W3C home > Mailing lists > Public > w3c-sgml-wg@w3.org > November 1996

Re: Recent ERB votes

From: <lee@sq.com>
Date: Fri, 8 Nov 96 12:59:28 EST
Message-Id: <9611081759.AA03778@sqrex.sq.com>
To: W3C-SGML-WG@w3.org, papresco@calum.csclub.uwaterloo.ca
> From: Paul Prescod <papresco@calum.csclub.uwaterloo.ca>

> If 
>  a)Panorama were a standard part of Netscape or IE, 
>  b)it had display mechanisms were based on standards (i.e. DSSSL-O), 
>  c)it had performance on par with regular web browsers, 
> 
> then XML would be irrelevant.

I wish I had known that the entire reason for XML was a performance issue
in Panorama.  It would have saved a lot of people a lot of work.
If we released a Netscape plugin version of Panorama (that also worked in
MSIE), included DSSSL-O support, and made it faster, XML would be a
waste of time?

I don't actually agree with this, by the way...

> Simple SGML would still be important, but that
> would be an ISO project, not a W3C one. XML should be ubiquitous (and thus
> simple), standards-based, and fast (and thus simple).

It will only be ubiquitous if lots of tools support it.

 
> >I agree.  Of course, empty elements aren't related to that.
> >And neither is downloading SGML databases over the internet...
> 
> They are related to that, through simplicity. If we can make parsing
> simpler, we will get better browser support, and thus ubiquity. We will also
> get better performance.
True, I concede that point -- which is why I wanted to use a form other
than <e> or <e a=value> for all EMPTY element.  I don't see how having
a fixed list of XML elements that are EMPTY for all XML documents, such
as BR, HR, IMG, META, LINK and so forth makes things easier.  Your TEI
documents had better not use any of those GIs except as EMPTY elemnts...

> >Then why wire in a list of HTML "du jour" EMPTY elements into
> >XML perpetuity?  The same program that converts today's HTML into XML
> >can change <BR> to <BR/> if it likes.
> 
> Because the ERB is _!NOT!_ interested in compatibility with HTML _documents_
> but with EXISTING HTML BROWSERS! They want to be able to do this, in the
> short term:

What you show does not contain any elements, and for all your shouting :-),\
does not answer my question.

> 
> <P>This is an <ABBREV>HTML-like</ABBREV> document with <ABBREV>TEI<ABBREV>
> style <gi>date</gi> elements, written on <date value='1980-02-21'>21 Feb
> 1980</date>. It also has a <ABBREV>TEI<ABBREV> header (though it is not
> shown here). Thanks to the magic of <ABBREV>CSS</ABBREV> style sheets, and
> <ABBREV>HTML</ABBREV> convention, the <ABBREV>TEI</ABBREV> tags can either
> disappear, or be mapped to special formatting for existing browsers.

RFC 1866 says that unrecognised elements are ignored.
A CSS-based browser must treat all those elements as if they were not there.
It is not allowed to let you assocuate styles with them, and, if it
does, is buggy (albeit useful).

Lee
Received on Friday, 8 November 1996 12:59:48 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 10:03:42 EDT