14:28:10 -40962758 (PDT)
X-Mailer: ELM [version 2.4 PL17]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 777       

I helped a local school setup a web server, and they've also setup a
CERN proxy server for outgoing access.  Being a high school, and students
being students, they've found certain, shall we say, "popular" sites
that the school would prefer they found elsewhere so as not to become
front page news.  Well known arguments pro and con aside, from a
perusal of the documentation, it looks like the CERN server can only
be told to Allow access to a set of sites, but cannot be told to
*Disallow* access.  Could someone point me in the right direction?

Thanks!

-- 
Alan Batie                     ------
batie@aahz.jf.intel.com        \    /        X.400: same day delivery
+1 503-264-8844 (voice)         \  /         ...in a nanosecond world.
+1 503-264-6388 (fax)            \/
eturn-Path: janos%specter.com@www10.w3.org 
Return-Path: <janos%specter.com@www10.w3.org>
Received: from mintaka.lcs.mit.edu by www19 (5.0/NSCS-1.0S) 
	id AT15409; Sat, 13 May 1995 23:59:29 +0500
Received: from www10.w3.org by MINTAKA.LCS.MIT.EDU id aa21769;
          12 May 95 15:48 EDT
Received: from xensei2.xensei.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA27119; Fri, 12 May 1995 15:48:21 +0500
Received: from poker (specter.com [198.151.175.140]) by xensei2.xensei.com (8.6.11/Xensei-M1.01/122994-BRC) with SMTP id PAA02073; Fri, 12 May 1995 15:47:19 -0400
Message-Id: <199505121947.PAA02073@xensei2.xensei.com>
X-Sender: janos@popmail.xensei.com
X-Mailer: Windows Eudora Version 1.4.4
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Fri, 12 May 1995 15:47:58 -0400
To: webwatch@xensei.com
From: janos@xensei.com (Joseph Janos)
Subject: WebWatch 1.0 released
Content-Length: 1576

Specter, Inc. is proud to announce the immediate availability 
of WebWatch 1.0 for Windows. WebWatch is a tool for keeping 
track of changes in selected Web documents.

Given an HTML document referencing URLs on the Web, WebWatch 
produces a filtered list, containing only those URLs that have 
been modified since a given time.

The input file can be e.g.
- the bookmark file of Netscape; or 
- the exported (to HTML format) hotlist of Mosaic; or 
- any other, standard HTML document, edited locally or 
retrieved from the Web.

The time used for filtering
- can be given as a global setting that applies to all URLs, or 
- WebWatch can derive it automatically, using the time of your 
last visit to the document, as recorded by your Web browser in 
your local HTML (bookmark) file.

WebWatch generates a local HTML document that contains links to 
only those documents which were updated after the given date. 
You can load this document into any Web browser and use it to 
navigate to the updated documents.

WebWatch stores its arguments in a parameter file. Once you 
have customized the program to your needs, using its graphical 
front-end, you can have it run in unattended mode, periodically.

WebWatch supports the use of proxy servers.

WebWatch is currently available for every (16 and 32 bit) 
Windows platforms.

WebWatch is shareware. Single user registration after 30 days 
evaluation is US $18. We offer free registration for every new 
bug found.

For further information and to download your copy, please visit 
http://www.specter.com/users/janos/specter/

eturn-Path: alain@hyperman.co.il 
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB08419; Sun, 14 May 1995 09:11:45 +0500
Received: from hyperman.co.il by www10.w3.org (5.0/NSCS-1.0S) 
	id AA05795; Sun, 14 May 1995 09:11:35 +0500
Received: by hyperman.co.il (4.0/SONY-4.0MX)
	id AA03166; Sun, 14 May 95 13:11:49+010
Date: Sun, 14 May 95 13:11:49+010
From: alain@hyperman.co.il
Return-Path: <alain@hyperman.co.il>
Message-Id: <9505141211.AA03166@hyperman.co.il>
To: www-talk@www10.w3.org
Subject: Viewer for UNIX
Content-Length: 346

I am not sure this is the right place to ask, but I hope somebody in this list will know:

I am looking for a place to find Viewers of documents like Adobe's Acrobat,
Does anybody know where (apart from Acrobat), I can find viewer 
that enable to exchange documents, running under UNIX System V.4?
Is there a forum discussing these?
Thanks
Alain
eturn-Path: timt%ee.pdx.edu@www10.w3.org 
Return-Path: <timt%ee.pdx.edu@www10.w3.org>
Received: from mintaka.lcs.mit.edu by www19 (5.0/NSCS-1.0S) 
	id AL25757; Mon, 15 May 1995 22:41:34 +0500
Received: from www10.w3.org by MINTAKA.LCS.MIT.EDU id aa27990;
          15 May 95 15:30 EDT
Received: from ursula.ee.pdx.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA17375; Mon, 15 May 1995 15:30:51 +0500
Received: from darwin.cat.pdx.edu (root@darwin.cat.pdx.edu [131.252.21.110])
        by ursula.ee.pdx.edu (8.6.10/CATastrophe-12/23/94-P) with ESMTP
        id MAA18275; Mon, 15 May 1995 12:33:38 -0700 for <www-talk@www10.w3.org>
Received: from localhost (timt@localhost [127.0.0.1])
        by darwin.cat.pdx.edu (8.6.10/CATastrophe-9/18/94-C) with ESMTP
        id MAA07654; Mon, 15 May 1995 12:27:51 -0700 for <www-talk@www10.w3.org>
Message-Id: <199505151927.MAA07654@darwin.cat.pdx.edu>
X-Mailer: exmh version 1.6 4/21/95
To: www-talk@www10.w3.org
Subject: netscape caches.
X-Uri: http://www.cat.pdx.edu/~timt
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Mon, 15 May 1995 12:27:49 -0700
From: Tim Trautmann <timt@ee.pdx.edu>
Content-Length: 1170

I am sorry if this is a faq OR the wrong place to ask but I haven't been able 
to keep up with my mailing lists lately.
The majority of our students (including myself) use netscape for browsing. 
This poses a slight delemma for us right now because netscape's private caches 
are disk hawgs. At this point we are low (very low) on disk space on our 
filesystems. Last time we did a find on the .netscape_cache directory, we were 
able to weed out about 250 megs of space. YES we are buying more hard drives 
for our fileservers, but being at a state university this takes time going 
through all the red tape. 
Anyways to make a long story even longer. Is there a way to globally set the 
size of the personal netscape caches (maybe through a secret app-defaults 
flag)? We are not running a proxy because of the above reasons, BUT if we did, 
would netscape still use it's own caching mechanism or would it honor a 
central proxy server as a cache?!

Thanks for all your answers in advance,

-- Tim

---
Tim Trautmann               | e-mail: timt@ee.pdx.edu
Portland State University   | phone:  (503)725-7056
Computer Action Team        | http://www.cat.pdx.edu/~timt
eturn-Path: Massimo.Albanesi@fao.org 
Return-Path: <Massimo.Albanesi@fao.org>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA02165; Tue, 16 May 1995 10:21:35 +0500
Received: from dxmint.cern.ch by www10.w3.org (5.0/NSCS-1.0S) 
	id AA25112; Tue, 16 May 1995 10:21:19 +0500
Received: from faov2b by dxmint.cern.ch
	id AA27594; Tue, 16 May 1995 16:21:10 +0200
Return-Receipt-To: Massimo.Albanesi@fao.org
Received: from Bogus.FAO.ORG by faov02.fao.org (PMDF V4.3-11 #3703)
 id <01HQKS9RURKG8Y546S@faov02.fao.org>; Tue, 16 May 1995 16:18:10 +0100
Received: from msmail.fao.org (168.202.2.28)
 by faov01.fao.org (PMDF V4.3-10 #3703) id <01HQKSAI4S9S000DO6@faov01.fao.org>;
 Tue, 16 May 1995 16:18:47 +0100
Received: by msmail.fao.org with Microsoft Mail id <2FC3BEEF@msmail.fao.org>;
 Tue, 16 May 95 16:19:11 +02
Date: Tue, 16 May 1995 16:19:00 +0002
From: "Albanesi, Massimo" <Massimo.Albanesi@fao.org>
To: "www-talk@info.cern.ch" <www-talk@w3.org>
Message-Id: <2FC3BEEF@msmail.fao.org>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN
Content-Transfer-Encoding: 7BIT
Content-Length: 12

SUBSCRIBE


eturn-Path: dmk@allegra.att.com 
Return-Path: <dmk@allegra.att.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AC21183; Tue, 16 May 1995 13:34:46 +0500
Received: from research.att.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA26846; Tue, 16 May 1995 13:34:30 +0500
Message-Id: <9505161734.AA26846@www10.w3.org>
Received: by research.att.com; Tue May 16 12:46 EDT 1995
Received: from phero.tempo.att.com by allegra.tempo.att.com; id AA06067; Tue, 16 May 95 12:46:10 EDT
Date: Tue, 16 May 95 12:46:08 EDT
From: dmk@allegra.att.com (Dave Kristol)
To: www-talk@www10.w3.org
Subject: TCP shutdown problems
Content-Length: 830

I've written my own Web server.  (Doesn't everyone? :-)  Unlike the
public domain servers I've seen, I don't time out sends (in response to
GET, for example).

I've observed that my server process sometimes hangs after a
transaction is complete, waiting for a TCP connection to shut down.
(On Unix systems, in the close() following shutdown().) I've looked at
the logs to try to associate this failing with particular client
programs or operating systems or remote hosts, but it seems pretty
general.

For now I've been forced to timeout the shutdown after 10 minutes.  But
I suspect there's an OS bug (on my side?:  SGI IRIX 5.2) in the area of
connection shutdown.  (I believe the send timeout on the other servers
handles this problem along the way.)

Can anyone shed any light on the problem and/or its solution?
Dave Kristol
eturn-Path: david@kwantlen.bc.ca 
Return-Path: <david@kwantlen.bc.ca>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA04487; Tue, 16 May 1995 16:29:00 +0500
Received: from gateway.kwantlen.bc.ca by www10.w3.org (5.0/NSCS-1.0S) 
	id AA28399; Tue, 16 May 1995 16:28:48 +0500
Received: from trex (trex.kwantlen.bc.ca) by gateway.kwantlen.bc.ca (5.65c/SMI-4.1)
	id AA04103; Tue, 16 May 1995 13:28:36 -0700
Received: by trex (5.0/SMI-SVR4)
	id AA01166; Tue, 16 May 1995 13:28:34 +0800
Date: Tue, 16 May 1995 13:28:33 -0700 (PDT)
From: David Dumaresq <david@kwantlen.bc.ca>
X-Sender: david@trex
To: www-talk@www10.w3.org
Subject: Netscape and Adobe
Message-Id: <Pine.SOL.3.91.950516132404.23936D-100000@trex>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 665

I was reading a rather old issue of InformationWeek (Apr 10/95) about 
Netscape and Adobe's agreement to add Acrobat to Navigator 1.1. Well, 
1.1N is out but I don't see any sign of Acrobat.

Anyone heard about this?


David.
-----------------------------------------------------------------------------
David Dumaresq                                	| email: david@kwantlen.bc.ca
Programmer/Analyst, Info. Systems & Computing   | phone: (604) 599-2120
Kwantlen University College, BC, Canada    	| fax:   (604) 599-2068

           "The world is one country and mankind its citizens."
-----------------------------------------------------------------------------

eturn-Path: luotonen@netscape.com 
Return-Path: <luotonen@netscape.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA00852; Wed, 17 May 1995 16:29:30 +0500
Received: from neon.netscape.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA07050; Wed, 17 May 1995 16:27:18 +0500
Received: (from luotonen@localhost) by neon.netscape.com (950215.SGI.8.6.10/8.6.9) id NAA06472; Wed, 17 May 1995 13:22:40 -0700
From: Ari Luotonen <luotonen@netscape.com>
Message-Id: <199505172022.NAA06472@neon.netscape.com>
Subject: Byte ranges -- formal spec proposal
To: www-talk@w3.org, http-wg@cuckoo.hpl.hp.com
Date: Wed, 17 May 1995 13:22:40 -0700 (PDT)
Cc: uri@bunyip.com
X-Mailer: ELM [version 2.4 PL24]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 7118      


Please review and comment.

     _________________________________________________________________
   
                        BYTE RANGES WITH URLS AND HTTP
                                       
   May 16, 1995
   
   John Franks <john@math.nwu.edu>
   Ari Luotonen <ari@netscape.com>
   
   
     _________________________________________________________________
   
   
   
Table of Contents

     * Overview
     * The byterange Parameter
          + Examples
     * The Range HTTP Response Header
          + Examples
     * Miscellaneous
   
     _________________________________________________________________
   
   
   
Overview

   There are number of Web applications that would benefit from being
   able to request the server to give a byte range of a document. It may
   be argued that this should be left as a server-specific feature in the
   opaque URL, as the "parameters" used in URLs that may be available or
   useful can vary from server to server. However, there are reasons why
   standardizing the byte range feature would be beneficial.
   
   One of the primary reasons is to be able to support byte ranges in
   proxy servers. Without a standard proxy servers will have to treat
   each different byte range of a given document as a separate document.
   Should the notion of a byte range be standard, not only would it
   prevent portions of documents to be multiply cached, but it would make
   it possible for the server to generate range responses directly from
   its cache.
   
   
     _________________________________________________________________
   
   
   
Description of the byterange URL Parameter

     * The byte range request is attached to the end of the URL,
       separated by a semi-colon.
       
     * The parameter name is byterange.
       
     * The range is two integers greater than zero, separated by a
       hyphen.
       
     * The first byte in file is byte number 1.
       
     * One of the numbers may be missing, but both of them cannot at the
       same time.
       
     * If the first number is missing, it means to return n bytes from
       the end of the file, where n is the second number.
       
     * If the second number is missing, it means the end of file.
       
     * If the second number is larger than the size of the file, it is
       taken to mean the size of the file.
       
     * The first integer must always be less than or equal to the second
       one.
       
     * The range includes both limits, that is, a range 500-1000 means
       the 501 bytes between 499 and 1001.
       
     * There may be multiple ranges, separated by a comma. The order of
       the ranges is the preferred order in which the ranges should be
       returned.
       
     * In place of a range there may be only a single positive integer.
       This means only a single byte. This is not meaningful for byte
       ranges, but if and when this scheme gets extended to allow other
       kinds of ranges (such as lines, chapters or pages), this will be
       useful.
       
   
     _________________________________________________________________
   
   
   
  EXAMPLES OF THE BYTERANGE URL PARAMETER
  
   The first 500 bytes:

        http://host/dir/foo;byterange=1-500

   The second 500 bytes:

        http://host/dir/foo;byterange=501-1000

   Bytes from 501 until the end of file:

        http://host/dir/foo;byterange=501-

   The last 500 bytes of the file:

        http://host/dir/foo;byterange=-500

   Two separate ranges:

        http://host/dir/foo;byterange=51-100,201-250

   The first 100 bytes, 1000 bytes starting from the byte number 500, and
   the remainder of the file starting from byte number 4000:

        http://host/dir/foo;byterange=1-100,500-1499,4000-

   The first 100 bytes, 1000 bytes starting from the byte number 500, and
   the last 200 bytes of the file:

        http://host/dir/foo;byterange=1-100,500-1499,-200

   
     _________________________________________________________________
   
   
   
Description of the HTTP Response Headers

   If the request includes multiple ranges, the response is sent back as
   a multipart MIME message.
   
   If there are overlapping ranges the behaviour for each range doesn't
   change. That is, a range will not be truncated, or left out, just
   because there is an overlap.
   
   The following HTTP response header is sent back to provide
   verification and information about the range and total size of the
   file:

        Range: bytes X-Y/Z

   where:
   X
          is the number of the first byte returned (the first byte is
          byte number one).
          
   Y
          is the number of the last byte returned (in case of the end of
          the file this is the same number as the size of the file in
          bytes).
          
   Z
          is the total size of the file in bytes.
          
   
     _________________________________________________________________
   
   
   
  EXAMPLES OF THE RANGE: HEADER
  
   The first 500 bytes of a 1234 byte file:

        Range: bytes 1-500/1234

   The second 500 bytes of the same file:

        Range: bytes 501-1000/1234

   Bytes from 501 until the end of the same file:

        Range: bytes 501-1234/1234

   The last 500 bytes of the same file:

        Range: bytes 735-1234/1234

   
     _________________________________________________________________
   
   
   
Miscellaneous

   There are other kinds of ranges that can be addressed in a similar
   fashion; this document does not define them, but both the URL
   parameter and the Range: header are defined so that it is possible to
   extend them. This byte range specification applies to any
   content-type. There may be range schemes that are meaningful to only
   certain types of documents.
   
   As an example, there might be a linerange URL parameter, with the same
   kind of range specification, and the Range: header would then specify
   the numbers in lines. Example:

        http://host/dir/foo;linerange=21-30

   The response from a 123 line file would be:

        Range: lines 21-30/123

   This could be useful for such things as structured text files like
   address lists or digests of mail and news, but isn't meaningful to
   such document types as GIF or PDF.
   
   Other examples might be document format specific ranges, such as
   chapters:

        http://host/dir/foo;chapterrange=1-3

        Range: chapters 1-3/12

   Or just the first chapter:

        http://host/dir/foo;chapterrange=1

        Range: chapters 1/12

  MULTIPLE URL PARAMETERS
  
   If at some point there will be multiple simultaneous URL parameters,
   they should be separated by the ampersand character (just like
   multiple values are encoded in the FORM request).
   
   
     _________________________________________________________________
   
   John Franks, Ari Luotonen 

Cheers,
--
Ari Luotonen				ari@netscape.com
Netscape Communications Corp.		http://home.netscape.com/people/ari/
501 East Middlefield Road
Mountain View, CA 94043, USA		Netscape Server Development Team
eturn-Path: frank@localhost.w3.org 
Return-Path: <frank@localhost.w3.org>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA11794; Wed, 17 May 1995 18:29:21 +0500
Received: from localhost (dialip25.tddc.net) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA08121; Wed, 17 May 1995 18:27:18 +0500
Received: (from frank@localhost) by localhost (8.6.9/8.6.9) id RAA01603; Wed, 17 May 1995 17:08:18 -0500
Date: Wed, 10 May 1995 08:27:33 -0500 (CDT)
From: Franklin Schmidt <frank@localhost.w3.org>
Reply-To: fschmidt@tddc.net
To: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Re: Session tracking
In-Reply-To: <199505100712.AAA20303@aslan.math.hmc.edu>
Message-Id: <Pine.LNX.3.91.950510081140.7070A-100000@localhost>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Resent-Date: Wed, 17 May 1995 17:07:35 -0500 (CDT)
Resent-From: Franklin Schmidt <frank@localhost.w3.org>
Resent-To: Multiple recipients of list <www-talk@www10.w3.org>
Resent-Message-Id: <Pine.LNX.3.91.950517170735.1597A@localhost>
Content-Length: 1762

On Wed, 10 May 1995, Jared Rhine wrote:

> [The topic was session-tracking proposals, esp the proposed "Session-ID"
> header.]

I have just joined this mailing list, so I have only seen one message in
this thread, but I have a few comments.

User-tracking is much better than session-tracking.  You can track a session
by tracking a user, and additionally you can keep user specific information
between sessions.

Magic cookies are an over-complicated idea that could also be much better
handled by user-tracking.  The shopping basket or any other user specific
info should be kept at the server side.  The problem of abandoned baskets
could be handled essentially as it is in the supermarket, by timing out
inactive baskets and then restocking the items.

>   JP> 1) identify sessions (esp. from firewalled domains)
>   JP> [...]
>   JP> If the 'From:' field is widely used, you'd have what you wanted(1).
> 
> No, I don't believe so, because one of the stated requirements is that we be
> able to identify sessions without compromising the identity of the user.

I have posted a suggestion for how to handle user-tracking in several
newsgroups, where it was ignored.  Here is the suggestion again:

There is no reasonable way to uniquely identify users from a CGI script. 
HTTP_FROM (which contains the user's e-mail address) would work, but it
generally isn't supported because people do not want their e-mail address
distributed without their permission.  So I would like to propose a new
environment variable called HTTP_FROM_CRYPT which would contain the user's
e-mail address encrypted (in a consistent way).  (For example, the Unix
crypt() function could be used with a standard "salt" value.) This would
provide a safe unique identifier for each user.

eturn-Path: rdaniel@acl.lanl.gov 
Return-Path: <rdaniel@acl.lanl.gov>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AC13868; Wed, 17 May 1995 19:02:40 +0500
Received: from acl.lanl.gov by www10.w3.org (5.0/NSCS-1.0S) 
	id AA08373; Wed, 17 May 1995 19:00:11 +0500
Received: from idaknow.acl.lanl.gov (idaknow.acl.lanl.gov [128.165.148.24]) by acl.lanl.gov (8.6.11/8.6.10) with ESMTP id QAA04601; Wed, 17 May 1995 16:59:39 -0600
From: "Ronald E. Daniel" <rdaniel@acl.lanl.gov>
Received: (rdaniel@localhost) by idaknow.acl.lanl.gov (8.6.9/8.6.4) id QAA24514; Wed, 17 May 1995 16:59:38 -0600
Date: Wed, 17 May 1995 16:59:38 -0600
Message-Id: <199505172259.QAA24514@idaknow.acl.lanl.gov>
To: http-wg@cuckoo.hpl.hp.com, luotonen@netscape.com, www-talk@w3.org
Subject: Re:  Byte ranges -- formal spec proposal
Cc: uri@bunyip.com
Content-Length: 4845

Thus spoke Ari Luotonen <luotonen@netscape.com> (at least on Wed, 17 May 1995)

>      _________________________________________________________________
>    
>                         BYTE RANGES WITH URLS AND HTTP


We have been putting off the problem of fragment identifiers, and this
is a good start on the problem. I have a few reflex objections about
details - such as preferring 0-based addressing to 1-based - but they
are very minor. My major objection is that I would like to see
byterange addressing as one component in a more general fragment
identification architecture. The "Miscellaneous" section, quoted below,
mentions the possibility of combining different addressing schemes, but
does not provide any specification. I would be a LOT happier if we
could have an overall scheme that byterange, paragraph, row/col, word,
stanza, and other addressing schemes could fit into.

For example, I might want queries such as:

 Get the value of the <title> element in an HTML file
 http://host/path;generic-id="title"

 Get bytes 1-5 of the second paragraph of a file
 http://host/path;para=2&byterange=1-5

 Get a portion of a JPEG
 http://host/path.jpg;rows=37-99&cols=53-200


When we start looking at the addressing needs of a variety of
specifiers (rows/cols, paragraphs, ...) then we may find that
we would prefer different choices of index base, inclusion or
exclusion of the elements at the extremes of the range, etc.



> Miscellaneous
> 
>    There are other kinds of ranges that can be addressed in a similar
>    fashion; this document does not define them, but both the URL
>    parameter and the Range: header are defined so that it is possible to
>    extend them. This byte range specification applies to any
>    content-type. There may be range schemes that are meaningful to only
>    certain types of documents.
>    
>    As an example, there might be a linerange URL parameter, with the same
>    kind of range specification, and the Range: header would then specify
>    the numbers in lines. Example:
> 
>         http://host/dir/foo;linerange=21-30
> 
>    The response from a 123 line file would be:
> 
>         Range: lines 21-30/123
> 
>    This could be useful for such things as structured text files like
>    address lists or digests of mail and news, but isn't meaningful to
>    such document types as GIF or PDF.
>    
>    Other examples might be document format specific ranges, such as
>    chapters:
> 
>         http://host/dir/foo;chapterrange=1-3
> 
>         Range: chapters 1-3/12
> 
>    Or just the first chapter:
> 
>         http://host/dir/foo;chapterrange=1
> 
>         Range: chapters 1/12
> 
>   MULTIPLE URL PARAMETERS
>   
>    If at some point there will be multiple simultaneous URL parameters,
>    they should be separated by the ampersand character (just like
>    multiple values are encoded in the FORM request).


We need to define more than just the syntax of how multiple
parameters will be seperated. We need to define the semantics
of foo=n1-n2&bar=n3-n4. Does the "bar" parameter apply to the
result of the "foo" parameter? Vice versa? Or do we return the
two selections seperately the way you specify with foo=n1-n2,n3-n4  ?

How are errors to be handled when we specify a range that is
longer than the file? What about when the starting offset of
the range is greater than the length of the file? 

Byteranges are pretty nice since they are broadly applicable,
but I am not sure what it means to ask for a byterange of
a database. This problem is even more acute when we get into
parameters such as "paragraph", "row/col", "stanza", etc.
How are we to indicate when a parameter is inappropriate for
a URL, such as paragraph for an image? Usually row/col
will be inappropriate for HTML files, but if we have previously
selected a table then it is the natural way to get a table
element. How do we do that?

If we do not develop a uniform architecture for fragment
identification, we are going to have a slew of partial solutions before
we wise up and develop a uniform treatment. Then everyone will be
pissed because of differing addressing conventions, code bloat, etc.
and a total inability to make the uniform scheme match the previous
partial solutions.

My understanding is that HyTime can handle this uniform
fragment identification. Can people knowledgeable about HyTime
talk about the good *and bad* points of using HyTime addressing
for URI fragment identification? Is there a way we can
start small, with just byterange selection, then grow our
capabilities?



Ron Daniel Jr.                email: rdaniel@acl.lanl.gov
Advanced Computing Lab        voice: (505) 665-0597
MS B-287  TA-3  Bldg. 2011      fax: (505) 665-4939
Los Alamos National Lab        http://www.acl.lanl.gov/~rdaniel/
Los Alamos, NM,  87545    tautology: "Conformity is very popular"
eturn-Path: connolly@beach.w3.org 
Return-Path: <connolly@beach.w3.org>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA16132; Wed, 17 May 1995 19:42:12 +0500
Received: from beach.w3.org by www10.w3.org (5.0/NSCS-1.0S) 
	id AA08736; Wed, 17 May 1995 19:39:46 +0500
Received: from beach.w3.org (connolly@localhost [127.0.0.1]) by beach.w3.org (8.6.9/8.6.9) with ESMTP id TAA10475; Wed, 17 May 1995 19:39:39 -0400
Message-Id: <199505172339.TAA10475@beach.w3.org>
To: Ari Luotonen <luotonen@netscape.com>
Cc: www-talk@w3.org, http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com, uri@bunyip.com
Subject: Re: Byte ranges -- formal spec proposal 
In-Reply-To: Your message of "Wed, 17 May 1995 13:22:40 PDT."
             <199505172022.NAA06472@neon.netscape.com> 
Date: Wed, 17 May 1995 19:39:35 -0400
From: "Daniel W. Connolly" <connolly@beach.w3.org>
Content-Length: 1027

In message <199505172022.NAA06472@neon.netscape.com>, Ari Luotonen writes:

A nice, clear, complete proposal. As you say, this could be done as a
server-private mechanism, but there's no reason why everybody
shouldn't do it the same way.

A couple nits:

>     * The first byte in file is byte number 1.

Blech. I'd rather it were 0. No biggie.

>  MULTIPLE URL PARAMETERS
>  
>   If at some point there will be multiple simultaneous URL parameters,
>   they should be separated by the ampersand character (just like
>   multiple values are encoded in the FORM request).

The ampersand character has odd interactions with SGML entity
reference syntax in HTML.

This URL:

	http://host/path;param1=val1&param2=val2

has to be written:

	<a href="http://host/path;param1=val1&#38;param2=val2">xxx</a>
	<a href="http://host/path;param1=val1&amp;param2=val2">xxx</a>

in HTML.

I suggest you separate parameters with ';' in stead:

	<a href="http://host/path;param1=val1;param2=val2">xxx</a>

Save everybody a little grief.


Dan

eturn-Path: cshotton@biap.com 
Return-Path: <cshotton@biap.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA16394; Wed, 17 May 1995 19:48:27 +0500
Received: from brain.biap.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA08797; Wed, 17 May 1995 19:46:11 +0500
Received: from [129.106.201.2] by brain.biap.com
 with SMTP (MailShare 1.0fc5); Wed, 17 May 1995 18:48:31 -0500
X-Sender: cshotton@198.64.246.129
Message-Id: <v0211010aabe0370e3425@[129.106.201.2]>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Wed, 17 May 1995 18:47:38 -0500
To: "Ronald E. Daniel" <rdaniel@acl.lanl.gov>,
        http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com, luotonen@netscape.com,
        www-talk@w3.org
From: cshotton@biap.com (Chuck Shotton)
Subject: Re:  Byte ranges -- formal spec proposal
Cc: uri@bunyip.com
Content-Length: 2817

At 4:59 PM 5/17/95, Ronald E. Daniel wrote:
>Byteranges are pretty nice since they are broadly applicable,
>but I am not sure what it means to ask for a byterange of
>a database. This problem is even more acute when we get into
>parameters such as "paragraph", "row/col", "stanza", etc.
>How are we to indicate when a parameter is inappropriate for
>a URL, such as paragraph for an image? Usually row/col
>will be inappropriate for HTML files, but if we have previously
>selected a table then it is the natural way to get a table
>element. How do we do that?

Worrying about stuff like this is WAY outside the bounds of asking a server
to send a particular range of bytes out of a data file. Remember, servers
typically are ignorant of the semantics of the data they serve. You are
running the discussion off into an area that is NOT a Good Thing (tm) for
server authors. Worrying about things like what a byte range means is
strictly for the client and the viewer app.

>If we do not develop a uniform architecture for fragment
>identification, we are going to have a slew of partial solutions before
>we wise up and develop a uniform treatment.

There are plenty of OO standards for defining the semantics of all the
different data examples you cite. However, they aren't really within the
scope of what servers (or WWW clients for that matter) need to worry about.
It really is an issue of a remote app wanting random access to the server's
file system and nothing more. Let the remote app worry about the semantics
of the data it is asking for and we'll all sleep a lot easier.

The Adobe people have done a LOT of work in this area already and as
complicated as Acrobat's document store is, they seem to be able to cope
with having the viewer/client ask for simple byte ranges. You have to make
an awfully good case for why simple byte range requests need to be
perverted into full-blown "fragment requests". There is a need for byte
range specs NOW. There is no apparent need for fragment requests,
especially since there is no common object model in place for
HTTP-delivered documents.

HOWEVER, things are going to be a LOT more complicated when new O/S
implementations start showing up with multi-fork files. The Mac O/S already
suffers under the Unix-driven limitations of HTTP because there is no
explicit support for the 2-part files in the Mac O/S. This will only get
worse because there are many new file systems in work that support the
concept of an unlimited number of data forks associated with a single
"file". How do byte range specifications work for this case?


-----------------------------------------------------------------------
Chuck Shotton
cshotton@biap.com                                  http://www.biap.com/
cshotton@oac.hsc.uth.tmc.edu                           "I am NOT here."


eturn-Path: cshotton@biap.com 
Return-Path: <cshotton@biap.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA16417; Wed, 17 May 1995 19:50:12 +0500
Received: from brain.biap.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA08823; Wed, 17 May 1995 19:47:56 +0500
Received: from [129.106.201.2] by brain.biap.com
 with SMTP (MailShare 1.0fc5); Wed, 17 May 1995 18:48:42 -0500
X-Sender: cshotton@198.64.246.129
Message-Id: <v0211010cabe039a4cfb0@[129.106.201.2]>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Wed, 17 May 1995 18:47:49 -0500
To: Ari Luotonen <luotonen@netscape.com>, www-talk@w3.org,
        http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
From: cshotton@biap.com (Chuck Shotton)
Subject: Re: Byte ranges -- formal spec proposal
Cc: uri@bunyip.com
Content-Length: 1639

>Description of the byterange URL Parameter
>
>     * The byte range request is attached to the end of the URL,
>       separated by a semi-colon.

Why not use the already supported "?" separator? Many file systems use the
semi-colon character to represent version numbers and overloading this
character could cause difficulties. In effect, you are asking the server to
"search" for a specific byte range in a document anyway, so it's not too
big of a stretch to adopt the "?" safe character instead of risking
possible conflicts with ";".

>     * The first byte in file is byte number 1.

This is good.


>     * The range includes both limits, that is, a range 500-1000 means
>       the 501 bytes between 499 and 1001.

This isn't clearly worded.

>     * In place of a range there may be only a single positive integer.
>       This means only a single byte. This is not meaningful for byte
>       ranges, but if and when this scheme gets extended to allow other
>       kinds of ranges (such as lines, chapters or pages), this will be
>       useful.

Please. Let's not expand on this. It forces servers to have a much more
intimate knowledge of the content they serve than is necessary. Let's
define how byte ranges work and leave the nasty WWW object model for
another day and another syntax. If you want a single byte, specify a range,
n-n.

Leave the content semantics between CGIs and Viewers.

-----------------------------------------------------------------------
Chuck Shotton
cshotton@biap.com                                  http://www.biap.com/
cshotton@oac.hsc.uth.tmc.edu                           "I am NOT here."


eturn-Path: agleckle@us.oracle.com 
Return-Path: <agleckle@us.oracle.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA17210; Wed, 17 May 1995 20:11:57 +0500
Received: from inet-smtp-gw-1.us.oracle.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA09131; Wed, 17 May 1995 20:09:30 +0500
Received:  from monday.us.oracle.com by inet-smtp-gw-1.us.oracle.com with SMTP (8.6.12/37.7)
	id RAA04110; Wed, 17 May 1995 17:09:23 -0700
Received:  by monday.us.oracle.com (5.59.11/37.7)
	id AA18968; Wed, 17 May 95 17:09:23 PDT
Message-Id: <9505180009.AA18968@monday.us.oracle.com>
Date: Wed, 17 May 95 17:09:23 PDT
From: Arthur Gleckler <agleckle@us.oracle.com>
To: www-talk@w3.org
In-Reply-To: <v0211010aabe0370e3425@[129.106.201.2]> (cshotton@biap.com)
Subject: Re:  Byte ranges -- formal spec proposal
Reply-To: agleckle@us.oracle.com
Content-Length: 194

Could someone please give a quick justification for adding the ability
to retrieve just parts of documents from web servers?  I didn't see
such an explanation in the original proposal.  Thanks!
eturn-Path: john@math.nwu.edu 
Return-Path: <john@math.nwu.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA17391; Wed, 17 May 1995 20:15:08 +0500
Received: from hopf.math.nwu.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA09183; Wed, 17 May 1995 20:12:46 +0500
Received: (from john@localhost) by hopf.math.nwu.edu (8.6.10/8.6.9) id TAA06661; Wed, 17 May 1995 19:12:40 -0500
From: John Franks <john@math.nwu.edu>
Message-Id: <199505180012.TAA06661@hopf.math.nwu.edu>
Subject: Re: Byte ranges -- formal spec proposal
To: connolly@beach.w3.org (Daniel W. Connolly)
Date: Wed, 17 May 1995 19:12:39 -0500 (CDT)
Cc: luotonen@netscape.com, www-talk@w3.org,
        http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com, uri@bunyip.com
In-Reply-To: <199505172339.TAA10475@beach.w3.org> from "Daniel W. Connolly" at May 17, 95 07:39:35 pm
X-Mailer: ELM [version 2.4 PL23]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1360      

According to Daniel W. Connolly:
> 
> A nice, clear, complete proposal. As you say, this could be done as a
> server-private mechanism, but there's no reason why everybody
> shouldn't do it the same way.
> 
> A couple nits:
> 
> >     * The first byte in file is byte number 1.
> 
> Blech. I'd rather it were 0. No biggie.
> 

Base 0 is fine for bytes but would be problematic for other ranges.
E.g. 

	http://host/book;chapterrange=3-5

would mean chapters 4 to 6 if base 0 is used.  This would be just too
confusing.  We thought it better to be consistent and use the same
base for everything.

> >  MULTIPLE URL PARAMETERS
> >  
> >   If at some point there will be multiple simultaneous URL parameters,
> >   they should be separated by the ampersand character (just like
> >   multiple values are encoded in the FORM request).
> 
> The ampersand character has odd interactions with SGML entity
> reference syntax in HTML.
> 
> This URL:
> 
> 	http://host/path;param1=val1&param2=val2
> 
> has to be written:
> 
> 	<a href="http://host/path;param1=val1&#38;param2=val2">xxx</a>
> 	<a href="http://host/path;param1=val1&amp;param2=val2">xxx</a>
> 
> in HTML.
> 
> I suggest you separate parameters with ';' in stead:
> 
> 	<a href="http://host/path;param1=val1;param2=val2">xxx</a>
> 
> Save everybody a little grief.
> 

This is a good point.

John Franks
eturn-Path: jag@scndprsn.eng.sun.com 
Return-Path: <jag@scndprsn.eng.sun.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB19195; Wed, 17 May 1995 20:50:39 +0500
Received: from Sun.COM (koriel.Sun.COM) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA09575; Wed, 17 May 1995 20:48:21 +0500
Received: from Eng.Sun.COM (engmail2.Eng.Sun.COM) by Sun.COM (koriel.Sun.COM)
	id AA00809; Wed, 17 May 95 17:47:54 PDT
Received: from scndprsn.Eng.Sun.COM by Eng.Sun.COM (5.x/SMI-5.3)
	id AA03445; Wed, 17 May 1995 17:47:49 -0700
Received: from norquay.Eng.Sun.COM by scndprsn.Eng.Sun.COM (5.0/SMI-SVR4)
	id AA01959; Wed, 17 May 1995 17:47:59 +0800
Received: by norquay.Eng.Sun.COM (5.0/SMI-SVR4)
	id AA28623; Wed, 17 May 1995 17:50:00 +0800
Date: Wed, 17 May 1995 17:50:00 +0800
From: jag@scndprsn.eng.sun.com (James Gosling)
Message-Id: <9505180050.AA28623@norquay.Eng.Sun.COM>
To: connolly@beach.w3.org, john@math.nwu.edu
Subject: Re: Byte ranges -- formal spec proposal
Cc: luotonen@netscape.com, www-talk@w3.org,
        http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com, uri@bunyip.com
X-Sun-Charset: US-ASCII
Content-Length: 1399

> According to Daniel W. Connolly:
> > 
> > A nice, clear, complete proposal. As you say, this could be done as a
> > server-private mechanism, but there's no reason why everybody
> > shouldn't do it the same way.
> > 
> > A couple nits:
> > 
> > >     * The first byte in file is byte number 1.
> > 
> > Blech. I'd rather it were 0. No biggie.
> > 
> 
> Base 0 is fine for bytes but would be problematic for other ranges.
> E.g. 
> 
> 	http://host/book;chapterrange=3-5
> 
> would mean chapters 4 to 6 if base 0 is used.  This would be just too
> confusing.  We thought it better to be consistent and use the same
> base for everything.

I don't think this is relevant: http should be kept simple and data-type
independent, leave out the higher level semantics.  Then 0 based
addressing is the most sensible.  Even for chapters, the argument is
weak: what chapter number is the title page?  What chapter number
applies to appendicies?  Does the number then need to be a string that
names a sub-entity?  This is a Pandora's Box that should stay closed.

Any thought on how this should interect with dynamic computed documents
(CGI-bin scripts)?  Supporting range addressing of computed documents
would require either re-computation on each fetch, or caching.  If
re-computed, how do you guarantee consistancy?  Imagine fetching a
document one byte at a time that contains the server's load average.
eturn-Path: jak@violet.berkeley.edu 
Return-Path: <jak@violet.berkeley.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA19934; Wed, 17 May 1995 21:05:29 +0500
Received: from violet.berkeley.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA09734; Wed, 17 May 1995 21:03:10 +0500
Received: by violet.berkeley.edu (8.6.10/1.33r)
	id SAA14003; Wed, 17 May 1995 18:02:46 -0700
Date: Wed, 17 May 1995 18:02:46 -0700
From: jak@violet.berkeley.edu (John A. Kunze)
Message-Id: <199505180102.SAA14003@violet.berkeley.edu>
To: http-wg@cuckoo.hpl.hp.com, luotonen@netscape.com, www-talk@w3.org
Subject: Re: Byte ranges -- formal spec proposal
Cc: uri@bunyip.com
Content-Length: 1705

> From: Ari Luotonen <luotonen@netscape.com>
> Date: Wed, 17 May 1995 13:22:40 -0700 (PDT)
> ...

The proposal in general sounds pretty reasonable to me.  It's not unlike
fragment proposals submitted for other applications (eg, Z39.50), but
I think the timing and especially the simplicity of this one favors it.

>    The following HTTP response header is sent back to provide
>    verification and information about the range and total size of the file:
> 
>         Range: bytes X-Y/Z

Without getting much more complicated, you might want to handle the case
when the server doesn't know the size, as when a file is growing.

Multiple ranges make me a little squeamish because you have to think about
packaging ranges in the (single) response.  For example, do you concatenate
all the returned ranges, and let the client count bytes to figure out where
range boundaries are?  (Things get weird fast when you move beyond bytes
into, say, paragraphs.)

If you stick with multiple ranges you might let the server be pig-headed
about them and return them in the order that's most convenient.  Good taste
dictates that it will do what's requested, but offers protection from some
pathological cases.  Reliability can be assured if the header reports back
to the client what the server ended up doing.

-John

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
John A. Kunze                          510-642-1530
Information Systems and Technology     Fax:  510-643-5385
293 Evans Hall                         Internet:  jak@violet.berkeley.edu
Berkeley, CA  94720                    or jak@nlm.nih.gov
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=
eturn-Path: john@math.nwu.edu 
Return-Path: <john@math.nwu.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA20129; Wed, 17 May 1995 21:14:14 +0500
Received: from hopf.math.nwu.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA09796; Wed, 17 May 1995 21:11:49 +0500
Received: (from john@localhost) by hopf.math.nwu.edu (8.6.10/8.6.9) id UAA06793; Wed, 17 May 1995 20:11:45 -0500
From: John Franks <john@math.nwu.edu>
Message-Id: <199505180111.UAA06793@hopf.math.nwu.edu>
Subject: Re: Byte ranges -- formal spec proposal
To: cshotton@biap.com (Chuck Shotton)
Date: Wed, 17 May 1995 20:11:44 -0500 (CDT)
Cc: luotonen@netscape.com, www-talk@w3.org,
        http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com, uri@bunyip.com
In-Reply-To: <v0211010cabe039a4cfb0@[129.106.201.2]> from "Chuck Shotton" at May 17, 95 06:47:49 pm
X-Mailer: ELM [version 2.4 PL23]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1496      

According to Chuck Shotton:
> 
> >Description of the byterange URL Parameter
> >
> >     * The byte range request is attached to the end of the URL,
> >       separated by a semi-colon.
> 
> Why not use the already supported "?" separator? Many file systems use the
> semi-colon character to represent version numbers and overloading this
> character could cause difficulties. In effect, you are asking the server to
> "search" for a specific byte range in a document anyway, so it's not too
> big of a stretch to adopt the "?" safe character instead of risking
> possible conflicts with ";".
> 

I see a potential problem with using '?' because it is already used
for forms.  What happens if someone has a form with a field named
"byterange"?  We are better off having the name space for form fields
separate from that of URL parameters.

> 
> Please. Let's not expand on this. It forces servers to have a much more
> intimate knowledge of the content they serve than is necessary. Let's
> define how byte ranges work and leave the nasty WWW object model for
> another day and another syntax.

I generally agree with this, though I might not have put it so strongly.
On the other hand it is fine for servers to implement server specific 
parameters with special meaning for that server.  There are a number of
such in the server that I wrote.  If, as in the case of byterange, it
turns out that several implementors want the same functionality it makes
sense to try to standardize.

John Franks
eturn-Path: masinter@parc.xerox.com 
Return-Path: <masinter@parc.xerox.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA27623; Thu, 18 May 1995 00:06:05 +0500
Received: from alpha.xerox.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA12793; Thu, 18 May 1995 00:03:50 +0500
Received: from gordo.parc.xerox.com ([13.1.100.192]) by alpha.xerox.com with SMTP id <14576(5)>; Wed, 17 May 1995 21:03:38 PDT
Received: by gordo.parc.xerox.com id <139716>; Wed, 17 May 1995 21:03:32 -0700
To: luotonen@netscape.com
Cc: www-talk@w3.org
In-Reply-To: Ari Luotonen's message of Wed, 17 May 1995 15:34:11 -0700 <199505172234.PAA20589@neon.netscape.com>
Subject: Re: Byte ranges -- formal spec proposal
From: Larry Masinter <masinter@parc.xerox.com>
Sender: Larry Masinter <masinter@parc.xerox.com>
Fake-Sender: masinter@parc.xerox.com
Message-Id: <95May17.210332pdt.139716@gordo.parc.xerox.com>
Date: Wed, 17 May 1995 21:03:19 PDT
Content-Length: 1807

Ari's proposal said:

>   There are number of Web applications that would benefit from being
>   able to request the server to give a byte range of a document.

and I asked

> Name two.

and Ari replied:

> 1. A custom application on the client side that works on some
> page-oriented document format, such as FrameMaker docs or PDF.

but I still ask:

Please give a single, specific and realistic example. If you think a
framemaker document can be retrieved a byte range at a time, what is
the initial byte range? How do you know?

Can PDF documents be retrieved partially? A priori? How many bytes
would you know to ask for in the first place? Is 'the first part of a
PDF file' still 'application/pdf'?


> 2. Just regular Web clients where image or document transfer is
> interrupted before the entire image/document is received, and then
> later restarted.  So instead of starting all over again you could only
> transfer the remaining part of it.

This is http specific, isn't it? Surely a FTP server won't support
this. So, does this belong in the URL at all? 

I think these are both scenarios that might benefit from a GET-PARTIAL
or some kind of modification to GET in HTTP, but I don't see either of
these scenarios being justification for a new feature in URLs.

telnet://host;byterange=1-5?

The discussion seems to have confounded fragment identifiers (where
the URL points to a part of another object, but the request is 'get
the whole object, but direct attention to this part') with some kind
of 'partial object retrieval' (retrieve this other resource which is a
particular subpiece of the original one.)

Fragment identifiers occur after #, but the ';' separator is reserved
in a variety of URL schemes. Unless you intend this to only work with
HTTP? That isn't what the proposal said.



eturn-Path: roscheis@cs.stanford.edu 
Return-Path: <roscheis@cs.stanford.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA28772; Thu, 18 May 1995 00:40:32 +0500
Received: from Xenon.Stanford.EDU by www10.w3.org (5.0/NSCS-1.0S) 
	id AA13015; Thu, 18 May 1995 00:38:17 +0500
Received:  by Xenon.Stanford.EDU (5.61+IDA/25-Xenon-eef) id AA06352; Wed, 17 May 95 21:38:01 -0700
Message-Id: <9505180438.AA06352@Xenon.Stanford.EDU>
To: www-talk@w3.org
Cc: mogens@cs.stanford.edu, winograd@cs.stanford.edu
Subject: FYI: Shared WWW Annotations: Code Available to Developers
From: rmr@cs.stanford.edu (Martin Roscheisen)
Date: Wed, 17 May 1995 21:37:58 -0700
Sender: roscheis@cs.stanford.edu
Content-Length: 2640


We are making available to WWW developers the experimental prototype
implementation which we developed as a vehicle for our research into a
scalable architecture for a generalized form of "group annotations".

Usages which we have found this useful for include structured
discussion about paper drafts, collaborative filtering, seals of
approval/content rating, all sorts of guided tours, shared hotlists
with section-based visibility control, usage indicators, co-presence,
and "Vannevar Bush type" value-added trails.

The code is a couple of months of work by two people.  It has been used 
mainly on RS6000s, but should also work on other machines.

Server Code: a system of cgi-bin scripts implemented in perl.

Browser Code: a derivative product of NCSA xMosaic. Extensions include

 - facilities for shared in-place annotations to arbitrary HTML document
   (including dynamically allocated extended/brief menus),
 - a META language module which parses meta information and deals with
   appropriate callback routines.
 - a user profile server which allows users to store their personal 
   configuration (including personal preferences and the hotlist) at a
   place where they can always reach it. (This "base station" is an 
   extended http server.) 
 - a light-weight viewing facility for looking at meta information 
   including annotations (a PostIt type popup window activated by 
   pressing the middle button), 
 - navigational aids (callbacks to a tour/map program, another HTML 
   widget in main layout for title/link display, augmented HTML parser 
   which recognizes the LINK tag), as well as 
 - a facility for printing collections of linked documents in Postscript. 
 - various fixes to and variations on Mosaic.

Please note that this "release" is not intended for general users; it
is targeted at letting other developers get a more tangible experience
beyond the papers which we have published.  Also, since the primary
purpose of the implementation was to explore new ideas and concepts,
we do not maintain the code and also do not give any guarantees.

For a description of the system, see the paper

Martin Roscheisen, Christian Mogensen, and Terry Winograd (1994). A
  Platform for Third-Party Value-Added Providers: Architecture,
  Protocols, and Usage Examples.  Technical Report, Computer Science
  Department, Stanford University. 
Available at URL http://www-diglib.stanford.edu/COMMENTOR/.

The code is also available from a link at this page, or directly at
ftp://www-pcd.stanford.edu/pub/pcd/brio/.

Cheers, 

Martin Roscheisen / rmr@cs.stanford.edu 
Christian Mogensen / mogens@cs.stanford.edu 


eturn-Path: masinter@parc.xerox.com 
Return-Path: <masinter@parc.xerox.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA29449; Thu, 18 May 1995 00:55:19 +0500
Received: from alpha.xerox.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA13130; Thu, 18 May 1995 00:52:51 +0500
Received: from golden.parc.xerox.com ([13.1.100.139]) by alpha.xerox.com with SMTP id <14507(1)>; Wed, 17 May 1995 21:52:39 PDT
Received: by golden.parc.xerox.com id <2761>; Wed, 17 May 1995 21:52:31 -0700
Cc: http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com, www-talk@w3.org, uri@bunyip.com
To: http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com, www-talk@w3.org, uri@bunyip.com
Subject: Re: Byte ranges -- formal spec proposal
From: Larry Masinter <masinter@parc.xerox.com>
Sender: Larry Masinter <masinter@parc.xerox.com>
Fake-Sender: masinter@parc.xerox.com
Reply-To: www-talk@w3.org
Message-Id: <95May17.215231pdt.2761@golden.parc.xerox.com>
Date: Wed, 17 May 1995 21:52:19 PDT
Content-Length: 298

I'm getting this discussion 3 times, http-wg, www-talk, and now on
uri.  I suggest keeping the discussion on www-talk for now.

The proposal is to add byte ranges to URLs (in general, it seems). I
don't think it belongs there; at best, byte ranges make sense as an
addon to the HTTP protocol.





eturn-Path: klute@heike.nads.de 
Return-Path: <klute@heike.nads.de>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA06750; Thu, 18 May 1995 02:37:35 +0500
Received: from heike.nads.de by www10.w3.org (5.0/NSCS-1.0S) 
	id AA14593; Thu, 18 May 1995 02:37:13 +0500
Received: from heike (localhost [127.0.0.1]) by heike.nads.de (8.6.12/8.6.12) with SMTP id IAA07823; Thu, 18 May 1995 08:36:06 +0200
Message-Id: <199505180636.IAA07823@heike.nads.de>
To: luotonen@netscape.com
Cc: Multiple recipients of list <www-talk@www10.w3.org>, klute@heike.nads.de
Subject: Re: Byte ranges -- formal spec proposal 
Reply-To: Rainer Klute <klute@nads.de>
Organization: NADS GmbH, Germany
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
In-Reply-To: Your message of Wed, 17 May 1995 18:16:12 +0500.
             <199505172022.NAA06472@neon.netscape.com> 
Date: Thu, 18 May 1995 08:36:06 +0200
From: Rainer Klute <klute@nads.de>
Content-Length: 1855

>                        BYTE RANGES WITH URLS AND HTTP
>                                       
>   May 16, 1995

A very good proposal and it's nice to have it published and
discussed before it is implemented! This is the way Netscape Comm.
should always take.

Some minor points follow:


>Description of the byterange URL Parameter
>
>     * The byte range request is attached to the end of the URL,
>       separated by a semi-colon.

What is the "end of an URL"? In

	<http://foo/bar?blah>

where is the "end"? Is it before the question mark or behind
"blah". Would a byterange URL be

	<http://foo/bar;byterange=4-44?blah>

or
	<http://foo/bar?blah;byterange=4-44?blah>

I'd prefer the former one because it integrates much more smothly
with existing applications.


>     * The first byte in file is byte number 1.

What is a file? There's not necessarily a file behind a URL, and to
the client this should be totally transparent. Let's call it
document or (even more general) entity.

As other responses indicate, it is more or less a matter of
taste whether to start with 0 or 1. Just pick one (not necessarily
1 :-)) and go for it.


>   If at some point there will be multiple simultaneous URL parameters,
>   they should be separated by the ampersand character (just like
>   multiple values are encoded in the FORM request).

Oh no, please don't pick the ampersand because of its SGMLish
implications!


BTW, some food for more thinking: It may be useful to invent some
notation for set operations, like
<http://foo/bar;(chapterrange=5-12)-((chapterrange=8)+byterange=100-1500))>


Best regards
Rainer Klute

  Dipl.-Inform. Rainer Klute        NADS - Advertising on nets
  NADS GmbH
  Emil-Figge-Str. 80                Tel.: +49 231 9742570
D-44227 Dortmund                    Fax:  +49 231 9742571

            <http://www.nads.de/~klute/>
eturn-Path: brian@organic.com 
Return-Path: <brian@organic.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA08169; Thu, 18 May 1995 03:01:22 +0500
Received: from eat.organic.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA14989; Thu, 18 May 1995 03:01:09 +0500
Received: (from brian@localhost) by eat.organic.com (8.6.11/8.6.9) id AAA07567; Thu, 18 May 1995 00:01:15 -0700
Date: Thu, 18 May 1995 00:01:15 -0700 (PDT)
From: Brian Behlendorf <brian@organic.com>
Subject: Re: Byte ranges -- formal spec proposal
To: www-talk@w3.org
In-Reply-To: <95May17.215231pdt.2761@golden.parc.xerox.com>
Message-Id: <Pine.3.89.9505172203.h15253-0100000@eat.organic.com>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 6243

On Wed, 17 May 1995, Larry Masinter wrote:
> I'm getting this discussion 3 times, http-wg, www-talk, and now on
> uri.  I suggest keeping the discussion on www-talk for now.

Done.  Are John Franks and Ari aboard?  

> The proposal is to add byte ranges to URLs (in general, it seems). I
> don't think it belongs there; at best, byte ranges make sense as an
> addon to the HTTP protocol.

Then how does one build a URL to point to minutes 6 through 8 of a 1-hour
60-megabyte DJ set?  Or to message 2064 (byte range 2254322-2257934) in a
4 megabyte mailbox archive?  Sure, if I'm using an HTTP-aware mailbox
reader or audio viewer that's a possibility... but then I can only launch
a range request from that type of viewer.  Ick. 

I do see what you're getting at, though.  There is *not* necessarily a 
direct mapping between a URL and the representation of the object that 
URL refers to as returned by the server.  Key definitions (and correct me 
if I'm wrong, but at least this is how I think most developers are 
conceiving the web):

1) A URL is a pointer to an *object* somewhere. (well, okay, object=resource)

2) An object can be anything - only *representations* of an object are 
directly viewable (flat files or program output, for example, are 
both representations).  

3) A *representation* is what we get back when we perform an action on a URL
(GET, POST, etc) in the context of an HTTP request. You do not "GET" the 
object itself.

4) *Representations* can be influenced by any part of the request, not 
just the URI.  For example, headers which are known to make a difference: 
Accept (content negotiation), If-Modified-Since, WWW-Authorization, hell 
even User-Agent in certain situations.

(quick question - isn't it inherently more scalable to distribute and 
cache the objects themselves rather than their (possibly 
numerous) representations?  Hmm..)

So, does a "byte range" constitute a variation of the object, or a new
object itself, which deserves a unique URL?  Compelling cases could be
made on either side, but I think in this situation it truely is a
variation of the object.  But now we have a problem - the WWW Link Model
(hi roy!) only lets me link to *objects* (i.e., URL's), not particular
variations/representations of objects, if I understand things correctly. 
For example, if I have an object that represents my home page, and my home
page object returns both HTML 2.0 and HTML 3.0 representations of itself,
there's no way for me to *force* an HTML 2.0 browser to see the HTML 3.0
representation without giving the HTML 3.0 representation its own, 
un-content-negotiated URL.  Feh. 

Okay, so here's the problem.  A URL must be able, not required, but able,
to *completely* describe the request for an object.  In other words, URL's
must be able to point to particular representations of webbable objects. 
The protocol "method" used.  The additional headers.  In fact, in most
situations today URL's are used to point to representations instead of
objects - content providers are simply creating unique URL's to every
representation.  So, we're not breaking anything fundamental here, it
seems.  Further more: 

1) There must be a clear distinction between the part of the URL that 
describes the *object*, and the part of the URL that describes its 
representation.

2) User-agents must be able to deal with the part of the URL that 
describes the representation at a higher level - for example, when a user 
goes to "bookmark" the object, they are asked to chose whether they want 
to bookmark the object in general or the particular representation of 
that object.  

3) Responses need to indicate which parts of that representation request 
influenced the output, so that caches know what to key on (and don't 
needlessly key on everything in the request.)  I think there's a "vary" 
header proposed somewhere....

4) There must be a defined list of "sanctimonious" headers in HTTP, ones 
which are always part of the request and are *not* modifiable by the 
representation-part of the URL.  For example, User-Agent:, or From:.  
Likewise, content providers should not vary content based on these headers.

Phew.

(btw, the CD I'm listening to now seems highly conducive to these kind of 
thought processes - Air, by Pete Namlook, on FAX)

So, here's how I think things should look.  The format:

  http://host/path/to/object?object_arguments;request_headers

  object_arguments: a url-encoded list of name-value pairs 
	i.e. name=brian&age=22

  request_headers: a url-encoded list of request headers, which only
        make sense in the context of the protocol used (in this case HTTP)
	This generality is so that URL's aren't hindered by HTTP-only
	specifications.

So that the browser's request looks something like

  (connect to host port 80)
  GET /path/to/object?object_arguments HTTP/1.0
  User-Agent: Godzilla
  request_header.name1=request_header.value1
  request_header.name2=request_header.value2

For the purposes of this exposition, the HTTP header referring to 
byteranges would be something like "ByteRange:".  Something more general 
is needed for other segments of course.

Some sample URL's:

   a pointer to a sound file of clinton's weekly radio address:
	http://www.npr.org/clinton/week23
   a pointer to an MPG version of clinton's weekly radio address:
        http://www.npr.org/clinton/week23;Accept=audio/x-mpeg
   a pointer to byte range 10234234-13244212 of clinton's weekly radio address:
	http://www.npr.org/clinton/week23;Accept=audio/x-mpeg&Byterange=10234234-13244212
   
I can already sense some problems.  Here's an interesting URL:

http://whitehouse.gov:25/;MAIL+FROM=madmad@bomber.org&RCPT+TOpresident&DATA\nFrontLawn,2pm,May16th\n.\n

Though I suppose some catches could be put in place for this situation, 
can we protect against that for every protocol?  At what point does a 
sufficiently obfuscated (to the human eye) extended URL become a malicious
virus-ish mechanism for mayhem?

Food for thought, hopefully I'm not too far off base on some of these.
Dan, Roy, let me have it.  :)

	Brian

--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--
brian@organic.com  brian@hyperreal.com  http://www.[hyperreal,organic].com/

eturn-Path: Carl.von.Loesch@arbi.informatik.uni-oldenburg.de 
Return-Path: <Carl.von.Loesch@arbi.informatik.uni-oldenburg.de>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AC14240; Thu, 18 May 1995 04:47:54 +0500
Received: from arbi.informatik.uni-oldenburg.de by www10.w3.org (5.0/NSCS-1.0S) 
	id AA16332; Thu, 18 May 1995 04:46:21 +0500
Received: by arbi.informatik.uni-oldenburg.de (smail3.1.18 + xalias);
	Thu, 18 May 95 10:45 CES
Received: by diamant.Informatik.Uni-Oldenburg.DE (Smail-3.1.28.1)
	id <m0sC123-0003CyC@diamant.Informatik.Uni-Oldenburg.DE>;
	Thu, 18 May 95 10:34 MET DST
Message-Id: <m0sC11z-000AgFC@beryll.Informatik.Uni-Oldenburg.DE>
Subject: Re: Byte ranges -- formal spec proposal
To: luotonen@netscape.com
Date: Thu, 18 May 1995 10:34:05 +0200 (MET DST)
From: Carl von Loesch <Carl.von.Loesch@arbi.informatik.uni-oldenburg.de>
Cc: www-talk@www10.w3.org
In-Reply-To: <199505172022.NAA06472@neon.netscape.com> from "Ari Luotonen" at May 17, 95 06:15:17 pm
X-Mailer: ELM [version 2.4 PL22]
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
Content-Length: 732       

I like this very much as it permits implementation of something like
FTP's REGET .. possibility to resume on broken TCP links...

Ari Luotonen typeth:
|         http://host/dir/foo;byterange=1-500
|         http://host/dir/foo;linerange=21-30
|         http://host/dir/foo;chapterrange=1-3

Something not very important, but I'd like to have something cozier
to type, so I suggest this:

     http://host/dir/foo;bytes=1-500
     http://host/dir/foo;lines=21-30
     http://host/dir/foo;chapters=1-3

Think it looks a little less technical...

KUTGW.
-- 
     _______					  http://home.pages.de/~lynx/
irc: SymLynX		Carl v. Loesch			  lynx@net.pages.de

LynX@You.might.aswell.use.This.as.my.Email.Addres�
eturn-Path: vania@gr.osf.org 
Return-Path: <vania@gr.osf.org>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB14913; Thu, 18 May 1995 04:55:47 +0500
Received: from postman.osf.org by www10.w3.org (5.0/NSCS-1.0S) 
	id AA16404; Thu, 18 May 1995 04:55:37 +0500
Received: from danton.gr.osf.org (danton.gr.osf.org [130.105.64.11]) by postman.osf.org (8.6.9/8.6.x) with SMTP
	id EAA20939; Thu, 18 May 1995 04:55:34 -0400
Message-Id: <199505180855.EAA20939@postman.osf.org>
Received: from alba.gr.osf.org by danton.gr.osf.org with SMTP
	(15.11/15.6) id AA15765; Thu, 18 May 95 11:02:33 met
To: jag@scndprsn.eng.sun.com
Cc: www-talk@www10.w3.org
Subject: Re: Byte ranges -- formal spec proposal 
In-Reply-To: Your message of "Thu, 18 May 95 03:20:39 +0500."
             <9505180050.AA28623@norquay.Eng.Sun.COM> 
Date: Thu, 18 May 95 10:55:41 +0200
From: "Vania Joloboff" <vania@gr.osf.org>
Content-Length: 920


> Any thought on how this should interect with dynamic computed documents
> (CGI-bin scripts)?  Supporting range addressing of computed documents
> would require either re-computation on each fetch, or caching.  If
> re-computed, how do you guarantee consistancy?  Imagine fetching a
> document one byte at a time that contains the server's load average.

Are you suggesting that instead of an addressing schema,
one should focus instead on a naming schema for parts,
the mapping of which would be left to the server ?


That is, servers could provide part identifiers
instead of part addresses, and client could query
those identifiers.


	Vania

========================================================================
Vania Joloboff                    Phone: (33) 76 63 48 92
OSF                               Fax:   (33) 76 51 05 32 
2, Avenue de Vignate
38610 Gieres - France             e-mail: vania@gr.osf.org	
eturn-Path: gtn@ebt.com 
Return-Path: <gtn@ebt.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA29630; Thu, 18 May 1995 08:42:48 +0500
Received: from ebt-inc.ebt.com (kirk.ebt.com) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA19051; Thu, 18 May 1995 08:42:37 +0500
Received: (from gtn@localhost) by ebt-inc.ebt.com (8.6.9/8.6.9) id IAA05039; Thu, 18 May 1995 08:44:55 -0400
Date: Thu, 18 May 1995 08:44:55 -0400
From: Gavin Nicol <gtn@ebt.com>
Message-Id: <199505181244.IAA05039@ebt-inc.ebt.com>
To: luotonen@netscape.com
Cc: www-talk@w3.org, http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
In-Reply-To: <199505172022.NAA06472@neon.netscape.com> (message from Ari Luotonen on Wed, 17 May 1995 13:22:40 -0700 (PDT))
Subject: Re: Byte ranges -- formal spec proposal
Content-Length: 564

I do not like the idea of byte ranges being part of a URL.
In DynaWeb for example

   http://www.ebt.com/collection/readers?byterange=1-500

is absolutely meaningless, because the actual file size is not
related to the size of the documents retrieved. Worse even,
there are cases (and DynaWeb is one) where a single URL
can possibly reference documents that differ in an
arbitrary manner.

In other words, byte ranges are not generally applicable to
all objects accessible via a URL, so we cannot make this
a requirement.

How about adding another method instead?
eturn-Path: cshotton@biap.com 
Return-Path: <cshotton@biap.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA01604; Thu, 18 May 1995 09:08:34 +0500
Received: from brain.biap.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA19489; Thu, 18 May 1995 09:08:20 +0500
Received: from [129.106.201.2] by brain.biap.com
 with SMTP (MailShare 1.0fc5); Thu, 18 May 1995 08:10:53 -0500
X-Sender: cshotton@198.64.246.129
Message-Id: <v02110124abe0f71451c6@[129.106.201.2]>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Thu, 18 May 1995 08:10:01 -0500
To: Gavin Nicol <gtn@ebt.com>, luotonen@netscape.com
From: cshotton@biap.com (Chuck Shotton)
Subject: Re: Byte ranges -- formal spec proposal
Cc: www-talk@w3.org, http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
Content-Length: 1045

At 8:44 AM 5/18/95, Gavin Nicol wrote:
>I do not like the idea of byte ranges being part of a URL.
>In DynaWeb for example
>
>   http://www.ebt.com/collection/readers?byterange=1-500
>
>is absolutely meaningless, because the actual file size is not
>related to the size of the documents retrieved. Worse even,
>there are cases (and DynaWeb is one) where a single URL
>can possibly reference documents that differ in an
>arbitrary manner.
>
>In other words, byte ranges are not generally applicable to
>all objects accessible via a URL, so we cannot make this
>a requirement.

I guess I don't see a problem here. If byte ranges don't apply to data
stored in your server, why would you ever have to worry about URLs
requesting them? Why should your server ever receive them, and why
shouldn't it just ignore them?

-----------------------------------------------------------------------
Chuck Shotton
cshotton@biap.com                                  http://www.biap.com/
cshotton@oac.hsc.uth.tmc.edu                           "I am NOT here."


eturn-Path: john@math.nwu.edu 
Return-Path: <john@math.nwu.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA06022; Thu, 18 May 1995 10:06:37 +0500
Received: from hopf.math.nwu.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA20250; Thu, 18 May 1995 10:06:27 +0500
Received: (from john@localhost) by hopf.math.nwu.edu (8.6.10/8.6.9) id JAA07624; Thu, 18 May 1995 09:01:29 -0500
From: John Franks <john@math.nwu.edu>
Message-Id: <199505181401.JAA07624@hopf.math.nwu.edu>
Subject: Re: Byte ranges -- formal spec proposal
To: gtn@ebt.com (Gavin Nicol)
Date: Thu, 18 May 1995 09:01:29 -0500 (CDT)
Cc: luotonen@netscape.com, www-talk@w3.org,
        http-wg%cuckoo.hpl.hp.com@hplb.hpl.hp.com
In-Reply-To: <199505181244.IAA05039@ebt-inc.ebt.com> from "Gavin Nicol" at May 18, 95 08:44:55 am
X-Mailer: ELM [version 2.4 PL23]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1843      

According to Gavin Nicol:
> 
> I do not like the idea of byte ranges being part of a URL.
> In DynaWeb for example
> 
>    http://www.ebt.com/collection/readers?byterange=1-500
> 
> is absolutely meaningless, because the actual file size is not
> related to the size of the documents retrieved. Worse even,
> there are cases (and DynaWeb is one) where a single URL
> can possibly reference documents that differ in an
> arbitrary manner.
> 
> In other words, byte ranges are not generally applicable to
> all objects accessible via a URL, so we cannot make this
> a requirement.
> 

First, let me say that as far as I am concerned we are discussing
byte ranges for HTTP URLs, not for every URL.  I assumed this was
understood.  I, at least, am not suggesting changes to ftp or gopher
URLs.

Secondly, a couple of people have expressed concern over "making this
a requirement."  I am not sure what this even means.  No server is
ever "required" to honor any request.  Requests are routinely denied
and there are HTTP status codes to communicate information about the
denial.   The fact that a URL

	http://host/path/foo

is honored in no way implies that the *different* URL 

	http://host/path/foo;byterange=1-500

is required also to be honored.  It is quite acceptable to return an
"access denied" status for the second URL.  Parameters like byterange
while very useful in some applications are useless, meaningless, hard
to serve, or even impossible to serve in other cases.  We can all come
up with our favorite example.  When a byterange is requested in such
an instance the server should simply refuse the request with an
appropriate status code.  If a server does not support byteranges
(e.g. most current servers) and receives a request for one it will
presumably send a "file does not exist" status code.  This is fine.

John Franks

eturn-Path: mvanheyn@pellet.spry.com 
Return-Path: <mvanheyn@pellet.spry.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AE11814; Thu, 18 May 1995 11:38:09 +0500
Received: from homer.spry.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA21419; Thu, 18 May 1995 11:35:26 +0500
Received: from pellet.spry.com (pellet [198.185.1.189]) by homer.spry.com (8.6.9/8.6.9) with ESMTP id IAA16210; Thu, 18 May 1995 08:35:50 -0700
X-Mailer: exmh version 1.5.3 12/28/94
From: marcvh@spry.com (Marc VanHeyningen)
To: brian@organic.com
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Re: Byte ranges -- formal spec proposal 
In-Reply-To: Your message of "Thu, 18 May 1995 06:30:02 +0500."
             <Pine.3.89.9505172203.h15253-0100000@eat.organic.com> 
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Thu, 18 May 1995 08:35:02 -0700
Message-Id: <14274.800811302@pellet.spry.com>
Sender: mvanheyn@pellet.spry.com
Content-Length: 2156

Brian Behlendorf sed:
> On Wed, 17 May 1995, Larry Masinter wrote:
> > The proposal is to add byte ranges to URLs (in general, it seems). I
> > don't think it belongs there; at best, byte ranges make sense as an
> > addon to the HTTP protocol.
> 
> Then how does one build a URL to point to minutes 6 through 8 of a 1-hour
> 60-megabyte DJ set?  Or to message 2064 (byte range 2254322-2257934) in a
> 4 megabyte mailbox archive?  Sure, if I'm using an HTTP-aware mailbox
> reader or audio viewer that's a possibility... but then I can only launch
> a range request from that type of viewer.  Ick. 

I'd also like to see something more general.  (I'd also like to see something
integrated with, rather than orthogonal to, existing fragment references
but that's just me.)

>   http://host/path/to/object?object_arguments;request_headers
> 
>   object_arguments: a url-encoded list of name-value pairs 
> 	i.e. name=brian&age=22
> 
>   request_headers: a url-encoded list of request headers, which only
>         make sense in the context of the protocol used (in this case HTTP)
> 	This generality is so that URL's aren't hindered by HTTP-only
> 	specifications.

This also has interesting potential to allow hacks like the extra anchor
stuff like DN and CRYPTOPTS in SHTTP to be folded into the URLs.

> I can already sense some problems.  Here's an interesting URL:
> 
> http://whitehouse.gov:25/;MAIL+FROM=madmad@bomber.org&RCPT+TOpresident&DATA\nFrontLawn,2pm,May16th\n.\n
> 
> Though I suppose some catches could be put in place for this situation, 
> can we protect against that for every protocol?  At what point does a 
> sufficiently obfuscated (to the human eye) extended URL become a malicious
> virus-ish mechanism for mayhem?

You can do that today with Gopher URLs; this vulnerability has been known
about for years, popular browsers are susecptible to it, and nobody seems to
be complaining, so obviously it's not a real problem. :-) * 0.5

- Marc

		  Marc VanHeyningen  marcvh@spry.com
Disclaimer:
 If this were an official announcement from Spry-CompuServe Internet Division,
 it would have begun with the phrase "FOR IMMEDIATE RELEASE."

eturn-Path: mvanheyn@pellet.spry.com 
Return-Path: <mvanheyn@pellet.spry.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB12689; Thu, 18 May 1995 11:51:56 +0500
Received: from homer.spry.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA21612; Thu, 18 May 1995 11:49:11 +0500
Received: from pellet.spry.com (pellet [198.185.1.189]) by homer.spry.com (8.6.9/8.6.9) with ESMTP id IAA16816 for <www-talk@w3.org>; Thu, 18 May 1995 08:49:41 -0700
X-Mailer: exmh version 1.5.3 12/28/94
From: marcvh@spry.com (Marc VanHeyningen)
To: www-talk@w3.org
Subject: Re: Byte ranges -- formal spec proposal 
In-Reply-To: Your message of "Wed, 17 May 1995 13:22:40 PDT."
             <199505172022.NAA06472@neon.netscape.com> 
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Thu, 18 May 1995 08:48:53 -0700
Message-Id: <14310.800812133@pellet.spry.com>
Sender: mvanheyn@pellet.spry.com
Content-Length: 1077

Per various requests, I'm sending this to www-talk only so people will not
see it multiple times.

My main question is what exactly a byte range is intended to mean.  For
example, if I'm a client and I have a fresh cached version of <http://foo>,
what should I display when dereferencing <http://foo;byterange=100-200>?
A representation of just that range, using the same content-type as the main
part?  The whole cached version of <http://foo> with the window aligned to
that portion of the document and everything else greyed out, but still
readable for context?

In essence, is the purpose of this merely to decrease network bandwidth
by allowing servers to send only part of a representation, or to provide
a mechanism for making an HTTP URL go to specific portions of an object?
Any more elaborate comments on the proposal would need to first be
grounded in which of these two objectives we're trying to attain.

(Aside -- apart from text/plain, how many widely used content-types
are there where a fragment of the object is a legal object of that
content-type?)

- Marc

eturn-Path: flavio@dimap.ufrn.br 
Return-Path: <flavio@dimap.ufrn.br>
Received: from dxmint.cern.ch by www19 (5.0/NSCS-1.0S) 
	id AA00433; Thu, 18 May 1995 12:13:44 +0500
Received: from www0.cern.ch by dxmint.cern.ch
	id AA27157; Thu, 18 May 1995 18:13:01 +0200
Received: from dxmint.cern.ch by www0.cern.ch (5.0/SMI-4.0)
	id AA01602; Thu, 18 May 1995 18:11:18 --100
Received: from sol.dimap.ufrn.br by dxmint.cern.ch
	id AA26029; Thu, 18 May 1995 18:06:27 +0200
Received: by sol.dimap.ufrn.br (4.1/SMI-4.1)
	id AA02432; Thu, 18 May 95 11:51:05 EST
Date: Thu, 18 May 95 11:51:05 EST
From: flavio@dimap.ufrn.br (Flavio Marcelo C B do Amaral)
Message-Id: <9505181451.AA02432@sol.dimap.ufrn.br>
To: www-talk@www0.cern.ch
Subject: RFC 822
Content-Length: 10

Hello All
eturn-Path: lazear@dockside.mitre.org 
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA01171; Thu, 18 May 1995 12:19:54 +0500
Received: from mwunix.mitre.org by www10.w3.org (5.0/NSCS-1.0S) 
	id AA21924; Thu, 18 May 1995 12:17:17 +0500
Return-Path: lazear@dockside.mitre.org
Received: from gateway.mitre.org (gateway.mitre.org [128.29.31.10]) by mwunix.mitre.org (8.6.10/8.6.4) with SMTP id MAA15926; Thu, 18 May 1995 12:16:59 -0400
From: lazear@dockside.mitre.org
Received: from dockside.mitre.org by gateway.mitre.org (4.1/SMI-4.1)
	id AA03432; Thu, 18 May 95 12:16:56 EDT
Received: by dockside.mitre.org (4.1/SMI-4.1)
	id AA06864; Thu, 18 May 95 12:16:07 EDT
Message-Id: <9505181616.AA06864@dockside.mitre.org>
To: Ari Luotonen <luotonen@netscape.com>
Cc: www-talk@w3.org, http-wg@cuckoo.hpl.hp.com, uri@bunyip.com,
        lazear@dockside.mitre.org
Subject: Re: Byte ranges -- formal spec proposal 
In-Reply-To: Your message of "Wed, 17 May 95 13:22:40 PDT."
             <199505172022.NAA06472@neon.netscape.com> 
Date: Thu, 18 May 95 12:16:02 -0400
Content-Length: 54

"byte range"...is this the new FTAM option?  :-)  :-)
eturn-Path: masinter@parc.xerox.com 
Return-Path: <masinter@parc.xerox.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA01553; Thu, 18 May 1995 12:22:44 +0500
Received: from alpha.xerox.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA21976; Thu, 18 May 1995 12:20:06 +0500
Received: from golden.parc.xerox.com ([13.1.100.139]) by alpha.xerox.com with SMTP id <14570(3)>; Thu, 18 May 1995 09:19:43 PDT
Received: by golden.parc.xerox.com id <2761>; Thu, 18 May 1995 09:19:35 -0700
To: john@math.nwu.edu
Cc: www-talk@w3.org
In-Reply-To: John Franks's message of Thu, 18 May 1995 08:55:25 -0700 <199505181555.KAA08055@hopf.math.nwu.edu>
Subject: Re: Byte ranges -- formal spec proposal
From: Larry Masinter <masinter@parc.xerox.com>
Sender: Larry Masinter <masinter@parc.xerox.com>
Fake-Sender: masinter@parc.xerox.com
Message-Id: <95May18.091935pdt.2761@golden.parc.xerox.com>
Date: Thu, 18 May 1995 09:19:26 PDT
Content-Length: 801

> It has been suggested that byte ranges could be supported by either a new
> HTTP method or by a new HTTP header.  The point of byteranges is to allow
> HTML anchors to request a range of bytes from a document.

This originally sounded like it was a proposal for a new form of URL.
Now, it looks like it is a proposal for a convention among HTTP
servers -- like CGI -- which HTTP server implementations could either
include or not, with the client completely unaware. The only issue is
whether proxy servers might be aware of the convention and deal with
overlapping byte ranges.

Except for the case of extracting individual segments out of larger
ascii files, though, I'm having trouble thinking of any real examples
of this, and, in particular, any examples where the documents aren't
text/plain.
eturn-Path: jonm@netscape.com 
Return-Path: <jonm@netscape.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AC03158; Thu, 18 May 1995 12:33:51 +0500
Received: from neon.netscape.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA22068; Thu, 18 May 1995 12:31:35 +0500
Received: from [198.93.92.20] (jonmmac.mcom.com [198.93.92.20]) by neon.netscape.com (950215.SGI.8.6.10/8.6.9) with SMTP id JAA17442; Thu, 18 May 1995 09:27:11 -0700
Date: Thu, 18 May 1995 09:27:11 -0700
X-Sender: jonm@pop.mcom.com
Message-Id: <v02110101abe0c340ec12@[198.93.92.20]>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
To: masinter@parc.xerox.com,
        Multiple recipients of list <www-talk@www10.w3.org>
From: jonm@netscape.com (Jon Mittelhauser)
Subject: Re: Byte ranges -- formal spec proposal
Content-Length: 1222

At 5:06 AM 5/18/95, Larry Masinter wrote:
>Ari's proposal said:
>but I still ask:
>
>Can PDF documents be retrieved partially?

Yes.

>A priori?

Yes.

>How many bytes
>would you know to ask for in the first place?

You would have to ask the PDF experts (e.g. Adobe).  However, if you saw the
demo (or heard about it) at the Seybold publishing conference, it was done
there.  We (Adobe & Netscape) demonstrated a prototype browser implementation
which treated PDF as a first class object and included streaming and byte range
retreival via server CGI.  Ari is simply attempting (for Netscape) to help
create a standard means of doing this very common procedure so that there are
not 100's of different implementations -- the other very positive aspect
(especially from Ari's point of view) is that if there is a standard way of
doing it, his proxy can understand it...(and cache it...etc...etc...)

>
>I think these are both scenarios that might benefit from a GET-PARTIAL
>or some kind of modification to GET in HTTP, but I don't see either of
>these scenarios being justification for a new feature in URLs.
>

One problem with a new method or a Header based method is that it breaks
all existing proxy setups...

-Jon


eturn-Path: kfoss@doit.wisc.edu 
Return-Path: <kfoss@doit.wisc.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA13447; Thu, 18 May 1995 14:29:55 +0500
Received: from brown-swiss.macc.wisc.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA23240; Thu, 18 May 1995 14:29:50 +0500
Received: from sogning.macc.wisc.edu by brown-swiss.macc.wisc.edu;
          id AA14161; NX5.67c/42; Thu, 18 May 95 13:29:01 -0500
Message-Id: <v0211011eabe143b91c22@[144.92.30.249]>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Thu, 18 May 1995 13:36:20 -0500
To: david@kwantlen.bc.ca
From: Kurt Foss <kfoss@doit.wisc.edu>
Subject: Re: Netscape and Adobe
Cc: <www-talk@www10.w3.org>
Content-Length: 869

>I was reading a rather old issue of InformationWeek (Apr 10/95) about
>Netscape and Adobe's agreement to add Acrobat to Navigator 1.1. Well,
>1.1N is out but I don't see any sign of Acrobat.
>
>Anyone heard about this?
>

Patience is a virtue. I wouldn't expect anything quite so soon, but I'd bet
that a forthcoming version of the browser will show signs of the recently
announced Adobe-Netscape arrangement.

Regards ~ Kurt

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Kurt Foss * Media Technology * U_Wisconsin-Madison Div. of Info. Technology
608/262-1738 * FAX 608/262-4679 * kfoss@doit.wisc.edu (UW) or kfoss@itis.com
          "Everything I did in my life that was worthwhile I caught hell for."
                  - Earl Warren, quoted in 'DO IT! Let's Get Off Our_ Buts_'
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


eturn-Path: kfoss@doit.wisc.edu 
Return-Path: <kfoss@doit.wisc.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB20116; Thu, 18 May 1995 16:28:25 +0500
Received: from brown-swiss.macc.wisc.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA24485; Thu, 18 May 1995 16:28:19 +0500
Received: from sogning.macc.wisc.edu by brown-swiss.macc.wisc.edu;
          id AA21136; NX5.67c/42; Thu, 18 May 95 15:27:45 -0500
Message-Id: <v0211012cabe15f91a6d7@[144.92.30.249]>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Thu, 18 May 1995 15:35:04 -0500
To: david@kwantlen.bc.ca
From: Kurt Foss <kfoss@doit.wisc.edu>
Subject: Re: Netscape and Adobe
Cc: www-talk@www10.w3.org
Content-Length: 1809


>I was reading a rather old issue of InformationWeek (Apr 10/95) about
>Netscape and Adobe's agreement to add Acrobat to Navigator 1.1. Well,
>1.1N is out but I don't see any sign of Acrobat.
>
>Anyone heard about this?

More ... here's the actual news item from Adobe's web site --->

>---
Roadmap Specifics

The companies disclosed a four step roadmap for delivering a complete suite of
Internet publishing tools.

First, the Macintosh(R) and Windows(tm) versions of Netscape Navigator(tm) 1.1
will support the Acrobat Weblink(tm) software plug-in, a free add-on application
from Adobe that allows Acrobat documents to link to other documents on the
Internet.

The companies will also collaborate on a future version of Netscape Navigator
that will seamlessly view documents in Adobe's Portable Document Format (PDF),
the open, cross-platform file format created by Acrobat software.

Additional integration efforts will include work on Netscape server software to
provide quick access to PDF documents across the Internet, allowing users to
download portions of PDF files at a time for faster on screen viewing.

Lastly, Adobe will extend its authoring applications to more fully support the
ability to import and export PDF files and will provide the ability to output to
Hypertext Markup Language (HTML) in a future version of Adobe PageMaker(tm).
>---

~ Kurt

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
Kurt Foss * Media Technology * U_Wisconsin-Madison Div. of Info. Technology
608/262-1738 * FAX 608/262-4679 * kfoss@doit.wisc.edu (UW) or kfoss@itis.com
          "Everything I did in my life that was worthwhile I caught hell for."
                  - Earl Warren, quoted in 'DO IT! Let's Get Off Our_ Buts_'
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


eturn-Path: brian@organic.com 
Return-Path: <brian@organic.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB23025; Thu, 18 May 1995 17:10:13 +0500
Received: from eat.organic.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA24820; Thu, 18 May 1995 17:10:05 +0500
Received: (from brian@localhost) by eat.organic.com (8.6.11/8.6.9) id OAA22331; Thu, 18 May 1995 14:10:04 -0700
Date: Thu, 18 May 1995 14:10:02 -0700 (PDT)
From: Brian Behlendorf <brian@organic.com>
Subject: Re: Byte ranges -- formal spec proposal 
To: Marc VanHeyningen <marcvh@spry.com>
Cc: Multiple recipients of list <www-talk@www10.w3.org>
In-Reply-To: <14310.800812133@pellet.spry.com>
Message-Id: <Pine.3.89.9505181329.r15253-0100000@eat.organic.com>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 1765

On Thu, 18 May 1995, Marc VanHeyningen wrote:
> (Aside -- apart from text/plain, how many widely used content-types
> are there where a fragment of the object is a legal object of that
> content-type?)

Absolutely no support for fragmentation: MPEG audio and video (I realized
this after using MPEG files as an example in my post last night - ack) I
would be willing to bet most other recent formats fit this description. 
Also: GIF, JPEG, others.  Most of these formats have smart readers which 
can *recover* at some point into the stream (sorta like when you turn 
your TV on and it's fuzzy for a second or two), so "no support for 
fragmentation" doesn't mean "a subpart doesn't mean anything".

Minimal support for fragmentation (i.e., there are places where the object
can be split): PDF, Quicktime (certain codecs), HTML (?), 
"mailbox-message" format plain text.

Completely fragmentable: plain text, aiff, wav, and mu-law sound formats.  
Basically, any unstructured and uncompressed file formats, which also 
happens to be the least useful type of file.


To repeat: just because a file's fragments aren't valid file types on 
their own doesn't mean they're not useful.  They might be valid in the 
context of some other file type which has inlined them.  For example, 
let's say I have an HTML file like this:

...
   An example of VRML:
   <PRE>
   <A REL="EMBED" HREF="http://host/foo.wrl?byterange=22342-22401"></A>
   </PRE>

Whereas the object representation the URL pointed to was not a valid VRML 
file, and thus was not typed VRML, but was perfectly embeddable in HTML.

Thoughts?

	Brian


--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--
brian@organic.com  brian@hyperreal.com  http://www.[hyperreal,organic].com/

eturn-Path: mvanheyn@pellet.spry.com 
Return-Path: <mvanheyn@pellet.spry.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB24344; Thu, 18 May 1995 17:28:31 +0500
Received: from homer.spry.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA24959; Thu, 18 May 1995 17:28:25 +0500
Received: from pellet.spry.com (pellet [198.185.1.189]) by homer.spry.com (8.6.9/8.6.9) with ESMTP id OAA04607; Thu, 18 May 1995 14:28:53 -0700
X-Mailer: exmh version 1.5.3 12/28/94
From: marcvh@spry.com (Marc VanHeyningen)
To: Brian Behlendorf <brian@organic.com>
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Re: Byte ranges -- formal spec proposal 
In-Reply-To: Your message of "Thu, 18 May 1995 14:10:02 PDT."
             <Pine.3.89.9505181329.r15253-0100000@eat.organic.com> 
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Thu, 18 May 1995 14:28:04 -0700
Message-Id: <15453.800832484@pellet.spry.com>
Sender: mvanheyn@pellet.spry.com
Content-Length: 1260

Brian awk^H^H^Hsed:
> On Thu, 18 May 1995, Marc VanHeyningen wrote:
> > (Aside -- apart from text/plain, how many widely used content-types
> > are there where a fragment of the object is a legal object of that
> > content-type?)
> 
> Minimal support for fragmentation (i.e., there are places where the object
> can be split): PDF, Quicktime (certain codecs), HTML (?), 
> "mailbox-message" format plain text.
> 
> Completely fragmentable: plain text, aiff, wav, and mu-law sound formats.  
> Basically, any unstructured and uncompressed file formats, which also 
> happens to be the least useful type of file.

Well, the most common format, audio/basic, has header information which
contains magic numbers and the Hz and stuff like that, and technically is
not a legal audio/basic file without them, though many players will tolerate
this.  I would put this in the same category as HTML; arbitrary segments
probably are not legal, but viewers might be able to tolerate them.

> To repeat: just because a file's fragments aren't valid file types on 
> their own doesn't mean they're not useful.

Indeed; you show a potentially useful, if rather brittle, example.
However, if the fragment is not a valid file type, how then should the
server label it?

- Marc

eturn-Path: frystyk@w3.org 
Return-Path: <frystyk@w3.org>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB14327; Thu, 18 May 1995 22:33:20 +0500
Received: from www20 (www20.w3.org) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA28092; Thu, 18 May 1995 22:33:12 +0500
Received: by www20 (5.0/NSCS-1.0S) 
	id AA29284; Thu, 18 May 1995 22:33:11 +0500
Date: Thu, 18 May 1995 22:33:11 +0500
From: frystyk@w3.org (Henrik Frystyk Nielsen)
Message-Id: <9505190233.AA29284@www20>
To: www-talk@w3.org
Subject: Re: Byte ranges -- formal spec proposal
Reply-To: Henrik Frystyk Nielsen <frystyk@w3.org>
X-Sun-Charset: US-ASCII
Content-Length: 2036


It doesn't look like this mail made it in the first go, sorry if
it pops up again!

Larry wrote:

> Please give a single, specific and realistic example. If you think a
> framemaker document can be retrieved a byte range at a time, what is
> the initial byte range? How do you know?
> 
> Can PDF documents be retrieved partially? A priori? How many bytes
> would you know to ask for in the first place? Is 'the first part of a
> PDF file' still 'application/pdf'?

I can see several applications where this would be a good and
useful idea, for example in audio and video - or as a general approach to
having external links into any compound data format - quite similar to
what HyperG is doing. 

However, I don't like the specific byte count so much as the idea
in general. If we can keep it to a set of logical names as a function
of media type then I would be a lot more happy, for example frame
count etc. Can't we just change the `Misc' part into the main part
of the proposal - this part talks about the logical names - not 
specific byte numbers.

In my opinion, URLs are already quite shaky and if we can avoid having
another update problem in URLs every time a new version of the resource
behind it is generated, if would certainly prefer this!

> > 2. Just regular Web clients where image or document transfer is
> > interrupted before the entire image/document is received, and then
> > later restarted.  So instead of starting all over again you could only
> > transfer the remaining part of it.

This is also one application but I think that the ones above are much more 
frequently.
 
> This is http specific, isn't it? Surely a FTP server won't support
> this. So, does this belong in the URL at all? 

No, you are right :-)


-- cheers --

Henrik Frystyk                                          frystyk@W3.org
World-Wide Web Consortium,                              Tel + 1 617 258 8143
MIT/LCS, NE43-356					Fax + 1 617 258 8682
77 Massachusetts Avenue
Cambridge MA 02154, USA



 


----- End Included Message -----

eturn-Path: fielding@avron.ics.uci.edu 
Return-Path: <fielding@avron.ics.uci.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA17007; Thu, 18 May 1995 23:22:23 +0500
Received: from paris.ics.uci.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA28682; Thu, 18 May 1995 23:22:19 +0500
Received: from avron.ics.uci.edu by paris.ics.uci.edu id aa20333;
          18 May 95 19:36 PDT
To: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Re: Byte ranges -- formal spec proposal 
In-Reply-To: Your message of "Thu, 18 May 1995 06:20:03 +0500."
             <Pine.3.89.9505172203.h15253-0100000@eat.organic.com> 
Date: Thu, 18 May 1995 19:36:11 -0700
From: "Roy T. Fielding" <fielding@avron.ics.uci.edu>
Message-Id:  <9505181936.aa20333@paris.ics.uci.edu>
Content-Length: 5361

Brian writes:

> So, does a "byte range" constitute a variation of the object, or a new
> object itself, which deserves a unique URL?  Compelling cases could be
> made on either side, but I think in this situation it truely is a
> variation of the object.  But now we have a problem - the WWW Link Model
> (hi roy!) only lets me link to *objects* (i.e., URL's), not particular
> variations/representations of objects, if I understand things correctly. 
> For example, if I have an object that represents my home page, and my home
> page object returns both HTML 2.0 and HTML 3.0 representations of itself,
> there's no way for me to *force* an HTML 2.0 browser to see the HTML 3.0
> representation without giving the HTML 3.0 representation its own, 
> un-content-negotiated URL.  Feh. 

That is quite correct -- in fact, I came to the same conclusion Tuesday
night (while lying in bed staring at the ceiling, of course) regarding
the need for URL parameters for versioning and content negotiation.
The same applies to byte ranges, page ranges, etc.

> Okay, so here's the problem.  A URL must be able, not required, but able,
> to *completely* describe the request for an object.  In other words, URL's
> must be able to point to particular representations of webbable objects. 

Yep.

> The protocol "method" used.

Nope.

>                              The additional headers.

Only insofar as they affected the chosen representation.

> In fact, in most
> situations today URL's are used to point to representations instead of
> objects - content providers are simply creating unique URL's to every
> representation.  So, we're not breaking anything fundamental here, it
> seems.  Further more: 
> 
> 1) There must be a clear distinction between the part of the URL that 
> describes the *object*, and the part of the URL that describes its 
> representation.

I'll disagree here -- there only needs to be a distinction between variants;
what that distinction is may vary from parameters to file name extensions.

> 2) User-agents must be able to deal with the part of the URL that 
> describes the representation at a higher level - for example, when a user 
> goes to "bookmark" the object, they are asked to chose whether they want 
> to bookmark the object in general or the particular representation of 
> that object.  

UAs need to be able to identify the non-variant part of the URL.

> 3) Responses need to indicate which parts of that representation request 
> influenced the output, so that caches know what to key on (and don't 
> needlessly key on everything in the request.)  I think there's a "vary" 
> header proposed somewhere....

Responses need to include a Location: header which defines the exact
variant chosen, and URI headers which define the available variants
for the resource.

> 4) There must be a defined list of "sanctimonious" headers in HTTP, ones 
> which are always part of the request and are *not* modifiable by the 
> representation-part of the URL.  For example, User-Agent:, or From:.  
> Likewise, content providers should not vary content based on these headers.

Er, well, there's no way to enforce that.  I prefer just requiring that
the UA be informed of what they've received.

> Phew.
> 
> (btw, the CD I'm listening to now seems highly conducive to these kind of 
> thought processes - Air, by Pete Namlook, on FAX)

I nice comfy bed works well too.

> So, here's how I think things should look.  The format:
> 
>   http://host/path/to/object?object_arguments;request_headers

*BOING*  phooey.  The answer is in the Relative URL draft.
I will define "the http URL" in the HTTP/1.0 draft, and it will
be based on the generic-RL syntax.  We can include an appendix
on http URL conventions if that is acceptable to the HTTP WG.

Question:  Why the verbose names?  I preferred "bytes" over "byterange",
and would prefer just "b" even more.  But, I can see where some names
are best left readable and around 4-5 characters:

     http://site/foo;byte=1-100000
     http://site/foo;line=1-100
     http://site/foo;page=4-7
     http://site/foo;chapter=1-100
     http://site/foo;language=en-us
     http://site/foo;version=1.1
     http://site/foo;type=text/html%25version%3D3

     http://site/foo;chapter=1-100?fred+barney
     http://site/foo;language=mi;chapter=1-100?haka+pakeha

NOTE: the similarity between the URI: header's vary values and
the parameter names is mandatory for caching to work sensibly.

     GET http://site/foo?haka+pakeha HTTP/1.0
     User-Agent: Me
     Accept: text/html;q=1, */*;q=0.5
     Accept-Language: mi;q=1, en;q=0.9

     HTTP/1.0 200 OK
     Content-type: application/pdf
     Location: http://site/foo;language=mi;chapter=1-100?haka+pakeha
     URI: <http://site/foo>;vary="byte,chapter,language",
          <http://site/foo;language="mi">;vary="byte,chapter",
          <http://site/foo;language="en-gb">;vary="byte,chapter",
          <urn:/NZ/treaties/waikato>;vary="byte,chapter,language"
     

It is my personal opinion that multiple byte ranges in a single
URL are not useful and only make life difficult -- multiple requests
are more appropriate.


 ....Roy T. Fielding  Department of ICS, University of California, Irvine USA
                                       <fielding@ics.uci.edu>
                      <URL:http://www.ics.uci.edu/dir/grad/Software/fielding>
eturn-Path: craig@passport.ca 
Return-Path: <craig@passport.ca>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB08457; Fri, 19 May 1995 04:13:28 +0500
Received: from diplomatic.passport.ca by www10.w3.org (5.0/NSCS-1.0S) 
	id AA02080; Fri, 19 May 1995 04:13:22 +0500
Received: by diplomatic.passport.ca (Smail3.1.29.1 #6)
	id m0sCNA9-001Bh1C; Fri, 19 May 95 04:12 EDT
Message-Id: <m0sCNA9-001Bh1C@diplomatic.passport.ca>
From: craig@passport.ca (Craig Hubley)
Subject: Re: HTML Link Type Model
To: murray@sco.com
Date: Fri, 19 May 1995 04:12:01 -0400 (EDT)
Cc: html-wg@oclc.org, www-talk@w3.org
Reply-To: craig@passport.ca
In-Reply-To:  <9505171303.aa21227@dali.scocan.sco.COM> from "Murray Maloney" at May 17, 95 01:10:16 pm
X-Mailer: ELM [version 2.4 PL23]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 2053      

> But by abandoning REL/REV, you lose the expressive power
> provided by describing the relationship in both directions.

Not quite.  The fact that some relationships/roles are invertible
can be adequately expressed in other ways.  You simply move the
statement that ROLE="indexes" is the inverse of ROLE="is_indexed_by",
somewhere else.

> Or we could abandon the use of verbs along with the "WWW link model"
> and just stick with nouns.  Would that work for you?

As I mentioned earlier, many of the useful words (like "index") are
both.  We can't abandon the "WWW link model" without IMPOSING A NEW
ONE on the WWW.  As I have stated now several times, HTML can only
define an *interface* to the rest of the WWW, as not all documents
on the WWW are in HTML.  This interface had better be consistent
with the "WWW link model", even if we end up having to rewrite the
"WWW link model" to do so.  I propose that that's what we have to
do.  At least, I think I will move my efforts in that direction,
and (for HTML) concentrate on link types that can safely default to 
'goto (HREF)', which is the only implemented part of the "WWW link
model".  That is, we *know* that WWW documents will know how to do
this (although for some, like RTF, it is not exactly clear how yet).

There *is* an "implemented WWW link model", and it is 'goto (HREF)'.
A means of supporting functions other than 'goto' has not yet been
defined.  Therefore it is dangerous to specify behavior that cannot
safely default to 'goto (HREF)' or require that browsers understand
it.  The basic links types SCO has implemented affect presentation
not navigation, and so their effects can be confined to HTML.  But
anything that might, for instance, present one page as two, with
two different URLs (such as "INCLUDE") has WWW-wide implications.
-- 
Craig Hubley                Business that runs on knowledge
Craig Hubley & Associates   needs software that runs on the net
mailto:craig@hubley.com     416-778-6136    416-778-1965 FAX
Seventy Eaton Avenue, Toronto, Ontario, Canada M4J 2Z5
eturn-Path: @post.demon.co.uk,@basis.demon.co.uk:perry@idl.idi.oclc.org 
Return-Path: <@post.demon.co.uk,@basis.demon.co.uk:perry@idl.idi.oclc.org>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AD11683; Fri, 19 May 1995 04:43:17 +0500
Received: from disperse.demon.co.uk by www10.w3.org (5.0/NSCS-1.0S) 
	id AA02287; Fri, 19 May 1995 04:42:56 +0500
Received: from post.demon.co.uk by disperse.demon.co.uk id aa07468;
          19 May 95 9:25 GMT-60:00
Received: from basis.demon.co.uk by post.demon.co.uk id aa14770;
          19 May 95 9:25 GMT-60:00
Date: Fri, 19 May 95 09:26:50 -0700
From: Perry Newton <perry@idl.idi.oclc.org>
X-Mailer: Mozilla 1.1b3 (Windows; I; 16bit)
Mime-Version: 1.0
To: www-talk@w3.org
Subject: Connecting Gopher and httpd
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset=iso-8859-1
Message-Id:  <9505190925.aa14770@post.demon.co.uk>
Content-Length: 548

Does anyone know how make a gopher server redirect requests to an httpd 
server. 
 
We wish to have users using both/either gopher and/or Web browsers but as we 
have an enhanced httpd server which includes document database access and text 
retrieval we do not want to duplicate file sets.  We want the requests that go 
to gopher to be mapped and directed to the httpd and then returned  
appropriately. 
 
Any suggestions or sources of information 
 
Perry Newton 
Information Dimensions 
perry@idl.idi.oclc.org 
http://www.id-london.co.uk/ 



eturn-Path: frystyk@w3.org 
Return-Path: <frystyk@w3.org>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA04023; Fri, 19 May 1995 21:01:00 +0500
Received: from www20 (www20.w3.org) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA07648; Fri, 19 May 1995 21:00:58 +0500
Received: by www20 (5.0/NSCS-1.0S) 
	id AA00305; Fri, 19 May 1995 21:00:57 +0500
Date: Fri, 19 May 1995 21:00:57 +0500
From: frystyk@w3.org (Henrik Frystyk Nielsen)
Message-Id: <9505200100.AA00305@www20>
To: www-talk@w3.org
Subject: Re: Byte ranges -- formal spec proposal
Reply-To: Henrik Frystyk Nielsen <frystyk@w3.org>
X-Sun-Charset: US-ASCII
Content-Length: 1995


It doesn't look like this mail made it in the first go, sorry if
it pops up again!

Larry wrote:

> Please give a single, specific and realistic example. If you think a
> framemaker document can be retrieved a byte range at a time, what is
> the initial byte range? How do you know?
> 
> Can PDF documents be retrieved partially? A priori? How many bytes
> would you know to ask for in the first place? Is 'the first part of a
> PDF file' still 'application/pdf'?

I can see several applications where this would be a good and
useful idea, for example in audio and video - or as a general approach to
having external links into any compound data format - quite similar to
what HyperG is doing. 

However, I don't like the specific byte count so much as the idea
in general. If we can keep it to a set of logical names as a function
of media type then I would be a lot more happy, for example frame
count etc. Can't we just change the `Misc' part into the main part
of the proposal - this part talks about the logical names - not 
specific byte numbers.

In my opinion, URLs are already quite shaky and if we can avoid having
another update problem in URLs every time a new version of the resource
behind it is generated, if would certainly prefer this!

> > 2. Just regular Web clients where image or document transfer is
> > interrupted before the entire image/document is received, and then
> > later restarted.  So instead of starting all over again you could only
> > transfer the remaining part of it.

This is also one application but I think that the ones above are much more 
frequently.
 
> This is http specific, isn't it? Surely a FTP server won't support
> this. So, does this belong in the URL at all? 

No, you are right :-)


-- cheers --

Henrik Frystyk                                          frystyk@W3.org
World-Wide Web Consortium,                              Tel + 1 617 258 8143
MIT/LCS, NE43-356					Fax + 1 617 258 8682
77 Massachusetts Avenue
Cambridge MA 02154, USA
eturn-Path: gtn%ebt.com@www10.w3.org 
Return-Path: US 
Received: from mintaka.lcs.mit.edu by www19 (5.0/NSCS-1.0S) 
	id AL28431; Sat, 20 May 1995 02:07:07 +0500
Received: from www10.w3.org by MINTAKA.LCS.MIT.EDU id aa14401;
          20 May 95 1:47 EDT
Received: from ebt-inc.ebt.com (kirk.ebt.com) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA03099; Fri, 19 May 1995 07:06:31 +0500
Received: (from gtn@localhost) by ebt-inc.ebt.com (8.6.9/8.6.9) id HAA05868; Fri, 19 May 1995 07:09:31 -0400
Date: Fri, 19 May 1995 07:09:31 -0400
From: Gavin Nicol <gtn@ebt.com>
Message-Id: <199505191109.HAA05868@ebt-inc.ebt.com>
To: www-talk@www10.w3.org
In-Reply-To: <Pine.3.89.9505181329.r15253-0100000@eat.organic.com> (message from Brian Behlendorf on Fri, 19 May 1995 05:38:55 +0500)
Subject: Re: Byte ranges -- formal spec proposal
Content-Length: 0

>Minimal support for fragmentation (i.e., there are places where the object
>can be split): PDF, Quicktime (certain codecs), HTML (?),
>"mailbox-message" format plain text.
 
HTML is not fragmentable in a meaningful manner. Tp fragment HTML, you
must first parse the document, and if you do that, byte ranges make
almost no sense as a unit of fragmentation. I will be sending an RFC
out one day about this.

>To repeat: just because a file's fragments aren't valid file types on
>their own doesn't mean they're not useful.  They might be valid in the
>context of some other file type which has inlined them.  For example,
>let's say I have an HTML file like this:
> 
>..
>   An example of VRML:
>   <PRE>
>   <A REL="EMBED" HREF="http://host/foo.wrl?byterange=22342-22401"></A>
>   </PRE>
> 
>Whereas the object representation the URL pointed to was not a valid VRML
>file, and thus was not typed VRML, but was perfectly embeddable in HTML.
 
What happens if this range of bytes actually references a document
that has changed recently? The user will get a page with a (probably)
garbage sample. Not robust....

eturn-Path: brian@organic.com 
Return-Path: <brian@organic.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA04486; Sat, 20 May 1995 23:49:24 +0500
Received: from eat.organic.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA17916; Sat, 20 May 1995 23:49:15 +0500
Received: (from brian@localhost) by eat.organic.com (8.6.11/8.6.9) id UAA10635; Sat, 20 May 1995 20:49:20 -0700
Date: Sat, 20 May 1995 20:49:20 -0700 (PDT)
From: Brian Behlendorf <brian@organic.com>
Subject: Re: Byte ranges -- formal spec proposal
To: www-talk@www10.w3.org
Message-Id: <Pine.3.89.9505202036.T15253-0100000@eat.organic.com>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 6261


One last thing I'd like to throw into the conversation here, and then I 
feel all has been said that needs to be said, and we should try and find 
rough consensus. (BTW, are there any "tools" for finding what 
can be considered rough consensus?  A straight WWW-based poll is 
probably out of line...)

************************************************************************
SUMMARY: all we really need, standards-wise, is a new 300-level HTTP 
response, "Contains".
************************************************************************

Several good examples have been brought up of files that can be comprised of
segments, where each of those segments is a valid file of the same data-type,
as an argument for this proposal.  However, in almost all of the examples,
there were only *specific* byte ranges which would work, in which the
requested object would really be usable.  Thus, for most of these examples,
you could just ask for "parts 0-3" or "2-5" or "3-end", and the right thing
would happen.  In only one of the examples was *true* random access
necessary, and that was to resume downloading of a file if it was interrupted
part of the way through.  Keep this example off to the side for the next few
paragraphs. 

Instead of thinking about one URL that represents a collection of objects,
why not give each object their own unique URL, and devise a way of addressing
a collection of URL's?  This is similar to byterange, but more general. 
Let's say somewhere a mapping takes place that translates URL1 into a
container for URL2, URL3, URL4, etc.  I have a hunch this is URC/URI
territory, but I don't know enough yet about the specific URC proposals
floating around yet to know if this is already being considered. 

So, it works like this:

Client asks for URL1. URL1 gets mapped at a server somewhere into a composite
body whose parts are URL2, URL3, and URL4. * If it doesn't find a place to
either inline or link URL3, URL4, etc., it's up to the browser to figure out
how to represent that "auxiliary" file.  Maybe it just keeps it around until
it can be represented later. 

Caches work just as they always have.  If they can cache that container 
mapping, so much the better.  The important thing is that URL2, URL3, 
URL4, etc., can be ANYTHING THEY WANT TO BE - there's no need to give 
them some sort of formal syntax, caches know from the mapping from URL1 
how they assemble together.  If the server prefers knowing them as 
byteranges, it doesn't matter.  I.e., we can have

  http://host/path/file 
    is-a-container-for
      http://host/path/file;byterange=0-30
      http://host/path/file;byterange=31-60

or

  http://host/path/file
    is-a-container-for
      http://host/path/file?part1
      http://host/path/file?part2

or even

  http://host/path/file
    is-a-container-for
      http://host/path/file2
      http://host2/path/script
      ftp://host3/path/file3

and either way the client or proxy will know when it has the whole 
object, or just its parts.

Finally, this also allows "parts" to be members of more than one 
container, something none of the byterange proposals had considered.  I 
think this is a good thing, can anyone think of a situation where this 
isn't?  In fact they can even be on completely separate servers.


Yes, THIS REQUIRES CHANGES TO BROWSERS AND SERVERS.  Minimally.  Why 
are we so afraid of that?

There are a couple really good side effects now that I think about it.
For example, right now Netscape's progressive-rendering algorithm has to 
wait until it recognizes a reference to an inlined image before it can 
start grabbing it.  If it could be told that "URL1 contains this HTML 
page and these inlined images" then it could possibly be more efficient 
in what it does.  Additionally, a content provider could "bundle" icons 
with one page that weren't necessarily inlined on that page, but which 
are used by subsequent pages, so that when visitors go to that subsequent 
page, the icons are already loaded.

I can give plenty of examples of how this could work for just about 
every application discussed so far.  It would seem to be pretty 
straightforward for servers to generate these mappings for a large PDF 
file, presuming there's some way for it to query the PDF file to know 
where it can be segmented.

So, now, back to the resume-downloading-at-point-x.  This is 
semantically a much different operation than "give me part x",
so let's just give it its own request header:

Startbyte: 204567

....would mean start the post-response-header transmission at byte 204567 into
the response, counting from the end of the response headers (\r\n\r\n, or
\n\n).  Who cares if this is a CGI script or actual file, eh?  :)


********************************************************************

So, I suppose in the end I'm proposing a new 300-level HTTP header, 
something like

  305 Contains Mapping

     o Following: anything
  
     o Required Headers: none

  The server returns an HTTP object comprised of a newline-delimited 
  list of URI's which this URL is said to "contain".  The client is expected
  to fetch these URL's and plug them together, representing this 
  requested URL as the canonical URL for this collection.  The other HTTP 
  headers on this object apply *only* to this object, and this response 
  should be cached where possible.

*******************************************************************

*Feedback*, please.  I hate having all these ideas and no time to 
implement them in a browser (though I'd be happy to implement this on the 
server side in Apache).

Roy?  Dan?  Henrik?

	Brian

--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--
brian@organic.com  brian@hyperreal.com  http://www.[hyperreal,organic].com/


* - Order is insignificant - a browser first starts rendering URL2 and looks
for where to start plugging in URL3, etc, but that should just be an
optimization, browsers can plug things together however they wish.  Some
network-aware file formats like VRML already have the concept of nesting
inlines, which HTML doesn't have (yet), so that order could to be
created by a depth- or breadth-first traversal of the scene to aid 
rendering, but in a real directed graph that's not necessary.


eturn-Path: gtn@ebt.com 
Return-Path: <gtn@ebt.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AC08658; Sun, 21 May 1995 08:43:19 +0500
Received: from ebt-inc.ebt.com (kirk.ebt.com) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA23374; Sun, 21 May 1995 08:43:07 +0500
Received: (from gtn@localhost) by ebt-inc.ebt.com (8.6.9/8.6.9) id IAA12490; Sun, 21 May 1995 08:46:09 -0400
Date: Sun, 21 May 1995 08:46:09 -0400
From: Gavin Nicol <gtn@ebt.com>
Message-Id: <199505211246.IAA12490@ebt-inc.ebt.com>
To: www-talk@www10.w3.org
In-Reply-To: <Pine.3.89.9505202036.T15253-0100000@eat.organic.com> (message from Brian Behlendorf on Sat, 20 May 1995 23:51:55 +0500)
Subject: Re: Byte ranges -- formal spec proposal
Content-Length: 2541

>Several good examples have been brought up of files that can be comprised of
>segments, where each of those segments is a valid file of the same data-type,
>as an argument for this proposal.  However, in almost all of the examples,
>there were only *specific* byte ranges which would work, in which the
>requested object would really be usable.  Thus, for most of these examples,
>you could just ask for "parts 0-3" or "2-5" or "3-end", and the right thing
>would happen.  In only one of the examples was *true* random access
>necessary, and that was to resume downloading of a file if it was interrupted
>part of the way through.  Keep this example off to the side for the next few
>paragraphs.

I've been meaning to write up an RFC on how DynaWeb handles large
files. As I've said, DynaWeb breaks a document into parts based on the
structure of the data. In particular, DynaWeb does runtime conversion
from SGML to HTML, and the smallest addressable part of a document in
DynaWeb is a single SGML element.

As you all probably know, an SGML document basically forms a
heirarchy of nested elements, or in other words, a tree. Filesystems,
in general, are also trees. It seemed natural to me to use the same
*type* of URL for files, and for sub-document addressing.

As such, DynaWeb actually supports 3 sub=document addressing modes,
which are pretty much taken straight from the TEI guidelines:

   http://www.ebt.com/collection/book/doc=1/chap=2/sect=3
   http://www.ebt.com/collection/book/1/2/3
   http://www.ebt.com/collection/book/1

The first form accesses elements in the heirarchy by *typed* child
number, the second form accesses elements based on child number,
irrespective of type, and the last is a direct element address. In
practice, because few people ever access the server except by
browsing, the last form can be used in most cases. I would like to
argue that such an addressing scheme is applicable to many other types
of data as well.   

As I said before, my real problem with byte-ranges is that generally,
they don't make sense. Ranges of *parts* does make sense however. One
other problem I have is that the format of a URL should really be
application dependent, so why make recommendations for cases where it
is meaningless? Let's leave it to the application (ie. the server),
until we are ready to design a far more general linking mechanism.

Loot at http://www.ebt.com/ to see how DynaWeb works.

PS. I should note that the above naming scheme is very, very useful in
our case, but it drives spiders wild....
eturn-Path: Carl.von.Loesch@arbi.informatik.uni-oldenburg.de 
Return-Path: <Carl.von.Loesch@arbi.informatik.uni-oldenburg.de>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA15820; Sun, 21 May 1995 17:29:56 +0500
Received: from arbi.informatik.uni-oldenburg.de by www10.w3.org (5.0/NSCS-1.0S) 
	id AA27163; Sun, 21 May 1995 17:29:21 +0500
Received: by arbi.informatik.uni-oldenburg.de (smail3.1.18 + xalias);
	Sun, 21 May 95 23:29 CES
Received: by diamant.Informatik.Uni-Oldenburg.DE (Smail-3.1.28.1)
	id <m0sDIUL-0003CyC@diamant.Informatik.Uni-Oldenburg.DE>;
	Sun, 21 May 95 23:24 MET DST
Message-Id: <m0sDIUK-000AgFC@tuerkis.Informatik.Uni-Oldenburg.DE>
Subject: Re: Byte ranges -- formal spec proposal
To: masinter@parc.xerox.com
Date: Sun, 21 May 1995 23:24:39 +0200 (MET DST)
From: Carl von Loesch <Carl.von.Loesch@arbi.informatik.uni-oldenburg.de>
Cc: www-talk@www10.w3.org
In-Reply-To: <95May18.091935pdt.2761@golden.parc.xerox.com> from "Larry Masinter" at May 18, 95 04:40:48 pm
X-Mailer: ELM [version 2.4 PL22]
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
Content-Length: 1400      

| > It has been suggested that byte ranges could be supported by either a new
| > HTTP method or by a new HTTP header.  The point of byteranges is to allow
| > HTML anchors to request a range of bytes from a document.

But I prefer the URL method much more, as it makes perfectly sense to
request a part of a binary from an FTP server, or just the signature of
a news posting, or a range in a gopher text file.

I like common syntax ("methods" in OOP).. even if the semantics differ
slightly.. whatever seems applicable sounds like a good choice to me.
I wouldn't want it in the headers.

Larry Masinter typeth:
| Except for the case of extracting individual segments out of larger
| ascii files, though, I'm having trouble thinking of any real examples
| of this, and, in particular, any examples where the documents aren't
| text/plain.

Ever tried to download netscape from overseas?

And no, I don't want to explain every netscape user how to download using
ftp and REGET, as it would circumvent any hope of having the stuff cached
by the WWW proxies.

Explain every user how to ftp from mirrors?
We're working on that, but the number of net newbies is growing too fast.

P.S.  Don't forget s/byterange/bytes/	;°)

-- 
     _______					  http://home.pages.de/~lynx/
irc: SymLynX		Carl v. Loesch			  lynx@net.pages.de

LynX@You.might.aswell.use.This.as.my.Email.Addres�
eturn-Path: masinter@parc.xerox.com 
Return-Path: <masinter@parc.xerox.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA26704; Sun, 21 May 1995 20:45:43 +0500
Received: from alpha.xerox.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA28169; Sun, 21 May 1995 20:45:42 +0500
Received: from golden.parc.xerox.com ([13.1.100.139]) by alpha.xerox.com with SMTP id <14421(4)>; Sun, 21 May 1995 17:45:38 PDT
Received: by golden.parc.xerox.com id <2761>; Sun, 21 May 1995 17:45:30 -0700
To: Carl.von.Loesch@arbi.informatik.uni-oldenburg.de
Cc: www-talk@www10.w3.org
In-Reply-To: Carl von Loesch's message of Sun, 21 May 1995 14:24:39 -0700 <m0sDIUK-000AgFC@tuerkis.Informatik.Uni-Oldenburg.DE>
Subject: Re: Byte ranges -- formal spec proposal
From: Larry Masinter <masinter@parc.xerox.com>
Sender: Larry Masinter <masinter@parc.xerox.com>
Fake-Sender: masinter@parc.xerox.com
Message-Id: <95May21.174530pdt.2761@golden.parc.xerox.com>
Date: Sun, 21 May 1995 17:45:24 PDT
Content-Length: 1135

> I said:

> | Except for the case of extracting individual segments out of larger
> | ascii files, though, I'm having trouble thinking of any real examples
> | of this, and, in particular, any examples where the documents aren't
> | text/plain.

> and Carl von Loesch <Carl.von.Loesch@arbi.informatik.uni-oldenburg.de> replied:

> Ever tried to download netscape from overseas?

> And no, I don't want to explain every netscape user how to download using
> ftp and REGET, as it would circumvent any hope of having the stuff cached
> by the WWW proxies.

which might be a clever rejoinder, except that the proposal doesn't
actually solve the problem. That is, this is not a proposal for
finishing a partial transmission when the original transmission
failed, and there doesn't seem to be any practical way to use URL byte
ranges to implement retransmission after failure.


> Explain every user how to ftp from mirrors?
> We're working on that, but the number of net newbies is growing too fast.

Whether there are a lot of net newbies is irrelevant to the question
of whether the proposal to include byte ranges in URLs is useful.



eturn-Path: a.shenker@fml.co.uk 
Return-Path: <a.shenker@fml.co.uk>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB29583; Mon, 22 May 1995 04:28:51 +0500
Received: from fml.co.uk (bleach.fml.co.uk) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA02091; Mon, 22 May 1995 04:28:34 +0500
Date: Mon, 22 May 95 09:32:39 BST
From: a.shenker@fml.co.uk (Adam Shenker)
Message-Id: <9505220832.AA01059@fml.co.uk>
To: www-talk@www10.w3.org
Subject: Security on the CERN server
Content-Length: 814


Dear All,


I am having difficulties in setting password control on our server (CERN HTTP 3.0
running on SUNOS 4.1.3_U1).

I have followed the User Guide instructions in detail, and have tried
setting up both the 'Embedded Protection setup in the Configuration File'
and the 'ACL' Setup.

With the Embedded protection, the server when initialised acknowledges that
certain file areas have protection set for them, but when that area is
accessed, the server (running in Verbose Mode) says that no protection has
been set up.

When using the ACL protection Methodology, all users are locked out of the
area.

Also, the User Guide only defines 'basic' as a legal 'AuthType' variable -
what other possibilities are there?

Has Anybody had similar problems to these? Does anybody know of any solutions?

Thanks,

Adam
eturn-Path: Ton.Verschuren@surfnet.nl 
Return-Path: <Ton.Verschuren@surfnet.nl>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB11170; Mon, 22 May 1995 06:32:27 +0500
Received: from survis.surfnet.nl by www10.w3.org (5.0/NSCS-1.0S) 
	id AA03062; Mon, 22 May 1995 06:31:59 +0500
Message-Id: <9505221031.AA03062@www10.w3.org>
Received: from surprice.surfnet.nl by survis.surfnet.nl with SMTP (PP);
          Mon, 22 May 1995 12:31:21 +0200
To: a.shenker@fml.co.uk
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Re: Security on the CERN server 
In-Reply-To: Your message of Mon, 22 May 1995 04:43:21 +0500.             <9505220832.AA01059@fml.co.uk> 
Organisation: SURFnet bv
Address: Cluetinckborch, P.O. Box 19035, 3501 DA Utrecht, NL
Phone: +31 30 310290
Telefax: +31 30 340903
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Id: <1465.801138676.1@SURFnet.nl>
Date: Mon, 22 May 1995 12:31:18 +0200
From: Ton Verschuren <Ton.Verschuren@surfnet.nl>
Content-Length: 1128

==> From: Adam Shenker

I think it's a bug (or feature ?-) in CERN httpd3.0: you can't use 
ACL's for user dirs (those URL's with a "~" or ~%7E").

Cheers,

Ton Verschuren - Network Development - SURFnet bv
http://www.nic.surfnet.nl/surfnet/persons/tv/

> 
> Dear All,
> 
> 
> I am having difficulties in setting password control on our server (CERN HTTP
  3.0
> running on SUNOS 4.1.3_U1).
> 
> I have followed the User Guide instructions in detail, and have tried
> setting up both the 'Embedded Protection setup in the Configuration File'
> and the 'ACL' Setup.
> 
> With the Embedded protection, the server when initialised acknowledges that
> certain file areas have protection set for them, but when that area is
> accessed, the server (running in Verbose Mode) says that no protection has
> been set up.
> 
> When using the ACL protection Methodology, all users are locked out of the
> area.
> 
> Also, the User Guide only defines 'basic' as a legal 'AuthType' variable -
> what other possibilities are there?
> 
> Has Anybody had similar problems to these? Does anybody know of any solutions
 ?
> 
> Thanks,
> 
> Adam


eturn-Path: Emmanuel.Ponsardin@edelweb.fr 
Return-Path: <Emmanuel.Ponsardin@edelweb.fr>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB21175; Mon, 22 May 1995 08:29:42 +0500
Received: from edelweb.fr ([193.51.12.16]) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA04077; Mon, 22 May 1995 08:29:36 +0500
Received: from champagne.edelweb.fr (champagne.edelweb.fr [193.51.12.33]) by edelweb.fr (8.6.10/8.6.9) with ESMTP id OAA26451 for <www-talk@www10.w3.org>; Mon, 22 May 1995 14:29:32 +0200
Received: from moet.edelweb.fr (moet.edelweb.fr [193.51.12.38]) by champagne.edelweb.fr (8.6.10/8.6.6) with ESMTP id OAA18508 for <www-talk@mail.w3.org>; Mon, 22 May 1995 14:29:28 +0200
From: Emmanuel Ponsardin <Emmanuel.Ponsardin@edelweb.fr>
Received: (ponsardi@localhost) by moet.edelweb.fr (8.6.10/8.6.6) id OAA04740 for www-talk@mail.w3.org; Mon, 22 May 1995 14:29:27 +0200
Date: Mon, 22 May 1995 14:29:27 +0200
Message-Id: <199505221229.OAA04740@moet.edelweb.fr>
To: www-talk@www10.w3.org
Subject: environment variables
Content-Length: 374

Hi,

I have home pages, and on each one, i used to have a "back" and "forward"
button as imagemaps. My problem is that imagine i have 3 pages named A, B and
C.html that lead to a D.html page; on the last one, the "back" button *must*
know which page the user came from...
Is there an environment variable to describe such a situation?

Thanks in advance

Emmanuel Ponsardin
eturn-Path: johnj@enigma.metro2.k12.mn.us 
Return-Path: <johnj@enigma.metro2.k12.mn.us>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AC25593; Mon, 22 May 1995 09:22:46 +0500
Received: from Enigma.metro2.k12.mn.us by www10.w3.org (5.0/NSCS-1.0S) 
	id AA04602; Mon, 22 May 1995 09:22:34 +0500
Received: from [204.72.193.74] (zurich.Metro2.k12.MN.US) by Enigma.metro2.k12.mn.us (5.0/SMI-SVR4)
	id AA19323; Mon, 22 May 1995 08:19:13 +0600
Message-Id: <9505221319.AA19323@Enigma.metro2.k12.mn.us>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Mon, 22 May 1995 08:22:34 -0600
To: www-talk@www10.w3.org
From: johnj@enigma.metro2.k12.mn.us (John Jaberi)
Subject: Help with WWW clickable images(hot pages)
Content-Length: 1289


>Hi,
>
>        I am trying to make a clickable image page and I put this two line
>        in httpd.conf
>
>        Map     /img/*  /htbin/htimage/services/metro2-www/images/*
>        Exec    /htbin/*        /usr/local/WWW/bin/*  #htimage script located 
>            
>        here
>        
>        When I press on image ,then I get error calling htimage.
>        I tried to run htimage script ,I get :
>        Error Calling HTIAMGE
>        Neither PATH_INFO nor PATH_TRANSLATED env. variable not set       
>        
>        I would appreciated your comment.
>
>Thanks,








        ........................................
        . Jahan Jaberi                         .                  
        . Network Support Engineer             .                           
                                                            
        . MetroII                              .                        
        . Phone:(612) 686-2272                 .
        . Fax  :(612)686-5501                  .               
        . Internet:johnj@metro2.k12.mn.us      .
        . Http://www.Metro2.k12.mn.us/John.html.
        ........................................                           
                                                                           
        


eturn-Path: jonm@netscape.com 
Return-Path: <jonm@netscape.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AC08514; Mon, 22 May 1995 12:16:44 +0500
Received: from neon.netscape.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA06355; Mon, 22 May 1995 12:16:39 +0500
Received: from [198.93.92.20] (jonmmac.mcom.com [198.93.92.20]) by neon.netscape.com (950215.SGI.8.6.10/8.6.9) with SMTP id JAA10035; Mon, 22 May 1995 09:11:54 -0700
Date: Mon, 22 May 1995 09:11:54 -0700
X-Sender: jonm@pop.mcom.com
Message-Id: <v02110100abe603b0ad63@[198.93.92.20]>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
To: masinter@parc.xerox.com,
        Multiple recipients of list <www-talk@www10.w3.org>
From: jonm@netscape.com (Jon Mittelhauser)
Subject: Re: Byte ranges -- formal spec proposal
Content-Length: 1432

At 8:51 PM 5/21/95, Larry Masinter wrote:
>> I said:
>> and Carl von Loesch <Carl.von.Loesch@arbi.informatik.uni-oldenburg.de>
>>replied:
>
>> Ever tried to download netscape from overseas?
>
>> And no, I don't want to explain every netscape user how to download using
>> ftp and REGET, as it would circumvent any hope of having the stuff cached
>> by the WWW proxies.
>
>which might be a clever rejoinder, except that the proposal doesn't
>actually solve the problem. That is, this is not a proposal for
>finishing a partial transmission when the original transmission
>failed, and there doesn't seem to be any practical way to use URL byte
>ranges to implement retransmission after failure.

Because that is an implementation issue not a specification issue.  If this
specification was in place, I could easily detect an aborted transfer
from within Netscape.  If the partially loaded document was cached and it
was then requested again, I could only ask for the parts that I was
missing (if the server supported the function).

However, as was explained earlier, this wasn't the original intention of
Ari's proposal.  We need to be able to do byte ranges for a number of
reasons (the only publicly announced one being Adobe PDF support).  We,
obviously, would like to make the method of doing this a standard rather
than have 20 different implementations...

-Jon


Jon Mittelhauser (jonm@netscape.com)
Netscape Communications


eturn-Path: John_Calcote@novell.com 
Return-Path: <John_Calcote@novell.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA23931; Mon, 22 May 1995 16:57:11 +0500
Received: from gw.provo.Novell.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA09456; Mon, 22 May 1995 16:57:01 +0500
Received: from novell.com by gw.provo.Novell.com (4.1/SMI-4.1)
	id AA19833; Mon, 22 May 95 14:57:57 MDT
Received: from PRV-INTERNET-Message_Server by novell.com
	with Novell_GroupWise; Mon, 22 May 1995 14:57:57 -0600
Message-Id: <sfc0a674.090@novell.com>
X-Mailer: Novell GroupWise 4.1
Date: Mon, 22 May 1995 14:56:39 -0600
From: John Calcote <John_Calcote@novell.com>
To: www-talk@w3.org
Subject:  INCLUDES questions
Content-Length: 798

I'm attempting to implement a clean version of HTTPD on the NetWare
(4.x) OS.  I have a daemon up and running now, (the easy part), and I
have rudimentary (if temporary) file transfers working for playing with
the server.  Works pretty good if I do say so myself!

Before I dive in any deeper, there is a concept or two that I need to
understand before I start designing the URL parsing mechanism.

First, what is the magic mime type:  text/x-server-parsed-html

>From my perusing of the NCSA source, this is some sort of INCLUDE
functionality, but I can't decipher what INCLUDE is all about, can an HTML
document include others to be sent along inline?  If so, what is the
general algorithm for parsing the include list, that is, how to you know
what these other files are?

Thanks!
John Calcote


eturn-Path: sjr@merit.edu 
Return-Path: <sjr@merit.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AD10426; Mon, 22 May 1995 21:30:57 +0500
Received: from home.merit.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA11530; Mon, 22 May 1995 21:30:33 +0500
Received: from localhost (sjr@localhost) by home.merit.edu (8.6.12/merit-2.0) with SMTP id VAA19435; Mon, 22 May 1995 21:30:22 -0400
Message-Id: <199505230130.VAA19435@home.merit.edu>
To: jag@scndprsn.eng.sun.com (James Gosling)
Cc: connolly@beach.w3.org, john@math.nwu.edu, luotonen@netscape.com,
        www-talk@w3.org, http-wg@cuckoo.hpl.hp.com, uri@bunyip.com
Subject: Re: Byte ranges -- formal spec proposal 
In-Reply-To: Your message of Wed, 17 May 1995 17:50:00 +0800.
             <9505180050.AA28623@norquay.Eng.Sun.COM> 
Date: Mon, 22 May 1995 21:30:21 -0400
From: "Steven J. Richardson" <sjr@merit.edu>
Content-Length: 2019

  > Date:  Wed, 17 May 1995 17:50:00 +0800
  > From:  jag@scndprsn.eng.sun.com (James Gosling)
  > To:    connolly@beach.w3.org, john@math.nwu.edu
  > CC:    luotonen@netscape.com, www-talk@w3.org, http-wg@cuckoo.hpl.hp.com,
	   uri@bunyip.com

  > > According to Daniel W. Connolly:
  > > > 
  > > > A nice, clear, complete proposal. As you say, this could be done as a
  > > > server-private mechanism, but there's no reason why everybody
  > > > shouldn't do it the same way.
  > > > 
  > > > A couple nits:
  > > > 
  > > > >     * The first byte in file is byte number 1.
  > > > 
  > > > Blech. I'd rather it were 0. No biggie.
  > > > 
  > > 
  > > Base 0 is fine for bytes but would be problematic for other ranges.
  > > E.g. 
  > > 
  > > 	http://host/book;chapterrange=3-5
  > > 
  > > would mean chapters 4 to 6 if base 0 is used.  This would be just too
  > > confusing.  We thought it better to be consistent and use the same
  > > base for everything.
  > 
  > I don't think this is relevant: http should be kept simple and data-type
  > independent, leave out the higher level semantics.  Then 0 based
  > addressing is the most sensible.  Even for chapters, the argument is
  > weak: what chapter number is the title page?  What chapter number
  > applies to appendicies?  Does the number then need to be a string that
  > names a sub-entity?  This is a Pandora's Box that should stay closed.

  As long as that sort of query at the _user level_ maps exactly, this is
  fine with me.  (I think that something akin to the above "chapterrange=3-5" 
  should actually mean what it seems to, at that level.)

  > 
  > Any thought on how this should interect with dynamic computed documents
  > (CGI-bin scripts)?  Supporting range addressing of computed documents
  > would require either re-computation on each fetch, or caching.  If
  > re-computed, how do you guarantee consistancy?  Imagine fetching a
  > document one byte at a time that contains the server's load average.


  Steve Richardson/Merit
eturn-Path: Joel.Crisp%bristol.ac.uk@www10.w3.org 
Return-Path: <Joel.Crisp%bristol.ac.uk@www10.w3.org>
Received: from mintaka.lcs.mit.edu by www19 (5.0/NSCS-1.0S) 
	id AO19701; Wed, 24 May 1995 03:29:50 +0500
Received: from www10.w3.org by MINTAKA.LCS.MIT.EDU id aa09282;
          23 May 95 8:13 EDT
Received: from dira.bris.ac.uk by www10.w3.org (5.0/NSCS-1.0S) 
	id AA15853; Tue, 23 May 1995 08:09:08 +0500
Received: from www.ets.bris.ac.uk by dira.bris.ac.uk with SMTP (PP);
          Tue, 23 May 1995 13:07:31 +0100
Received: by www.ets.bris.ac.uk (5.0/SMI-SVR4)	id AA18280;
          Tue, 23 May 1995 13:03:30 +0000
Date: Tue, 23 May 1995 13:03:30 +0000
From: Joel Crisp <Joel.Crisp@bristol.ac.uk>
Message-Id: <9505231203.AA18280@www.ets.bris.ac.uk>
To: www-talk@www10.w3.org
Subject: Tutorial Markup Language
X-Sun-Charset: US-ASCII
Content-Length: 970

Hi

We have a large ammount of interest in delivering interactive question
and answer type tutorials using WWW.

To that end, we have defined a super-set of HTML3 ( Note : we have no 
intention of forcing this into the HTML3 spec ), for describing the
semantic content of questions. We support 4 types directly, but believe
the language is sufficiently flexible to support many more types.

This DTD, minimal ( and I mean it ) documentation, and a PERL cgi
script to actually run a tutorial dialogue are available from 
   <URL:ftp://ftp.ets.bris.ac.uk/ftp-inbound/tutorial.tar>
for anyone who is interested.

Comments are welcome, but bear in mind this is my first attempt at an
SGML DTD...

Joel Crisp
----------
Joel.Crisp@bris.ac.uk | ets-webmaster@bris.ac.uk
<URL:http://www.ets.bris.ac.uk/>
    Software Engineer 
    Educational Technology Service, The University of Bristol, UK
Opinions expressed within are mine alone, and may not reflect those
of my employer.
eturn-Path: @www10.w3.org:bsobilof@inet.ed.gov 
Received: from LCS.MIT.EDU (mintaka.lcs.mit.edu) by www19 (5.0/NSCS-1.0S) 
	id AB28556; Wed, 24 May 1995 05:08:45 +0500
Received: from www10.w3.org by MINTAKA.LCS.MIT.EDU id aa10522;
          23 May 95 9:13 EDT
Received: from inet.ed.gov by www10.w3.org (5.0/NSCS-1.0S) 
	id AA16193; Tue, 23 May 1995 09:08:43 +0500
Return-Path: <bsobilof@inet.ed.gov>
Message-Id: <v02120c02abe78c875dfc@[192.239.34.71]>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Date: Tue, 23 May 1995 09:10:09 -0400
To: www-talk@www10.w3.org
From: bsobilof@inet.ed.gov (Blake Sobiloff)
Subject: Re: INCLUDES questions
Content-Length: 1582

At 8:14 AM 5/22/95, John Calcote wrote:
>>From my perusing of the NCSA source, this is some sort of INCLUDE
>functionality, but I can't decipher what INCLUDE is all about, can an HTML
>document include others to be sent along inline?  If so, what is the
>general algorithm for parsing the include list, that is, how to you know
>what these other files are?

The following URL points to NCSA's documentation for the use of server-side
includes (from v1.4 of httpd, but it doesn't appear to have changed from
1.3):

        http://hoohoo.ncsa.uiuc.edu/docs-1.4/tutorials/includes.html

Pet peeve: NCSA's httpd suppresses the "Last-modified:" (as well as
"Content-length:" the last time I looked) field if it parses a document.
While this may have been the most expedient way to handle the issue of
"just how to you figure the last-modified date of a compound document," I
believe it to be pretty bogus for many documents. If the includes point to
other documents, it is likely that those other documents also have
last-modified dates associated with them. Simply returning the most recent
date solves this problem, plus allows caching browsers and proxy hosts to
cache the documents distributed by your server -- a big win from my side of
a 14.4K swizzle-stick. :-)

--
Blake Sobiloff                             <bsobilof@inet.ed.gov>
Internet Systems Analyst/Webmaster     (speaking only for myself)
Decision Systems Technologies, Inc.           Voice: 301/441-3377
Greenbelt, MD  20770  USA                       Fax: 301/441-4571
                  http://inet.ed.gov/~bsobilof/


eturn-Path: Somers%sisc.ucl.ac.be@www10.w3.org 
Return-Path: 1.0
Received: from mintaka.lcs.mit.edu by www19 (5.0/NSCS-1.0S) 
	id AO28556; Wed, 24 May 1995 05:26:00 +0500
Received: from www10.w3.org by MINTAKA.LCS.MIT.EDU id aa11550;
          23 May 95 10:21 EDT
Received: from ifdh.sc.ucl.ac.be by www10.w3.org (5.0/NSCS-1.0S) 
	id AA16690; Tue, 23 May 1995 10:17:04 +0500
Received: by ifdh.sc.ucl.ac.be (5.65/Ultrix3.0-C (AN/FYNU v%I%))
	id AA11478; Tue, 23 May 1995 16:16:59 +0200
Date: Tue, 23 May 1995 16:16:59 +0200
From: Somers@sisc.ucl.ac.be (Francis Somers)
Message-Id: <9505231416.AA11478@ifdh.sc.ucl.ac.be>
To: www-talk@w3.org
Subject: Re: Multiple language efforts?
Cc: Somers@sisc.ucl.ac.be
Content-Length: 3382

> Gvran Vberg writes:
>> Alan Batie writes:
>>> I'm wondering if anyone has other options for dealing with multiple language
>>> content management. The "obvious", but not very appealing, solutions are to
>>> have separate logical content trees, whether on the same server or not, and
>>> whether in parallel directory trees or in the same directories. A suggestion
>>> was recently made to me that it would be nice if the browsers could select a
>>> "preferred" language, and retrieve that if available, else retrieve the
>>> default. I know that something similar is already possible with respect to
>>> images, but I don't think it's really used --- the server doesn't send back
>>> the embedded images unless the browser asks for them anyhow. Thoughts?
>> 
>> I'm pretty sure you can do this with the cern server via
>> content-negotiation and the Accept-Language parameter. I don't have my
>> docs with me right now though, so I can't check. Anyone else have any
>> thoughts?
> I use language encoding of files at www.luth.se in a way that would let a
> language-aware client request a document in a specific language. It is
> done by having suffixes express the content-language. I don't know the
> CERN_httpd source code by heart by I recall it having quite extensive
> routines to really handle content and encoding negotiation.
> Today I use these suffixes primarily as a comfortable way to name different
> versions of a document. The language specification in the HTTP-reply is
> today only a nice side-effect as it seems to be ignored by all common clients.
> I won't go into detail about how it's done unless someone really wants
> me to. I recommend that anyone interested in this reads the documentation:
> <URL:http://www.w3.org/hypertext/WWW/Daemon/User/Config/Suffixes.html>

This is a naming solution, but it don't resolves the mutiple language problem
(perhaps I'am wrong ?)

If you have docA in french and english, docB only in french and
docC in french and english. From docA, you get docB and then docC.
Which is the language of the received docC ?

I have made a little change in the NCSA httpd server to help us.
When you choice a language (with href=URL?language=xx, where xx is
the ISO 639 code of the language), your client is registered on
the server (his IP address) with the choiced language. when you get 
a document, the server should give you the document in the right language
(the choice or the default if it doesn't exists).
example:
	http://www.ac.ucl.ac.be/LLN/vie.html?language=en
- you go to UCL (in the bottom) <exists only in french>
- then go to "La recherche"	-> you have the english version !!

The both versions have the same URL !
if the URL is http://myhost/path/file.html
the english file is /htpath/path/file-en.html
the french file is /htpath/path/file-fr.html

You can add a new language version without changing any anchor !!

But the best way in not to register the language of the client on the server ;
The client should be do a GET URL?language=fr,nl,en, and if the server
supports multi-language, it should give the fr version, if not, the nl
version, if not, the en version, if not the default version. But
this needs to change the client side ...

Ir. Francis Somers		e-mail: Somers@sisc.ucl.ac.be
University of Louvain		tel: +32-10-47.32.06
SC/SISC				fax: +32-10.45.21.83
Chemin du cyclotron,2		B-1348 Louvain-la-Neuve
eturn-Path: wlehotz%mail.coin.missouri.edu@www10.w3.org 
Return-Path: <wlehotz%mail.coin.missouri.edu@www10.w3.org>
Received: from mintaka.lcs.mit.edu by www19 (5.0/NSCS-1.0S) 
	id AV03356; Wed, 24 May 1995 06:46:49 +0500
Received: from www10.w3.org by MINTAKA.LCS.MIT.EDU id aa12618;
          23 May 95 11:47 EDT
Received: from mail (mail.coin.missouri.edu) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA17367; Tue, 23 May 1995 11:43:25 +0500
Received: from bigcat by mail (5.x/SMI-SVR4)
	id AA14002; Tue, 23 May 1995 10:43:06 -0500
Date: Tue, 23 May 1995 10:43:28 -0500 (CDT)
From: Willy Lehotz <wlehotz@mail.coin.missouri.edu>
X-Sender: wlehotz@bigcat
To: www-talk@www10.w3.org
Subject: NetScape error & Mosaic
Message-Id: <Pine.SUN.3.91.950523102129.8833A-100000@bigcat>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 1269

Greetings...

I have two problems/concerns.
	1. When users at the USDA\Consolidated Farm Service Agency
	   in Washington D.C. use Netscape v1.0, they recieve a
	   "malformed URL" error code. They first click on the USDA
	   Homepage<Http://www.usda.gov/>, then they goto the Agencies
	   and Programs. They click on the Consolidated Farm Service
	   Agency URL and sometimes receive the error and never get
	   our Homepage. 

	   Netscape has been used to view our pages from other sites and
	   no errors were received. What is a "Malformed URL" and what is 
	   happening in WDC?

	2. When using the NCSA Mosaic 2.04b browser some of our GIF files
	   do not display. We receive errors. The Organizational Photos
	   were scanned in from a high glossy newsletter and saved as 8 bit
	   B&W. The other GIFS are (I believe) 24 bit RGB. NCSA Mosaic 2.0
	   displays the GIFS just fine. What am I doing wrong? Our Agency is
	   also using Cello 1.1a and the GIFS dispay fine. 

Our WEB Server is running on a 486 PC under MSwindows 3.1 running 
Galaticomm's <Major BBS>. We shall beta test Gcomm's new World Group upgrade
in the next few weeks. It's Client/Server based.

Thank You for any and all comments

Willy Lehotz 
Mastech Systems Corp.
A05wlehotz@attmail.com

eturn-Path: nazgul@utopia.com 
Received: from www10.w3.org ([18.23.0.20]) by www19 (5.0/NSCS-1.0S) 
	id AA18892; Thu, 25 May 1995 09:09:57 +0500
Received: from utopia.com (utopia.wing.net) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA04433; Thu, 25 May 1995 09:09:50 +0500
Received: from [204.57.39.3] by utopia.com (8.6.12/SONY-4.0MX)
	id HAA22716; Thu, 25 May 1995 07:18:18 -0400
Return-Path: <nazgul@utopia.com>
X-Sender: nazgul@mailhost.utopia.com
Message-Id: <v0211010dabea156c4d7a@[204.57.39.3]>
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Date: Thu, 25 May 1995 07:20:24 -0400
To: gtn@ebt.com
From: nazgul@utopia.com (Kee Hinckley)
Subject: Re: Byte ranges (actually robots)
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Content-Length: 2134

At 8:46 AM 5/21/95, Gavin Nicol wrote:
>   http://www.ebt.com/collection/book/doc=3D1/chap=3D2/sect=3D3
>   http://www.ebt.com/collection/book/1/2/3
>   http://www.ebt.com/collection/book/1
...
>PS. I should note that the above naming scheme is very, very useful in
>our case, but it drives spiders wild....

That actually brings up an issue I've been meaning to mention somewhere
(the robots list would be appropriate, but I don't have time to join a
mailing list in order to post one issue).

We use a technology we call Dynamic View(tm) to present large amounts of
information in managable chunks, without creating a large hierarchy.  For
instance a mailing list might be broken into chunks by months, where you
would only see the details for the current month, and the rest would expand
when you click on the month name (see
http://www.utopia.com/mailings/edupage/ for an example).

This is great for people, however when a robot indexer comes to call I
don't want to confuse things. At best the structure will slow down the
indexing process. At worst it will look at a URL like
        http://www.utopia.com/mailings/edupage/?NAME=3DApril+95#April 95
and decide not to follow it at all.

So what to do? If we recognize the user-agent as a known robot, we give it
a flat presentation of the structure.  The catch of course, is that there
is no standard way to recognize a robot.  Some convention in the UserAgent
field would be sufficient, but....

Has anyone considered a move in that direction?  (Note, this also has a
bearing on URL-based shopping carts - I'd like the robot to be able to
browse the store without acquiring an ID, since you don't want the ID to
end up in some index somewhere. There of course, the problem is due to a
hack, so finding a solution is less critical if you're in a purist camp
:-).

Kee Hinckley      Utopia Inc. - Cyberspace Architects=81    617/721-6100
nazgul@utopia.com                               http://www.utopia.com/

I'm not sure which upsets me more: that people are so unwilling to accept
responsibility for their own actions, or that they are so eager to regulate
everyone else's.


eturn-Path: gtn@ebt.com 
Return-Path: <gtn@ebt.com>
Received: from www10.w3.org ([18.23.0.20]) by www19 (5.0/NSCS-1.0S) 
	id AB19150; Thu, 25 May 1995 09:18:54 +0500
Received: from ebt-inc.ebt.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA04494; Thu, 25 May 1995 09:18:44 +0500
Received: (from gtn@localhost) by ebt-inc.ebt.com (8.6.9/8.6.9) id JAA13703; Thu, 25 May 1995 09:21:10 -0400
Date: Thu, 25 May 1995 09:21:10 -0400
From: Gavin Nicol <gtn@ebt.com>
Message-Id: <199505251321.JAA13703@ebt-inc.ebt.com>
To: nazgul@utopia.com
Cc: www-talk@www10.w3.org
In-Reply-To: <v0211010dabea156c4d7a@[204.57.39.3]> (nazgul@utopia.com)
Subject: Re: Byte ranges (actually robots)
Content-Length: 329


>So what to do? If we recognize the user-agent as a known robot, we give it
>a flat presentation of the structure.  The catch of course, is that there
>is no standard way to recognize a robot.  Some convention in the UserAgent
>field would be sufficient, but....

Isn't there a standard permissions file that robots can fetch?

eturn-Path: nazgul@utopia.com 
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB21090; Thu, 25 May 1995 10:27:42 +0500
Received: from utopia.com (utopia.wing.net) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA04998; Thu, 25 May 1995 10:27:35 +0500
Received: from [204.57.39.3] by utopia.com (8.6.12/SONY-4.0MX)
	id KAA23062; Thu, 25 May 1995 10:25:37 -0400
Return-Path: <nazgul@utopia.com>
X-Sender: nazgul@mailhost.utopia.com
Message-Id: <v0211010fabea3bb44bee@[204.57.39.3]>
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Date: Thu, 25 May 1995 10:27:31 -0400
To: Gavin Nicol <gtn@ebt.com>
From: nazgul@utopia.com (Kee Hinckley)
Subject: Re: Byte ranges (actually robots)
Cc: www-talk@www10.w3.org
Content-Length: 1088

At 9:21 AM 5/25/95, Gavin Nicol wrote:
>>So what to do? If we recognize the user-agent as a known robot, we give it
>>a flat presentation of the structure.  The catch of course, is that there
>>is no standard way to recognize a robot.  Some convention in the UserAgent
>>field would be sufficient, but....
>
>Isn't there a standard permissions file that robots can fetch?

Yes, but that just let's me tell them what to stay away from. The issue
here is not that they should stay away, but that I should recognize them.
The permissions file is not fetched everytime they come around, and the
spacing between visits tends to be long (to reduce load on any one system),
and I can't just flag anyone who fetches robot.txt as a robot, since of
course a person may fetch it too.

Kee Hinckley      Utopia Inc. - Cyberspace Architects=81    617/721-6100
nazgul@utopia.com                               http://www.utopia.com/

I'm not sure which upsets me more: that people are so unwilling to accept
responsibility for their own actions, or that they are so eager to regulate
everyone else's.


eturn-Path: dmh@hpsgml.fc.hp.com 
Return-Path: <dmh@hpsgml.fc.hp.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AD22326; Thu, 25 May 1995 11:04:35 +0500
Received: from hpfcla.fc.hp.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA05405; Thu, 25 May 1995 11:04:25 +0500
Received: from hpsgml.fc.hp.com by hpfcla.fc.hp.com with ESMTP
	(1.37.109.16/15.5+IOS 3.20) id AA242204257; Thu, 25 May 1995 09:04:18 -0600
Received: from localhost by hpsgml.fc.hp.com with SMTP
	(1.37.109.16/15.5+IOS 3.22) id AA161874163; Thu, 25 May 1995 09:02:43 -0600
Message-Id: <199505251502.AA161874163@hpsgml.fc.hp.com>
To: www-talk@www10.w3.org
Cc: dmh@hpsgml.fc.hp.com
Subject: Browser Displayed URL
Date: Thu, 25 May 1995 09:02:42 -0600
From: Dave Hollander <dmh@hpsgml.fc.hp.com>
Content-Length: 1400


Hi,

I have talked to a few browser developers about this issue and 
would also like to get the opinions of this list. The question
is - what should be the known URL for any web node?

For example, if a node has this base statement: 
	<base href="http://www.hp.com/go/AccessGuide">

while the actual file location for this node is: 
	{htdocs}/AccessGuide/AccessGuide.html. 

and most readers access this node via an imagemap redirect:
	"rect /AccessGuide/AccessGuide.html 56,364  162,393"

When the node is reached, what location should browsers display 
to the user and use in hot lists?
  http://www.hp.com/cgi-bin/imagemap/ahp/ahpAccessHP.map?106,379
   or
  http://www.hp.com/AccessGuide/AccessGuide.html
   or
  http://www.hp.com/go/AccessGuide

I have seen browsers use all three forms: 
	how I got there, 
	where I was redirected, 
	what the node says it is.


This is real important to me (and I suspect the web community) because 
we wish to help users maintain their links to our material. Clearly, 
the first form is the hardest for me to maintain.  All URLS in the 
document are server relative so there is no issue about relative paths.

It seems to me that the known location for a node should be under 
control of the server/document author not the browser/reader.  If 
not, how is a developer going to help readers maintain links to 
the developer's site? 

Regards,
Dave Hollander

eturn-Path: dnew@sgf.fv.com 
Return-Path: <dnew@sgf.fv.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB23539; Thu, 25 May 1995 11:48:31 +0500
Received: from sgf.fv.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA05771; Thu, 25 May 1995 11:47:59 +0500
Received: by sgf.fv.com (Smail3.1.28.1 #52)
	id m0sEf4z-000GruC; Thu, 25 May 95 11:44 EDT
Date: Thu, 25 May 1995 11:44:08 +0100
From: Darren New <dnew@sgf.fv.com>
Subject: Re: Byte ranges (actually robots)
To: Kee Hinckley <nazgul@utopia.com>
Cc: Multiple recipients of list <www-talk@www10.w3.org>
In-Reply-To: <v0211010fabea3bb44bee@[204.57.39.3]>
Message-Id: <Pine.3.89.9505251146.B1315-0100000@sgf.fv.com>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 203

> and I can't just flag anyone who fetches robot.txt as a robot, since of
> course a person may fetch it too.

And of course not all robots will fetch robot.txt, either.  Welcome to 
the internet!  :-)

eturn-Path: gtn@ebt.com 
Return-Path: <gtn@ebt.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA23718; Thu, 25 May 1995 11:54:54 +0500
Received: from ebt-inc.ebt.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA05841; Thu, 25 May 1995 11:54:45 +0500
Received: (from gtn@localhost) by ebt-inc.ebt.com (8.6.9/8.6.9) id LAA21338; Thu, 25 May 1995 11:57:15 -0400
Date: Thu, 25 May 1995 11:57:15 -0400
From: Gavin Nicol <gtn@ebt.com>
Message-Id: <199505251557.LAA21338@ebt-inc.ebt.com>
To: nazgul@utopia.com
Cc: www-talk@www10.w3.org
In-Reply-To: <v0211010fabea3bb44bee@[204.57.39.3]> (nazgul@utopia.com)
Subject: Re: Byte ranges (actually robots)
Content-Length: 406

>>Isn't there a standard permissions file that robots can fetch?
> 
>Yes, but that just let's me tell them what to stay away from. The issue
>here is not that they should stay away, but that I should recognize them.

Fair enough. I'd also like to recognise them, and some kind of
convention in User-Agent or somesuch seems desirable. 

Putting a robot on our server is a great way of stress-testing it ;-)
eturn-Path: jmeritt@smtpinet.aspensys.com 
Return-Path: <jmeritt@smtpinet.aspensys.com>
Received: from dxmint.cern.ch by www19 (5.0/NSCS-1.0S) 
	id AA24636; Thu, 25 May 1995 12:30:13 +0500
Received: from www0.cern.ch by dxmint.cern.ch
	id AA28922; Thu, 25 May 1995 18:30:09 +0200
Received: from dxmint.cern.ch by www0.cern.ch (5.0/SMI-4.0)
	id AA26927; Thu, 25 May 1995 18:30:04 --100
Received: from aspensys by dxmint.cern.ch
	id AA28898; Thu, 25 May 1995 18:30:03 +0200
Received: from smtpinet.aspensys.com by aspensys (5.0/SMI-SVR4)
	id AA25122; Thu, 25 May 1995 12:27:39 +0500
Received: from cc:Mail by smtpinet.aspensys.com
	id AA801430289 Thu, 25 May 95 12:31:29 EST
Date: Thu, 25 May 95 12:31:29 EST
From: jmeritt@smtpinet.aspensys.com (Meritt, Jim)
Message-Id: <9504258014.AA801430289@smtpinet.aspensys.com>
To: www-talk@www0.cern.ch, wais-talk@wais.com
Subject: Mosaic/WAIS
Content-Length: 186

     The Comet (mosaic) has interfaces to http (duh!), gopher, ftp, mail, 
     and news that I know about & use.  Is there a WAIS interface hidden in 
     there?
     
     Jim Meritt
eturn-Path: lentz@annie.astro.nwu.edu 
Return-Path: <lentz@annie.astro.nwu.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AB27655; Thu, 25 May 1995 14:27:14 +0500
Received: from relay.acns.nwu.edu (nwu.edu) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA07234; Thu, 25 May 1995 14:27:10 +0500
Received: from annie.astro.nwu.edu by relay.acns.nwu.edu with SMTP
	(1.37.109.16/20.3) id AA190876533; Thu, 25 May 1995 13:28:53 -0500
Received: by annie.astro.nwu.edu; (5.65/1.1.8.2/15Jul94-0806PM)
	id AA20087; Thu, 25 May 1995 13:27:07 -0500
From: Robert A. Lentz <lentz@annie.astro.nwu.edu>
Message-Id: <9505251827.AA20087@annie.astro.nwu.edu>
Subject: Re: Browser Displayed URL
To: www-talk@www10.w3.org
Date: Thu, 25 May 1995 13:27:07 -0500 (CDT)
In-Reply-To: <199505251502.AA161874163@hpsgml.fc.hp.com> from "Dave Hollander" at May 25, 95 12:17:55 pm
Reply-To: lentz@annie.astro.nwu.edu
X-Mailer: ELM [version 2.4 PL24alpha3]
Mime-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 1050      

> 
> For example, if a node has this base statement: 
> 	<base href="http://www.hp.com/go/AccessGuide">
> 
> while the actual file location for this node is: 
> 	{htdocs}/AccessGuide/AccessGuide.html. 
> 
> and most readers access this node via an imagemap redirect:
> 	"rect /AccessGuide/AccessGuide.html 56,364  162,393"
> 
> When the node is reached, what location should browsers display 
> to the user and use in hot lists?
>   http://www.hp.com/cgi-bin/imagemap/ahp/ahpAccessHP.map?106,379
>    or
>   http://www.hp.com/AccessGuide/AccessGuide.html
>    or
>   http://www.hp.com/go/AccessGuide
> 
> I have seen browsers use all three forms: 
> 	how I got there, 
> 	where I was redirected, 
> 	what the node says it is.

The last is the most important, being the author's statement of how they
wish the document referred to as; very important when copying the URL.

-Robert
-- 
lentz@annie.astro.nwu.edu            http://www.astro.nwu.edu/lentz/plan.html
	"You have to push as hard as the age that pushes against you."
					-Flannery O'Connor
eturn-Path: sjk@amazon.com 
Return-Path: <sjk@amazon.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA09607; Thu, 25 May 1995 22:00:27 +0500
Received: from bert.amazon.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA11633; Thu, 25 May 1995 22:00:18 +0500
Received: (from sjk@localhost) by bert.amazon.com (8.6.12/8.6.9) id SAA09172; Thu, 25 May 1995 18:57:40 -0700
Date: Thu, 25 May 1995 18:57:40 -0700
From: Shel Kaphan <sjk@amazon.com>
Message-Id: <199505260157.SAA09172@bert.amazon.com>
To: www-talk@w3.org
Subject: caching dilemma
Content-Length: 4106



Hi,

I have some problems regarding caching that I hope someone here will
have some good suggestions how to solve.  I've already read the
discussion on http://www.ics.uci.edu/WebSoft/caching.html, and it
didn't really help very much.

The problem is that the current round of browsers and the
current protocol do not seem to interact very smoothly when it comes
to caching pages.  I'll explain what I mean:

The "expires" feature should cover the issue of when pages should be
flushed, but the world is apparently not ready for it, because:

- If you set documents to expire immediately, some major browsers
display "Data Missing" or equivalently scary messages when you use
browser commands to "back up" to that page.  Since many users are not
going to understand what is going on and will be confused by such
messages, and may not know to "reload" the page at that point, it
would be better for them never to see messages like that.  (I've
already had problems with some naive beta testers tripping over that.
They tend to think something must have broken.  You can't argue that
we need more sophisticated users, because we don't have a choice!)

- Some browsers (such as Prodigy's) appear to ignore the "expires" header 
and cache pages anyway.  (and that's just their *browser*...)

So, I have a question and I have suggestions.

First, the question:

Is there any good workaround for the current problem, that would have
the properties of:
- forcing browsers to reload expired pages when someone explicitly requests
  one, and
- either:
  - allowing pages on the browser's history stack (for instance) to remain
    in the local cache even if they are expired, or, 
  - *somehow* causing the browsers to gracefully and silently reload 
    expired pages when re-visited through history mechanisms.

No?  I suspected as much...

The suggestions:

To make the web work more smoothly, it would be nice if browsers would
handle this situation more gracefully, by, for instance, not displaying
errors like "Data Missing", but just automatically reloading the page.

However, I also think it is worth considering for browser writers that
history stacks (that can be re-viewed with browser navigation
controls) are in a class of their own when it comes to caching.
However, while it might make sense to back up and see an expired
document, since history mechanisms are for "history", it does not make
sense to go through a link and see a cached copy of an expired
document.  It is REALLY BAD for browsers to display cached copies of
expired documents when they are meant to be freshly displayed in
response to a direct user command, because a URL may be a request to a
program that is displaying dynamic information related to the user's
extended "session" with the server.  (This is the core of the issue).

I realize these considerations may have no role in the HTTP spec,
however I feel there are serious problems in this area, which can only
be resolved by coordinating the behavior of browsers and servers.

Another thing that might help: perhaps there should be a way for
servers to "force" the URL (the *name*) handled by clients to something other
than the requested URL.  This would allow, for example, the
requestor's URL to be used to encode information relating to a query,
but would then result in a single cache entry in the client.

To explain this a little more, if there were two GET requests, one for
/cgi-bin/food/hamburgers and one for /cgi-bin/food/french-fries, which
would result in a single page that ought to be cached as one page,
then the server ought to be able to say, "you asked for
/food/french-fries, but the page is called /food/generic-junk-food",
and to have the browser use that info to uniquely identify a cache
entry and update it with the newly fetched data.  This might not help
to avoid fetching documents extra times, but it would help on cache
coherence if the intent was to display a dynamically generated document.

Anyway, just some thoughts.  If you have any ideas, pointers or references
for me, I would really appreciate it.

--Shel Kaphan
  sjk@amazon.com
eturn-Path: nazgul@utopia.com 
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA13928; Fri, 26 May 1995 01:26:29 +0500
Received: from utopia.com (utopia.wing.net) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA13118; Fri, 26 May 1995 01:26:24 +0500
Received: from [204.57.39.3] by utopia.com (8.6.12/SONY-4.0MX)
	id BAA02255; Fri, 26 May 1995 01:24:22 -0400
Return-Path: <nazgul@utopia.com>
X-Sender: nazgul@mailhost.utopia.com
Message-Id: <v02110132abeb0c38377f@[204.57.39.3]>
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Date: Fri, 26 May 1995 01:26:18 -0400
To: sjk@amazon.com
From: nazgul@utopia.com (Kee Hinckley)
Subject: Re: caching dilemma
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Content-Length: 2469

At 10:20 PM 5/25/95, Shel Kaphan wrote:
>To make the web work more smoothly, it would be nice if browsers would
>handle this situation more gracefully, by, for instance, not displaying
>errors like "Data Missing", but just automatically reloading the page.

Automatic reloading of a page in my history stack seems rather
user-unfriendly. I expect history loading to be fast and not go off over
the net. I guess I could see it as a user-specified option, but...

>document.  It is REALLY BAD for browsers to display cached copies of
>expired documents when they are meant to be freshly displayed in
>response to a direct user command, because a URL may be a request to a

There I agree.

>program that is displaying dynamic information related to the user's
>extended "session" with the server.  (This is the core of the issue).
>
>I realize these considerations may have no role in the HTTP spec,
>however I feel there are serious problems in this area, which can only
>be resolved by coordinating the behavior of browsers and servers.
>
>Another thing that might help: perhaps there should be a way for
>servers to "force" the URL (the *name*) handled by clients to something oth=
er
>than the requested URL.  This would allow, for example, the
>requestor's URL to be used to encode information relating to a query,
>but would then result in a single cache entry in the client.
>
>To explain this a little more, if there were two GET requests, one for
>/cgi-bin/food/hamburgers and one for /cgi-bin/food/french-fries, which
>would result in a single page that ought to be cached as one page,
>then the server ought to be able to say, "you asked for
>/food/french-fries, but the page is called /food/generic-junk-food",
>and to have the browser use that info to uniquely identify a cache
>entry and update it with the newly fetched data.  This might not help
>to avoid fetching documents extra times, but it would help on cache
>coherence if the intent was to display a dynamically generated document.
>
>Anyway, just some thoughts.  If you have any ideas, pointers or references
>for me, I would really appreciate it.
>
>--Shel Kaphan
>  sjk@amazon.com

Kee Hinckley      Utopia Inc. - Cyberspace Architects=81    617/721-6100
nazgul@utopia.com                               http://www.utopia.com/

I'm not sure which upsets me more: that people are so unwilling to accept
responsibility for their own actions, or that they are so eager to regulate
everyone else's.


eturn-Path: S.Nisbet@doc.mmu.ac.uk 
Return-Path: <S.Nisbet@doc.mmu.ac.uk>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA17353; Fri, 26 May 1995 04:11:17 +0500
Received: from ehlana.mmu.ac.uk by www10.w3.org (5.0/NSCS-1.0S) 
	id AA14083; Fri, 26 May 1995 04:11:12 +0500
Received: from patsy.doc.aca.mmu.ac.uk by ehlana with SMTP (PP);
          Fri, 26 May 1995 09:10:29 +0100
Received: from raphael.doc.aca.mmu.ac.uk 
          by patsy.doc.aca.mmu.ac.uk (4.1/SMI-4.1)	id AA08160;
          Fri, 26 May 95 09:09:18 BST
Date: Fri, 26 May 95 09:09:18 BST
From: Steve Nisbet <S.Nisbet@doc.mmu.ac.uk>
Message-Id: <9505260809.AA08160@patsy.doc.aca.mmu.ac.uk>
To: www-talk@www10.w3.org
Subject: WEB Info
Content-Length: 1197

I have two points I would like to see some bilateral info on:

Firstly, for my own research, I am wondering if anyone knows what the share
of the platform market is for web software, ie. does anyone know whether there
are more UNIX machines running webs than PC's etc.

Secondly, I am experiencing a lot of hostile censorship and power battles at
my local site, where one autonomous section of the University seeks to hold
all other web servers on the site in thrall, standardising everything and
putting in place .
I quote
'A user hostile front end'.
then they are going to firewall the whole thing.

I may be wrong but I get the impression that this would make us somewhat
unique in the internet world, as the one institution refusing to talk to
anyone on equal terms, in fact, more or less refusing to talk to anyone.
I was wondering if there are opinions on this, or if similar situations
occur elsewhere. To me it seems to be at extreme odds with the philosophy of
the web and I (and others) are quite upset that this kind of Draconian attitude
persists in an advancing world of open communication.

Comments?

Yours in his politically sound bunker
Steve Nisbet
http://www.doc.mmu.ac.uk/
 


eturn-Path: Somers@sisc.ucl.ac.be 
Return-Path: <Somers@sisc.ucl.ac.be>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AZ19449; Fri, 26 May 1995 05:27:22 +0500
Received: from ifdh.sc.ucl.ac.be by www10.w3.org (5.0/NSCS-1.0S) 
	id AA14620; Fri, 26 May 1995 05:26:37 +0500
Received: from localhost by ifdh.sc.ucl.ac.be (5.65/Ultrix3.0-C (AN/FYNU v%I%))
	id AA04448; Fri, 26 May 1995 11:26:35 +0200
Message-Id: <9505260926.AA04448@ifdh.sc.ucl.ac.be>
To: www-talk@www10.w3.org
Subject: Re: Multiple language efforts?
Date: Fri, 26 May 95 11:26:34 +0200
From: Somers@sisc.ucl.ac.be
X-Mts: smtp
Content-Length: 3386

> Gvran Vberg writes:
>> Alan Batie writes:
>>> I'm wondering if anyone has other options for dealing with multiple language
>>> content management. The "obvious", but not very appealing, solutions are to
>>> have separate logical content trees, whether on the same server or not, and
>>> whether in parallel directory trees or in the same directories. A suggestion
>>> was recently made to me that it would be nice if the browsers could select a
>>> "preferred" language, and retrieve that if available, else retrieve the
>>> default. I know that something similar is already possible with respect to
>>> images, but I don't think it's really used --- the server doesn't send back
>>> the embedded images unless the browser asks for them anyhow. Thoughts?
>> 
>> I'm pretty sure you can do this with the cern server via
>> content-negotiation and the Accept-Language parameter. I don't have my
>> docs with me right now though, so I can't check. Anyone else have any
>> thoughts?
> I use language encoding of files at www.luth.se in a way that would let a
> language-aware client request a document in a specific language. It is
> done by having suffixes express the content-language. I don't know the
> CERN_httpd source code by heart by I recall it having quite extensive
> routines to really handle content and encoding negotiation.
> Today I use these suffixes primarily as a comfortable way to name different
> versions of a document. The language specification in the HTTP-reply is
> today only a nice side-effect as it seems to be ignored by all common clients.
> I won't go into detail about how it's done unless someone really wants
> me to. I recommend that anyone interested in this reads the documentation:
> <URL:http://www.w3.org/hypertext/WWW/Daemon/User/Config/Suffixes.html>

This is a naming solution, but it don't resolves the mutiple language problem
(perhaps I'am wrong ?)

If you have docA in french and english, docB only in french and
docC in french and english. From docA, you get docB and then docC.
Which is the language of the received docC ?

I have made a little change in the NCSA httpd server to help us.
When you choice a language (with href=URL?language=xx, where xx is
the ISO 639 code of the language), your client is registered on
the server (his IP address) with the choiced language. when you get 
a document, the server should give you the document in the right language
(the choice or the default if it doesn't exists).
example:
	http://www.ac.ucl.ac.be/LLN/vie.html?language=en
- - you go to UCL (in the bottom) <exists only in french>
- - then go to "La recherche"	-> you have the english version !!

The both versions have the same URL !
if the URL is http://myhost/path/file.html
the english file is /htpath/path/file-en.html
the french file is /htpath/path/file-fr.html

You can add a new language version without changing any anchor !!

But the best way in not to register the language of the client on the server ;
The client should be do a GET URL?language=fr,nl,en, and if the server
supports multi-language, it should give the fr version, if not, the nl
version, if not, the en version, if not the default version. But
this needs to change the client side ...

Ir. Francis Somers		e-mail: Somers@sisc.ucl.ac.be
University of Louvain		tel: +32-10-47.32.06
SC/SISC				fax: +32-10.45.21.83
Chemin du cyclotron,2		B-1348 Louvain-la-Neuve
eturn-Path: jim.warwick@analysys.co.uk 
Return-Path: <jim.warwick@analysys.co.uk>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA22223; Fri, 26 May 1995 07:10:37 +0500
Received: from eagle.analysys.co.uk by www10.w3.org (5.0/NSCS-1.0S) 
	id AA15272; Fri, 26 May 1995 07:10:29 +0500
Date: Fri, 26 May 95 11:40:28 BST
Message-Id: <9505261040.AA12098@eagle.analysys.co.uk>
Received: from djw-computer (partridge.analysys.co.uk) by eagle.analysys.co.uk (4.1) id AA12098; Fri, 26 May 95 11:40:28 BST
X-Sender: djw@eagle
X-Mailer: Windows Eudora Version 1.4.4
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
To: www-talk@www10.w3.org
From: jim.warwick@analysys.co.uk (Jim Warwick)
Subject: Proxy server interaction with password protected URLs?
Content-Length: 1141

I've been setting up URL's that refer to FTP accounts which require a
password, the URL's are of the form (for example) :

ftp://user@ftp.race.analysys.co.uk/user_dir/

This allows us to build a nice interface to files which require a simple
password to retrieve them.  On selecting this URL, the client prompt's the
user for their password, and then lets them into the FTP server.

This works well, but some users never get prompted for the password.  This
appears to be the case when they are accessing our site through a proxy
server.  In this case, they just get the reply :

Can't Access Document: ftp://user@ftp.race.analysys.co.uk/user_dir/
Reason: FTP-server replies: User user cannot log in.

I know very little about proxy servers, but this worries me.
Will they block this kind of request by default, or is it a configuration
issue at the proxy server end?
Is this a general issue with accesses that require user-id and passwords, or
is it only an issue for this type of FTP url?

Jim Warwick @ Analysys

PS: We're running the EMWAC HTTP server on a NT 3.5 platform and also using
the shipped MS FTP server on the same machine.


eturn-Path: dnew@sgf.fv.com 
Return-Path: <dnew@sgf.fv.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA24693; Fri, 26 May 1995 08:43:01 +0500
Received: from sgf.fv.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA15793; Fri, 26 May 1995 08:42:51 +0500
Received: by sgf.fv.com (Smail3.1.28.1 #52)
	id m0sEygn-000GrqC; Fri, 26 May 95 08:40 EDT
Date: Fri, 26 May 1995 08:40:29 +0100
From: Darren New <dnew@sgf.fv.com>
Subject: Re: Byte ranges -- formal spec proposal 
To: Marc VanHeyningen <marcvh@spry.com>
Cc: Multiple recipients of list <www-talk@www10.w3.org>
In-Reply-To: <15453.800832484@pellet.spry.com>
Message-Id: <Pine.3.89.9505260820.A369-0100000@sgf.fv.com>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 747

> Well, the most common format, audio/basic, has header information which
> contains magic numbers and the Hz and stuff like that, and technically is
> not a legal audio/basic file without them, though many players will tolerate
> this.  I would put this in the same category as HTML; arbitrary segments
> probably are not legal, but viewers might be able to tolerate them.

Huh?  Audio/basic is what ISDN uses.  It has no headers at all.

>From RFC1521:
   The content of the "audio/basic" subtype is audio encoded using 8-bit
   ISDN mu-law [PCM].  When this subtype is present, a sample rate of
   8000 Hz and a single channel is assumed.

If this doesn't mean the format is "raw", please let me know where the 
headers are defined.  --Darren


eturn-Path: davis@dri.cornell.edu 
Return-Path: <davis@dri.cornell.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AC25018; Fri, 26 May 1995 08:54:27 +0500
Received: from willow.tc.cornell.edu by www10.w3.org (5.0/NSCS-1.0S) 
	id AA15864; Fri, 26 May 1995 08:54:24 +0500
Received: by willow.tc.cornell.edu id AA10660
  (5.65c/IDA-1.4.4 for www-talk@www10.w3.org); Fri, 26 May 1995 08:54:23 -0400
Date: Fri, 26 May 1995 08:54:23 -0400
From: Jim Davis <davis@dri.cornell.edu>
Message-Id: <199505261254.AA10660@willow.tc.cornell.edu>
To: www-talk@www10.w3.org
Subject: [aw@bae.bellcore.com: Netscape to incorporate Java]
Content-Length: 352

  To: hotjava-interest@java.eng.sun.com
  Subject: Netscape to incorporate Java
  Date: Thu, 25 May 95 17:19:07 -0400
  From: Andrew Wason <aw@bae.bellcore.com>

  Netscape announced Tuesday that they have licensed Java from Sun
  and will incorporate it into the Netscape browser.

  See http://sfgate.com/examiner/new/stories/61.html for the story.

eturn-Path: nazgul@utopia.com 
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA28424; Fri, 26 May 1995 10:46:18 +0500
Received: from utopia.com (utopia.wing.net) by www10.w3.org (5.0/NSCS-1.0S) 
	id AA17040; Fri, 26 May 1995 10:46:08 +0500
Received: from [204.57.39.3] by utopia.com (8.6.12/SONY-4.0MX)
	id KAA03067; Fri, 26 May 1995 10:42:27 -0400
Return-Path: <nazgul@utopia.com>
X-Sender: nazgul@mailhost.utopia.com
Message-Id: <v02110137abeb98ea4596@[204.57.39.3]>
Mime-Version: 1.0
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
Date: Fri, 26 May 1995 10:44:24 -0400
To: dnew@sgf.fv.com
From: nazgul@utopia.com (Kee Hinckley)
Subject: Re: Byte ranges -- formal spec proposal
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Content-Length: 1012

At 8:44 AM 5/26/95, Darren New wrote:
>> Well, the most common format, audio/basic, has header information which
>> contains magic numbers and the Hz and stuff like that, and technically is
>> not a legal audio/basic file without them, though many players will toler=
ate
>> this.  I would put this in the same category as HTML; arbitrary segments
>> probably are not legal, but viewers might be able to tolerate them.
>
>Huh?  Audio/basic is what ISDN uses.  It has no headers at all.

Technically you are correct. However there are a large number of viewers
out there (on all platforms) which are used for audio/basic, but which
choke if handed anything that doesn't have a Sun AU header on it.

Kee Hinckley      Utopia Inc. - Cyberspace Architects=81    617/721-6100
nazgul@utopia.com                               http://www.utopia.com/

I'm not sure which upsets me more: that people are so unwilling to accept
responsibility for their own actions, or that they are so eager to regulate
everyone else's.


eturn-Path: worasuch@mail.cs.orst.edu 
Return-Path: <worasuch@mail.cs.orst.edu>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA03085; Fri, 26 May 1995 13:15:15 +0500
Received: from python.CS.ORST.EDU by www10.w3.org (5.0/NSCS-1.0S) 
	id AA18493; Fri, 26 May 1995 13:15:13 +0500
Received: from viper.CS.ORST.EDU (viper.CS.ORST.EDU [128.193.36.50]) by python.CS.ORST.EDU (8.6.9/8.6.9) with ESMTP id KAA00197 for <www-talk@www10.w3.org>; Fri, 26 May 1995 10:15:04 -0700
Received: (worasuch@localhost) by viper.CS.ORST.EDU (8.6.8/8.6.4) id KAA04589; Fri, 26 May 1995 10:15:04 -0700
Date: Fri, 26 May 1995 10:15:03 -0700 (PDT)
From: Chukiat Worasucheep <worasuch@mail.cs.orst.edu>
Subject: Regarding HTTP/1.0 specification
To: Multiple recipients of list <www-talk@www10.w3.org>
In-Reply-To: <Pine.3.89.9505260820.A369-0100000@sgf.fv.com>
Message-Id: <Pine.3.89.9505261015.B4560-0100000@viper.CS.ORST.EDU>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Content-Length: 569


I have some questions regarding specification of HTTP/1.0, dated March 8, 
1995.  Does anybody help me for more clarafications?

1)  LINK in request message:
One link end-point is defined by Request-URI, and where is the other 
end-point defined?

2)  Content-Language:
When not specified, the default is that intended for all language 
audience.  Do you mean enlish (en)?

For an entity that has more than 1 Content-Language, says 2, how is the
data sent?  In 2 set of entitties, each of which in each language, but 
representing to the same data?

Thanks in advance
eturn-Path: sjk@amazon.com 
Return-Path: <sjk@amazon.com>
Received: from www10.w3.org by www19 (5.0/NSCS-1.0S) 
	id AA03682; Fri, 26 May 1995 13:32:31 +0500
Received: from bert.amazon.com by www10.w3.org (5.0/NSCS-1.0S) 
	id AA18646; Fri, 26 May 1995 13:32:21 +0500
Received: (from sjk@localhost) by bert.amazon.com (8.6.12/8.6.9) id KAA10368; Fri, 26 May 1995 10:28:20 -0700
Date: Fri, 26 May 1995 10:28:20 -0700
From: Shel Kaphan <sjk@amazon.com>
Message-Id: <199505261728.KAA10368@bert.amazon.com>
To: nazgul@utopia.com
Cc: Multiple recipients of list <www-talk@www10.w3.org>
Subject: Re: caching dilemma
In-Reply-To: <v02110132abeb0c38377f@[204.57.39.3]>
References: <v02110132abeb0c38377f@[204.57.39.3]>
Content-Length: 1333

Kee Hinckley writes:
 > At 10:20 PM 5/25/95, Shel Kaphan wr