[thelist] HTML Tables for layout rather than strictly for data presentaion

[aardvark]

> From: "Joshua OIson" <joshua at alphashop.net>
>
> I read through part of the w3c recommendation about accessibility (as
> was posted earlier today) and noticed the section about tables.  It
> very blatantly said that tables should not be used for layout:

well, as you've seen, it recommends against them, but it is not 
forbidden by them... they even offer tips on how to make tables for 
layout less of a barrier...

> http://www.w3.org/TR/1999/WAI-WEBCONTENT-19990505/#gl-table-markup
> 
> I immediately asked myself whether or not I should provide the usual
> array of design templates to a client for consideration (graphics,
> etc) plus one or two mockups that are heavy accessibility and very
> minimal in table-heavy layout and graphics.  I'm still weighing both

i have a question, what changed in the web between the time you 
used tables and the time you read that?  did something happen of 
some import, or did you just happen to stumble across that?  that's 
obviously rhetorical...

the gist is, you don't necessarily change the way you do things the 
way you learn of a different way...  i am not coding in XHTML, and 
probably won't for quite a while... however, i *am* coding my 
HTML4.01 to be compliant (as i always do) and forward-compatible 
to XHTML for when the time comes... but as long as there is a 4.01 
DTD and browser support for it, i doubt i'll need to change existing 
pages...

> sides of this one, no pun intended.  Why is it that if the "experts",
> w3c, Neilson, etc, all say not to use tables for layout, why do so
> many of us use tables primarily for layout?

well, to answer the question as-is, people use tables for layout 
because for 4 generations of browsers, that's how you created 
layouts... it worked then, it still works, and it will continue to 
work... as the ALA redesign has tested, and as we have tested 
with the evolt.org redesign, tables hit more users when you want to 
preserve the layout for as many people as possible...

now, to qualify, i have never seen Nielsen, the W3C, or many 
others, expressly cry foul on the use of tables... they have all 
recognized the de facto use of tables, and instead have counseled 
to use tables *correctly*, in a way that degrades nicely for older 
browsers, newer browsers, and alternative browsers... this includes 
logical flow, summary data, simple structure, etc...

now, don't get me wrong, i'd rather use CSS-P the way it's 
described in the specs, but until 90+% of the folks out there 
support it correctly (and similarly), i'll probably stick with tables 
(that degrade quite well, thank you very much)...

on a related note, a little insight into the evolt.org redesign:

Inside the evolt.org Rebuild: The HTML and CSS
http://evolt.org/article/list/20/5816/index.html

also has a link to the ALA article...