Re: Enhancements to Forms - another loss??

James Aylett (sja20@hermes.cam.ac.uk)
Wed, 12 Feb 1997 22:01:39 +0000 (GMT)


Date: Wed, 12 Feb 1997 22:01:39 +0000 (GMT)
From: James Aylett <sja20@hermes.cam.ac.uk>
To: Walter Ian Kaye <walter@natural-innovations.com>
cc: www-html@w3.org
Subject: Re: Enhancements to Forms - another loss??
In-Reply-To: <199702122057.MAA11850@shellx.best.com>
Message-ID: <Pine.LNX.3.95.970212214836.530E-100000@crystal.clare.cam.ac.uk>

On Wed, 12 Feb 1997, Walter Ian Kaye wrote:

> I would favor the space-delimited ALTSRC/ALTREF="url#1 url#2 url%20#3"
> idea that was one of the suggestions in last year's thread. As to the
> related issue of retries, perhaps that could be a stylesheet item?

The first idea (ALTSRC, etc.) I would grant is a quick and fairly useful
solution to the problem (in the same way that FONT was for the developing
requirement of presentational data attached to the structural markup). For
this reason it will probably make its way into some implementations in the
same way Netscape introduced LOWSRC. As can probably be ascertained from
my previous message on the list, I don't consider this a particularly good
idea - it certainly doesn't fit the 'shape' of HTML as a structural markup
language, at least in my mind.

As to the idea that retry/error response should go into stylesheets, I
wonder where you get that idea from. It doesn't appear to me to be related
to anything currently embedded in stylesheets, and seems quite out of
place. In particular, I don't see how this sort of information can be
determined at all usefully by the page author for all possible clients.
Consider a link which takes ten seconds to open a connection between the
client and server. Our nice page author is sitting on a T3 link, with fast
ethernet connection, and so never considers this problem - and codes a
timeout of five seconds into all his links, reasoning that if it isn't
loading by then then the server is probably down, and a mirror should be
used instead. This may be even slower for the poor user, resulting in a
very annoying site.
In my opinion, it puts power into the hands of the authors which can't
really be used sensible - so what's the point?

Cheers,
James

-- 
/-----------------------------------------------------------------------------\
  James Aylett - Crystal Services (crystal.clare.cam.ac.uk): BBS, Ftp and Web
     Clare College, Cambridge, CB2 1TL -- sja20@cam.ac.uk -- (0976) 212023