W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > July to September 2005

[techs] Table Summary Tests (111, 112, 113, 114, 203)

From: Mirabella, Mathew J <Mathew.Mirabella@team.telstra.com>
Date: Thu, 18 Aug 2005 12:40:11 +1000
Message-ID: <73388857A695D31197EF00508B08F29816655514@ntmsg0131.corpmail.telstra.com.au>
To: "w3c-wai-gl list" <w3c-wai-gl@w3.org>

>Vicente Luque Centeno wrote:
>> I agree that ``summary="" and no <caption>'' SHOULD indicate a layout

>> table, but I am afraid that ``summary="" and no <caption>'' DOES NOT 
>> indicate a layout table.

> ACK. A layout table _should_ have an empty summary and no caption. But

> this cannot be a criterion to decide whether it _is_ a layout table.
> can also be a badly marked-up data table.

I know that not all tests will be automated, but... The above point does
the difficulties in really providing automated tests for a lot of

I would suggest that it is pretty easy to code up a good and a bad
layout table, a
Good and a bad data table, and show what you require for each to be
Once there is agreement on what is required etc.

What is much more difficult is to take a page with a table
(or anything else) and (via automated test) work out whether it really
a data table or layout table before working out whether the 
Developer has done the right thing by the table And it's purpose.

So determining what you require for good (accessible) web content
Is only the first part of having a way to really test for this on a

I am sure this has been talked about before... But I think we do need to
We keep a focus on what really can be used as a criterion to say what
web content
Actually is, and what is a criterion for what we require for that web
content to be
Accessible given what it actually is.
Received on Thursday, 18 August 2005 02:44:19 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 16 January 2018 15:33:55 UTC