W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > January to March 2001

Layout tables [ was summary attribute required? history.]

From: Leonard R. Kasday <kasday@acm.org>
Date: Mon, 05 Mar 2001 15:24:59 -0500
Message-Id: <4.3.2.7.2.20010305145839.01fb1350@pop3.concentric.net>
To: Charles McCathieNevile <charles@w3.org>
Cc: Al Gilman <asgilman@iamdigex.net>, William Loughborough <love26@gorge.net>, <w3c-wai-gl@w3.org>

At 12:55 PM 3/5/01 -0500, Charles McCathieNevile wrote:
>It's more a general use thing. If you are habitually using tablin or
>something to linearise tables, and you run across some that make no sense but
>whose summary says "layout table" then you understand why.  [with a few 
>typo fixes by LRK]

Would it not be better to have some machine readable way to indicate that 
it's a table?  e.g.

summary=""
class="layoutTable"

Or better yet, define a new XHTML attribute for TABLE, say "purpose", with 
values "layout" or "data".  It would default to data.

In other words, the following would be a layout table:

<table purpose="layout">
<TR> <TD> The </TD> <TD> Cow</TD> </TR>
</table>

I know of course that this is slightly heretical: there's a philosophical 
position that CSS should be used for all layout.  But the argument there is 
that layout violates the intended "purpose"  of table.  Once we make layout 
an intended purpose, that objection vanishes by definition.  Also, that 
philisophical argument is flawed anyway, since even data tables confound 
content and layout, so we don't have a clean separation to maintain.

Meanwhile, from a practical point of view, designers can keep on using 
tables, and screenreaders can handle them sensibly.

And designers need to use tables at least for a while because, for one 
thing, CSS positioning is loaded with bugs still (e.g. see 
www.richinstyle.com for a bug list).

Also, are there any CSS constructs that guarantee absence of overlap?  See, 
for example the overlap that takes place even when em is used in

http://astro.temple.edu/~kasday/wai/css-em.html

Is there any way to avoid the overlap there without tables?  If not, then 
we REALLY need layout tables until CSS gets the proper functionality.

I think XHTML and WCAG 2.0 gives us an ideal opportunity to fix this.

Len

P.S.
I'd still like a teeny erratum or clarification to WCAG 1.0 explicitly 
allowing summary to be omitted to set null in layout tables.
P.P.S.
It was great meeting all the folks who were able to make the face to face.

--
Leonard R. Kasday, Ph.D.
Institute on Disabilities/UAP and Dept. of Electrical Engineering at Temple 
University
(215) 204-2247 (voice)                 (800) 750-7428 (TTY)
http://astro.temple.edu/~kasday         mailto:kasday@acm.org

Chair, W3C Web Accessibility Initiative Evaluation and Repair Tools Group
http://www.w3.org/WAI/ER/IG/

The WAVE web page accessibility evaluation assistant: 
http://www.temple.edu/inst_disabilities/piat/wave/
Received on Monday, 5 March 2001 15:24:31 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:47:09 GMT