W3C home > Mailing lists > Public > w3c-wai-gl@w3.org > October to December 2004

Re: Notes from discussion about programmatic object, non-text content

From: Doug Gibson <doug@dgibson.net>
Date: Fri, 1 Oct 2004 21:28:52 -0400
To: <w3c-wai-gl@w3.org>
Message-ID: <004701c4a81f$2e197790$6e01a8c0@death>

On Thu, 30 Sep 2004 15:55:20 -0400, Wendy Chisholm <wendy@w3.org> wrote:

> Thinking specifically about javascript, we wanted to list all current
> inaccessible techniques that do not degrade if scripting is not
> available. We listed: dynamic menus, inserting content, and
> (e.g., calculator).  Client-side validation doesn't degrade well to
> server validation not because of accessibility issues but because of
> security and cost issues.

I'm still curious as to where the line of "inaccessible" is drawn on
some of the above examples (as generalizations). I have worked on
numerous dynamic navigation/menus (including expand-collapse and pop-out
menus) scripts in the past that I considered accessible by the fact that
with scripting disabled, the actions of the activating links were still
valid and the equivalent "drilldown" or expanded navigation was revealed
on subsequent page loads (with or without the script, actually) via
server-side logic rather than dynamically on the current page. The
sub-navigation was never really obscured or made inaccessible to the
user by having scripting disabled.

The same could be envisioned for the above mentioned calculator of a
simple nature (such as shipping calculator).

Looking back, my scripts were not really accessible because they were
lacking in keyboard support and probably other technique-related
inaccessibilities (still perfecting the techniques in many cases), but
with enough diligence and the same content or effect remaining
accessible or available, many of these sorts of scripts could remain
accessible solutions, IMO.

How do you distinguish or at what point is a script simply not enough on
its own to be an accessible solution vs it being inherently
inaccessible? Form validation could be accessible as well, but as
discussed, should not be used in lieu of server-side validation for
security reasons more than accessibility reasons.

Doug Gibson
Cold Fusion Web/Application Developer
Received on Saturday, 2 October 2004 01:29:07 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 16 January 2018 15:33:51 UTC