W3C home > Mailing lists > Public > www-amaya@w3.org > April to June 2000

TAN: replies to your bloatware

From: P. T. Rourke <ptrourke@mediaone.net>
Date: Mon, 3 Apr 2000 14:11:16 -0400
Message-ID: <002801bf9d98$03f8f3a0$c3843ccf@psicorp.com>
To: "J. S. Brady" <jamesbrady@home.com>
Cc: <www-amaya@w3.org>
JSB, Good comments, thank you.  Obviously my rhetoric was intentionally over
the top (i.e., "so twentieth century" - technically the twentieth century
doesn't end until midnight, December 31 this year at the International
Dateline; the phrase was an allusion, though I can't remember what to).  But
dismiss CSS as just so much bloat is at best misguided
(and to do so in such a disrespectful manner is deserving of just the level
dismissal I leveled).  There are obvious problems (mostly with regard to
poor execution on the part of UA vendors, and legacy browser use) with using
it as a top-to-bottom solution right now, but careful use (looking out for
graceful degradation, as JSB says; I like that phrase) can be very, very

As you said, I didn't mean to slam Lynx.  It is VERY VERY important to use
it as you've suggested to support accessibility.  My point was to slam the
idea that new = bad, old = good.  CSS isn't even that new; all three of the
major commercial graphical browsers support it now to one degree or another
(obviously the support is cranky, at best, but I think we all wish they
supported it completely).  Good page design should be able to manage pages
that look good with or without CSS, to support legacy browsers and
non-visual browsers.  If my page isn't readable in Lynx *and* either Mosaic
3 or HotJava, as well as NS 3/4, IE 4/5, and the current Moz, I don't put it
up until it does pass that test (I'm gradually working Amaya into that test
routine, too). But I don't worry any more about making my page beautiful in
Netscape 3, just easy to read.

CSS provides web publishers a resource for vastly improving the readability
of web pages and reducing download time (how many font codes can a single
CSS line eliminate?).  Two pages can be readable, but one more
readable than the other: for instance, I use a little bit of color to
emphasize defining occurrences and product names and such through CSS - and
they show up as em, dfn or strong in NS 3.  The key thing is to avoid using
it for positioning and other dynamic purposes until such time as it is
properly supported and non-supporting browsers are as rare as Mosaic 3 is

Let us point out as well that aural style sheets are part of the CSS2
definition, and promise to vastly improve the user experience for non-visual

As for the claim that with Rainbows and Apple IIs we'd still have the web -
uh, no, we wouldn't - because if no one had had the ambition to improve upon
the Apple II, no one would have had the ambition to try to hyperlink
documents across computer networks, either.  A rhetorical trick, I'll admit,
but the point behind my comments remains valid.  Apple IIs were great for
what we wanted to do the first couple of years after they came out.  Now we
expect desktops to do a great deal more than the Apple II ever did or could

Ultimately Amaya is a demonstration that number of new technologies (CSS,
clean structural markup, and most importantly editing in the display
browser) are possible.  No point calling it bloatware and slamming the
developers when you've already closed your mind to the value of the
capabilities the software was designed to demonstrate. (and no, I do not
mean JSB).  I think LKM's and JSB's postings are good examples of how to
contribute to a project rather than ridicule it because things didn't work
the way you wanted them to the first time you tried.

PTR (no O', thank you)


Left in because it all bears repeating:

> While I don't encounter much of a problem with HTML4, I don't use CSS
> much currently, because of the problems I've encountered with
> cross-platform compatibility and accessibility concerns.  Until differing
> browsers handle CSS similarly, I tend to avoid it.  If you don't, good on
> you.  My coding needs and user base may be very different from yours.

> As to Lynx and 300 bps modems:  While 300bps is overkill (underkill?),
> it's easy to forget that Lynx is still widely used (I use it frequently on

> my Linux box, when I want quick information and don't care about
> graphics), and given the increase in 'web-enabled' cellphones, PDAs, &c.,
> non-graphical compatibility (including 'graceful degridation')is an issue
> that should be taken more seriously by web developers.

> This issue is increasingly important as more and more blind and visually
> impaired users are discovering the internet (especially the WWW and
> e-commerce) as ways to increase their independence.  The W3C has released
> accessibility recommendations which address both 'web-enabled'
> non-graphical compatibility and accessibility for the blind / visually
> impaired, but I don't see their recommendations being adopted much.

> Oh, and there'd still be a WWW if everyone were still running on Apple
> II's and Rainbows.  It simply wouldn't be graphical.  While it might not
> be as popular, it certainly would still exist.  It was a good method for
> presenting information even before the "Here's a photo of me so you can
> see what I look like in RL" craze that lead to the graphical explosion
> Mosaic ushered in.

> > Rather, NOBODY in their right mind as their first response to
encountering a
> > problem with a program launches a tirade at the developers; those in
> > right mind describe the problem to the developers and hope they
> > referent: user and developers) can together isolate it and fix it.
> Now _that_ is a sentiment I can (mostly) agree with.  While I understand
> why compiling the source rather than the binaries would be preferrable,
> and can certainly sympathize with the frustration involved in a downloaded
> source that simply will *not* compile when one had expected it'd be an
> easy job, posting tirades like the message starting this thread is not
> only unadvisable and incindiary, but foolish as well.
Received on Monday, 3 April 2000 14:11:31 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 14:53:21 UTC