W3C home > Mailing lists > Public > ietf-http-wg-old@w3.org > September to December 1998

Re: Netscape 4.5 and HTTP/1.1 Accept-Encoding

From: David W. Morris <dwm@XPASC.COM>
Date: Tue, 3 Nov 1998 12:06:25 -0800 (PST)
To: John Franks <john@math.nwu.edu>
Cc: Dave Kristol <dmk@research.bell-labs.com>, http-wg@hplb.hpl.hp.com
Message-Id: <Pine.GSO.3.96.981103120207.19335B-100000@shell1.aimnet.com>


On Tue, 3 Nov 1998, John Franks wrote:

> On Tue, 3 Nov 1998, Dave Kristol wrote:
> 
> I  think gzip and compress are different encodings.  The fact that
> a program called "gzip" understands both is not relevant.

me too ...

> 
> > 
> > 2) My server should not send 406, since it's only a SHOULD requirement
> > anyway.  Or perhaps it should send 406 only for HTTP/1.1 requests.
> > 
> 
> What does your server do if a client sends a request with *no* 
> Accept-Encoding?  I think that's what it should do in this case.
> (E.g. send it as application/octet-stream).
> 
> It is quite likely that there will be clients which support some
> content encodings but not all the ones you support.  I don't think you
> want your server to be in the situation where it completely refuses to
> serve a file to such clients.

Better I think is that the server maintain two versions or in this case,
uncompress on the fly ... or send a choice page that lets the user
click the link for the explicit file... my assumption being that the
request was for the resource w/o the _Z ... if the _Z is requested,
then John's approach is reasonable ... one of the things missed in
this whole mess is that the user *MAY* simply want the file delivered
unmolested and saved on the hard drive. It is quite inconvenient for
me when something is unziped or uncompressed by browsers, servers, etc.

Dave Morris
Received on Tuesday, 3 November 1998 20:31:06 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 06:33:24 EDT