RE: Table Techniques - Summary

I concur with John, we really need to prevent the accidental or
indifference of missing summary attributes.  Although, Roberto's
position is valid in the view that machine validators like Bobby must
adhere to the WCAG standards, we must keep in mind the limitations or
extra burdens we place on software manufacturers.

The more we can set up for machine validators to test, the easier it
will be for designers to validate with an objective view.  As discussed
in our teleconference, and pointed out by Gregg I think, human
validators must be highly reliable in order for visual inspections to
work.  This leads us to the objective view of machine testers like Bobby
and the many others.

We are not designing the standards to meet machine testers, we are only
providing an easier way to ensure the tables are set up correctly.  By
requiring a null summary for layout tables and a content filled summary
for data tables we can easily perform a visual and machine validation.
By allowing the summary to be ignored only puts people into the habit of
not providing a summary and then opening up the situation to
accidentally not including a summary on data tables.  There are more
layout tables than there are data tables - habits form easily and are
transferrable to all development processes.  We need to prevent bad
habits if at all possible.

If the summary is null, then the test (machine or HIHR) could simply
validate for no captions and no th elements.  It's simply easy to do and
IMHO makes perfect sense.

Sincerely,
Lee Roberts
President/CEO
Rose Rock Design, Inc.
(405) 321-6372
http://www.roserockdesign.com


-----Original Message-----
From: w3c-wai-gl-request@w3.org [mailto:w3c-wai-gl-request@w3.org] On
Behalf Of John M Slatin
Sent: Friday, August 15, 2003 8:23 AM
To: Roberto Scano - IWA/HWG; Matt May; Chris Ridpath
Cc: WAI WCAG List
Subject: RE: Table Techniques - Summary



Roberto makes a good point when he asks "Why use the [summary]
attribute" if the resource contains no text describing the purpose or
organization of the table?

The primary reason is to confirm that the absence of explanatory text is
deliberate rather than accidental.

It occurs to me that the problem of identifying table types could be
solved by providing a type attribute for the table element; legal values
could be "layout" or "data." or An attribute called "layout" whose
values could be either "yes" or "no," could also work.  These would
become part of the XHTML spec.

John  


John Slatin, Ph.D.
Director, Institute for Technology & Learning
University of Texas at Austin
FAC 248C
1 University Station G9600
Austin, TX 78712
ph 512-495-4288, f 512-495-4524
email jslatin@mail.utexas.edu
web http://www.ital.utexas.edu
 


-----Original Message-----
From: Roberto Scano - IWA/HWG [mailto:rscano@iwa-italy.org] 
Sent: Thursday, August 14, 2003 3:56 pm
To: John M Slatin; Matt May; Chris Ridpath
Cc: WAI WCAG List
Subject: Re: Table Techniques - Summary



----- Original Message ----- 
From: "John M Slatin" <john_slatin@austin.utexas.edu>
To: "Matt May" <mcmay@w3.org>; "Chris Ridpath"
<chris.ridpath@utoronto.ca>
Cc: "WAI WCAG List" <w3c-wai-gl@w3.org>
Sent: Thursday, August 14, 2003 9:43 PM
Subject: RE: Table Techniques - Summary


>
> I agree that null summary (summary="") should be allowed for layout
> tables.  As Matt points out, this indicates a postive intention on the

> author's part to force screen reader behavior, just as the null alt
> attribute for images does.  Use of the <th> element is *another* good 
> indicator of the author's intent, in this case to create a data table 
> rather than a layout table.
>
> By contrast, the absence of a summary attribute, like the absence of
> an alt attribute, may simply indicate ignorance or indifference on the

> author's part.

I could understand and agree with John but the HTML reference [1]
referring to "summary" explain that:

"This attribute provides a summary of the table's purpose and structure
for user agents rendering to non-visual media such as speech and
Braille."

"summary" is different from the ALT attribute [2] that must be specified
for the IMG and AREA elements.

So, if there is no text that presents the table's purpose and structure,
why use the attribute?

Roberto Scano
---
[1] http://www.w3.org/TR/html401/struct/tables.html#adef-summary
[2] http://www.w3.org/TR/html401/struct/objects.html#adef-alt

Received on Friday, 15 August 2003 12:17:57 UTC