And I maintain that the forms section of the standard was written
with windows and IE in mind.

I agree that that piece is more subjective than other standards, and
that Jaws and IE would have been the most common experience that the
authors had to draw upon.  I do no agree that there is anything in the
standards that is unfairly binding to Windows and IE.

Accessible means I can use it with a heavy browser at least.

I have no idea what you mean by heavy.  Your previous use of the term
accessible has been amorphous.

How would you use forms like this on the phone?

I would expect an auditory phone browser to have a mechanism for
navigating data tables.

I know they'd have to be heavily reworked before
being sent over the wire.

Well constructed web sites, say WCAG double A compliant, would not
have to be worked up much.

It is *safari* and not Voice Over which is at issue here

How can you make that assertion with such conviction?  Safari is
pretty closely scrutinized by the web standards folks.  I have
followed some of those conversations and I do not recall handling of
data tables nor forms to have been an issue.

though for unlike msaa and bolted on apps, It is a good deal the
responsibility of the app to provide the information to the screen
reader and

At the very least, it is convenient that the screen reader developer
cannot just point a finger at the developer of the operating system
without pointing the finger at himself!

Safari does not expose the information to Voice Over

How do you know this?  Please provide an authoritative reference.

even if it were prepared to use it.  If this were the case, we would be
able to use the table commands available in Voice Over with safari.

Again, pure speculation.  Wishful thinking maybe.

A form is not a data table it is a layout table

You are wrong.  A conservative definition of data table is a table
that uses column or row headers to apply to multiple cells.  You could
be more liberal and allow that it could be be one header cell on just
one data cell.  A more general test is this:  Does  the order cells
are read change the meaning?  If so, it is not just a layout table.
Either of these last two definitions clearly applies to form elements
in tables.

In any case, the original thing I asked for help with is an array of
radio buttons in a data table.  It is a data table because the intent
is for the labels in the top most row and the left most column to both
apply to a particular data cell.  Correct structural markup has been
included to reflect this relationship.  The form satisfies the
requirements of 508/WCAG, but is really not useable by someone who
relies on VoiceOver.

if it is a table at all.

Right, forms do not have to use tables.  But I would guess that the
majority of them do.

The wild has lots of bad stuff in it.

Agreed.  But one cannot just expect to avoid data tables altogether.

I agree that information should not be coded in liniar fashion

Really?  Okay, so you agree that data tables can be acceptable.

but there should be ways to code which allow it not to be
liniarized and still functional in a liniar way.

Right.  That is what I did.  I coded it as a correctly marked up data table.

O wonder how .31 would play here?

The functional performance criteria are not applicable because we have
the more discrete standards for web content.  In any case, the content
*can* be used in a none-visual mode, but the assistive technology
needs to be sufficiently robust.

Why not come up with a way to code a form which meets the
standards and is as accessible as possible to all?

I regret that I am at a loss as to how to prepare data tables in html
so that they are compatible with VoiceOver.

Reply via email to