W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > July to September 1998

Re: CSS vs Tables For Achieving Complex Visual Layout

From: Todd Fahrner <fahrner@pobox.com>
Date: Wed, 22 Jul 1998 22:25:20 -0700
Message-Id: <v04011701b1dc5434a555@[206.245.203.103]>
To: raman@adobe.com, WAI GL <w3c-wai-gl@w3.org>, Charles McCathieNevile <charlesn@sunrise.srl.rmit.edu.au>
Cc: w3c-css-wg@w3.org
T. V. Raman wrote (6:50 PM -0700 7/22/98):

 " This message addresses the importance of transitioning the
 " WWW to using CSS for complex visual layout rather than
 " continuing down the current hack of abusing tables.

I've coded such pages and worse, and in fact I'm working late tonight
trying to edit several >50K table confections originally extruded from the
latest version of a popular "WYSIWYG" editor. You are absolutely right that
it's abominable. The best that can be said for it is unfortunately all that
matters for the parties who pay for such activity: it barely works in the
great majority of UAs in use today.

True columnar layout via CSS, while desirable, is not necessary. I believe
CSS1 would be enough to retire at least the worst abuses of tables.
Unfortunately again, CSS1 has been grossly misimplemented thus far, and
every "not quite" release pushes the birthdate of CSS's mainstream
commercial viability a few years beyond the horizon.

Complete, consistent support for the following (conspicuously neglected)
CSS1 properties <em>ACROSS ALL APPLICABLE MARKUP ELEMENTS</em> is a minimum
condition for retirement of tables-as-layout:

1. The display property
2. The box model
3. The float property
4. The width property

I will be very happily shocked if either of the upcoming "big two" 5.0
releases ships with all of these right. I do not disparage the skills of
the engineers working on these things - I'm sure it's challenging,
especially with effort divided between ever-more intensive reinforcement of
the hack regime with tweaks to the table rendering code, and implementing
CSS3 features.

I do think, however, that there's a widespread failure to acknowledge how
damaging it is to ship patently buggy, patchwork CSS1 implementations.
Shipping with such "support" off by default would at least not retard the
adoption of CSS after its implementations achieve a degree of coherence
compelling to design professionals.


Back to those tables now.... I'm fantasizing about some kind of surgical
electromagnetic pulse bomb that can obliterate all copies and source of the
software that generated them.

Todd Fahrner
mailto:fahrner@pobox.com
http://www.verso.com/agitprop/

The printed page transcends space and time. The printed page, the
infinitude of books, must be transcended. THE ELECTRO-LIBRARY.
	- El Lissitzky, 1923
Received on Thursday, 23 July 1998 01:24:55 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:46:58 GMT