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

Re: Tables and the Summary attribute

From: Terrence Wood <tdw@funkive.com>
Date: Thu, 29 Sep 2005 12:03:46 +1200 (NZST)
Message-ID: <25284.>
To: w3c-wai-ig@w3.org

David Woolley said:
> You have an upstream accessibility problem.  If you can't understand the
content, how is the reader supposed to understand it?

Umm no, I don't think so. The audience is other accountants and economists
- not web designers.

> If this is a statutory table, the best long term solution would be the
use of a specific namespace, so that one has proper EDI (electronic data
interchange) capabilities.

Yes it would, however that is way beyond the scope of this project. As
part of the solution I have developed some inhouse micro-formats, which at
the very least will be easy transform again in the future.

>> 2) The content owners and I are, in many instances, legally obligated
>> *not* lead any conclusions that may be drawn from interpreting the data
(e.g. by describing trends).

> I'd suggest a fairly short summary that says "Table in form mandated by
<name of legislation>, which forbids any interpretation of the data." You
could also quote the reasons that the legislation requires the table to be

Hmmm.... a bit verbose and falls in the same category as alt="decorative
image" IMHO. That type of information is better expressed once in a legal
disclaimer for the site, not on every table.

>> 3) Preferably the summary is generated programmatically and is
> The authors of AT are going to be better able to do that.  In any case
it would require rather more powerful markup.

I'm not sure what you mean by authors of AT. In any case, markup isn't
needed to produce two or three sentences for a single attibute. Looking at
the sample I provided in my last message the only changes required (for
year on year reuse) are the dates.

>> summary:
>> There are 5 columns. Columns 1 and 2 show forcasted results for 2006.
Column 1 is the original budget. Column 2 is the estimated actual
> This far exceeds the approximately 7 things that people can track in
short term memory and is pretty much useless as a result.

If you are a screenreader user, an accountant or have a degree in
cognitive psychology then I will take this comment seriously.

I personally don't have a problem developing a mental model of this table
by the description, and I think accountant types, having a mathematical
predisposition, probably won't stuggle with it either. Fortunately there
is a budget for usability testing with actual screenreader users.

kind regards
Terrence Wood.
Received on Thursday, 29 September 2005 00:03:51 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 20:36:26 UTC