general notes, concerning DOM and UA issues from the archives

Hi

Per a request from Jon, the UA Chairperson, I've tried to pull together
and digest everyone's ideas and related comments concerning the Document
Object Model (DOM) from a User Agent perspective prior to the DOM
teleconference call on Wed., March 31st, for people to review.  In the
UA archives, most of this discussion took place during February, 1999.
I hope these notes are helpful.

If anyone on this list is still unfamiliar with DOM, I'd suggest reading
two short notes.  First, read the introduction to DOM1, titled "What is
the Document Object Model" at:

http://www.w3.org/TR/REC-DOM-Level-1/introduction.html

Second, read the Frequently Asked Questions notes regarding DOM at:

http://www.w3.org/DOM/faq.html

mark

--

1.  There was a lot of email discussion regarding specific accessibility
APIs and their implementation regarding DOM.  However, I think T.V.
Raman's comments summed this entire discussion best, as follows:

     " The DOM is the underlying document model for the WWW
platform-- from an accessibility perspective, the significance of the DOM
to future accessibility far outweighs the importance of any single
platform-specific or vendor-specific API.

If a specific browser or platform can offer access functionality over
and beyond what the DOM can offer today, more power to that platform
-- and we should all whole-heartedly applaud the efforts of said
platform; however, this should not be an excuse to twist the arms
of access vendors into providing accessibility exclusively through
platform-specific APIs -- nor should it be used as the justification
for discouraging developing appropriate access relevant features in
the DOM."


2.  I think the best way to look at DOM and problems for accessibility
was best summed up by Hans Riesebos as follows:

     "Assistive technology has to tackle two problems: How to
make the UI accessible and how to make the documents accessible.  If
one part can be platform independent this would be great.  DOM is the
answer there."

I think it is very important that the UA group make this distinction.


3.  There were several email discussions mentioning things like
"exposing DOM, interfaces to external programs from DOM, etc.".
Quoting Hans Riesebos again, I think best summed up this email
thread as follows:

     "I feel that the term "exposing DOM" is in a sense a
misleading. What we want is to use or have a DOM.  Exposing DOM
is not the action we are looking for.  In this way many exposures of
DOM will lead to the same pitfall.  Every exposure will need
developers attention ..."


Going one step further, I applaud the efforts going into DOM 2, but
I am concerned, for example, that additions of things like the UI
Events are creating platform dependencies.  Rich Schwerdtfeger
and I have talked regarding the idea of an "extensible" DOM (Rich
has also called this an "application model").  I'm not sure if I would
use the syntax "DOM" at this point, if for no other reason than to try
and separate what the DOM working group needs to be sure is
"built-into DOM", and for the UA group to define what the UA needs
to be able to do to provide the information (structure, navigation,
semantic, connection, call-back, etc.) in the DOM "extension or
application model" as the UA/user prefers it.

Received on Tuesday, 30 March 1999 12:14:55 UTC