[6427] in www-talk@info.cern.ch
Re: Deploying new versions [Was: Versioning HTML at the
daemon@ATHENA.MIT.EDU (Brian Behlendorf)
Sun Oct 30 22:49:45 1994
Date: Mon, 31 Oct 1994 04:37:54 +0100
Errors-To: listmaster@www0.cern.ch
Errors-To: listmaster@www0.cern.ch
Reply-To: brian@wired.com
From: Brian Behlendorf <brian@wired.com>
To: Multiple recipients of list <www-talk@www0.cern.ch>
Well, fine, servers serve HTML 2.0 docs as "text/html" and
HTML 3.0 docs as "text/html; version=3.0". Browsers that implement all
or part of HTML 3.0 should send an Accept field of "text.html;
version=3.0", so servers can optionally be intelligent about what
versions of documents they serve. The only missing link I see is how
document authors tell servers their documents are HTML 3.0 - most servers
base this on file extension (i.e., .html files are "text/html") so I
don't see an easy way around defining a common file extension like .htm3
or something.
Brian
On Mon, 31 Oct 1994, Marc VanHeyningen wrote:
> [ This barfed and died on my previous attempts to send it to the
> www-talk list, here goes again... ]
>
> > Thirdly, if I create some HTML 3 documents and put them on my server,
> > which I intend to do, if I get my server to spit out
> >
> > Content-type: text/html; version=3.0
> >
> > will this break anything? Will it offend anyone? Will it help anyone?
>
> This will break almost all existing clients. Most of them are not smart
> enough to parse content-types according to MIME rules but will simply
> determine that "text/html; version=2.0" is not a recognized content-type
> and prompt the user to save the result in a file.
>
> Of those clients I have handy, Mosaic 2.5b for X, Netscape 0.9b for X, and
> Lynx 2.3b all do this. Only Arena 0.91b does the right thing.
>
> If you want to test this, a simple file with that content-type is
> available at <http://www.cs.indiana.edu/test.params>.
>
> It's a pity, because it's obviously the right way to provide versioning
> information but existing software won't work with it.
> --
> Marc VanHeyningen <http://www.cs.indiana.edu/hyplan/mvanheyn.html>
>
