W3C home > Mailing lists > Public > public-html-a11y@w3.org > December 2010

Re: Issue 130

From: Steve Faulkner <faulkner.steve@gmail.com>
Date: Wed, 15 Dec 2010 12:26:45 +0000
Message-ID: <AANLkTi=YzvdHQDKnWoLiOtNN6bS8QD9Ah7Qr6KzMqqyL@mail.gmail.com>
To: Benjamin Hawkes-Lewis <bhawkeslewis@googlemail.com>
Cc: Richard Schwerdtfeger <schwer@us.ibm.com>, public-html@w3.org, HTML Accessibility Task Force <public-html-a11y@w3.org>, Sam Ruby <rubys@us.ibm.com>
thanks for the info,

i note in the comments of the webkit implementation it says
// This employs a heuristic to determine if this table should appear.
// Only "data" tables should be exposed as tables.
// Unfortunately, there is no good way to determine the difference
// between a "layout" table and a "data" table.

The mozilla implementation provides an indicator that a table may be
used for layout via an object attribute, which can be picked up by AT.

In either case whether its browser based or AT based heuristics there
appears to be a distinct possibility that a false identification will


On 15 December 2010 12:01, Benjamin Hawkes-Lewis
<bhawkeslewis@googlemail.com> wrote:
> On Wed, Dec 15, 2010 at 10:35 AM, Steve Faulkner
> <faulkner.steve@gmail.com> wrote:
> > hi benjamin,
> >>The change proposal claims it is "impractical" for "a user agent
> >>or an assistive technology to create their own set of heuristics for
> >>determining if a table is to be used for layout".
> >
> > It is true that SOME assitive technology have implemented heuristics
> > to detect what they think are layout tables, this sometimes does not
> > work correctly (in my experience testing web sites and web
> > applications) and it also sometimes falsely hides data tables.
> Indeed.
> Given unannotated layout tables aren't going anywhere, and given the
> limited development resources at the disposal of accessibility
> concerns, I think the poverty of these independently implemented
> heuristics is a strong argument for optimizing and then standardizing
> them, just as the poverty of reverse engineered tag soup parsing was a
> strong argument for optimizing and then standardizing it. But that's
> perhaps orthogonal to this change proposal.
> > you state "Multiple user agents" have heauristics, which of the major
> > browsers have such heusristics implemented?
> At least Gecko, WebKit.
> http://mxr.mozilla.org/mozilla-central/source/accessible/src/html/nsHTMLTableAccessible.cpp
> https://trac.webkit.org/browser/trunk/WebCore/accessibility/AccessibilityTable.cpp
> --
> Benjamin Hawkes-Lewis

with regards

Steve Faulkner
Technical Director - TPG

www.paciellogroup.com | www.HTML5accessibility.com |
HTML5: Techniques for providing useful text alternatives -
Web Accessibility Toolbar - www.paciellogroup.com/resources/wat-ie-about.html
Received on Wednesday, 15 December 2010 12:27:39 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 19:55:49 UTC