W3C home > Mailing lists > Public > www-talk@w3.org > January to February 1996

Re: HTML variants and content negotiation

From: Brian Behlendorf <brian@organic.com>
Date: Mon, 8 Jan 1996 11:36:33 -0800 (PST)
To: Kee Hinckley <nazgul@utopia.com>
Cc: www-talk@w3.org
Message-Id: <Pine.SGI.3.91.960108112938.10147j-100000@fully.organic.com>
On Mon, 8 Jan 1996, Kee Hinckley wrote:
> Brian Behlendorf wrote:
> >	    [a benefit of option II, conditional HTML]
> >         Proxy caches don't have to worry about feature-set negotiation
> >           either - one document, not 2^(# feature-sets) possible documents.
> 
> That is definitely an advantage, proxy servers are a problem with server-side solutions. 
> Although in practice I think the browsers on the other side of proxy servers tend for the 
> most part to support the same type of capabilities.  Organizations with the wherewithall to 
> put up proxy servers tend to also have standardized on browsers.

Though browser evolution does tend to be constant, and an organization 
might be able to standardize on one browser but not necessarily a 
particular version of that browser.  I see lots of different levels of 
AOL browsers being used.

> Broken/different browser implementations are a problem.  One advantage of server-side 
> solutions is that we can work around them.  You mentioned the tables-within-tables and 
> inline-within-tables ones.  There are more subtle onces such as whether or not percentage 
> cell widths are supported (not in Netcruiser, with disastrous results (30% becomes 30 
> pixels)).

There's not much we can do, with any solution, that makes writing a 
browser foolproof, accounting for bugs.  If a browser says it can handle 
"tables", and can't, then who loses?  Directly, yes, the user, but I 
would tend to say that should compell the user to improve their browser 
rather than compell the content creator to account for the bug.  The 
content companies that want to stay bugwards compatible have no choice 
but to do user-agent negotiation, I don't think that's an unfair 
conclusion.  At least with conditional HTML it becomes easy to let 
browsers decide (or even users, given the right UI) which features to 
parse.

	Brian

--=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=--
brian@organic.com  brian@hyperreal.com  http://www.[hyperreal,organic].com/
Received on Monday, 8 January 1996 14:34:30 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 27 October 2010 18:14:19 GMT