W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > April to June 2003

Re: Question about Data Tables

From: Al Gilman <asgilman@iamdigex.net>
Date: Tue, 22 Apr 2003 15:02:51 -0400
Message-Id: <>
To: Josh Hughes <josh@deaghean.com>, Nick Kew <nick@webthing.com>
Cc: w3c-wai-ig@w3.org

At 02:02 PM 2003-04-22, Josh Hughes wrote:

>Nick Kew wrote:
>>Unless perhaps you were misusing the word validator to mean some
>>kind of smart-heuristic checker?
>Well, this is I have so far:
>To help you through the user checks, the validator generates a text-only, 
>de-styled, and de-scripted versions of the page. The text-only version is 
>the one I'm currently concerned about. The idea is to re-write the table 
>as it might be read in a screen reader.

This is a somewhat obsolescent notion.  You may want to implement the function,
but please be aware that many screen reader users won't want it.

Better the assistive technology should (and many do) navigate and explain
the table.

So make the linearization of tables an option that the user can control.  Not
a hard-wired feature of a text-only or text-mostly morph of the site.


See also the comma tool at w3.org known as 'tablin' for behavior to compare
with, and the linear-option tables on the Trace site as output to compare with.



>For example:
>Original: http://condor.gmu.edu/josh/validator/tutorial/tables.php
>I've still got a lot of work to do, but you kinda get the idea.
>I just wanted to make sure that "guessing" the scope was part of the 
>standard, and could reliably be expected from screen readers.

Yes screen readers will generally treat the cell contents in the first row
and column as default row and column labels.  They do this because they have
to help their users with all kinds of tables, not just HTML tables.

An interesting resource in this regard is the workshop that the Federal 
[interest group] in the U.S. did on the topic of accessible tables.


It includes some [not exhaustive, but interesting] information on experience
with actual screen readers.

Received on Tuesday, 22 April 2003 15:03:00 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 20:36:15 UTC