W3C home > Mailing lists > Public > www-html@w3.org > August 2006

RE: XHTML Applications and XML Processors [was Re: xhtml 2.0 noscript]

From: John Foliot - WATS.ca <foliot@wats.ca>
Date: Wed, 2 Aug 2006 19:09:22 -0700
To: "'Jim Ley'" <jim@jibbering.com>
Cc: <www-html@w3.org>
Message-ID: <009e01c6b6a1$dc15bd80$6501a8c0@bosshog>

Jim Ley wrote:
> That's an unfortunate limitation of the screen reading technology, it
> does not mean that we should limit other groups access to content to
> make them the same, the current model isn't harmed if scripts only
> execute at the end, it still works, and it still works interopably,
> there are of course problems with scripts and screen readers, we
> certainly know that, but changing the processing model in the way
> suggested here isn't going to improve matters for screen reader
> users, and is going to harm others. 

OK, and this is where I am confused - you are advocating the ability for
scripts to fire in mid-stream or certain user-groups are "harmed".  How
are they harmed - exactly? (And please do excuse the fact that at some
levels this is technical above my head, but I am following).  

As I am understanding, Mark is simply advocating that scripts fire after
the document has loaded completely, and not mid-load.  How is this
really harmful?  What I understand him to be saying is that by waiting
for the document to finish loading, it allows for any matter of
post-processing ("..SAX and DOM-style processing (any style,
actually)...") - and I further suggest that one of these post-process
functions is the further ability for AT to "do it's thing".  Now it may
in fact add to the overall time required for total processing for that
user group, but on another level it also "levels the playing field",
which rightly or wrongly is and may significantly be affected by
non-technical requirements such as legislation that insists on
"equality".  I don't mean to pull out the legal trump card in a
technical discussion, but I cannot see a real technical harm caused by a
model such as what Mark proposes, yet I *do* see some potential real
harm, or at least unequal treatment from a process that favors one user
group over the other, even if one group represents 99% of the users.
After all, that is what equality legislation is in place for in the
first place.

Now it can be argued that recommendations and best practices (and yes,
even laws) should not affect capability and capacity - I've heard that
argument and understand the talking points. But since I cannot see the
"harm" in waiting for Mark's suggested model (full download and then
post-process), and can see the harm of your proposal - and since, in
practical terms they essentially arrive at the same place (yours in
microsecond faster)... Well Jim, just because we can, does it really
mean we should?

> I'm not sure where you've got the idea that I would in anyway dislike
> the fact non-visual users rely on AT's, or indeed any other
> developers, or the fact some people only see complete documents,
> however because one group only has linear ordering, it doesn't mean
> we should harm those people who have other models.

I am oft concerned that one stream of W3C activity often "forgets" other
streams... Such as WAI.  One of the "idealistic" goals of the W3C, as
articulated by TBL himself is: "The power of the Web is in its
universality. Access by everyone regardless of disability is an
essential aspect."  Yes, I know, almost over-worn by now - but truths
often appear that way...

> There has been no suggestion in this thread to add to their burden,
> the only suggestion has been to add to other users burdens.

And again, pardon my ignorance, but how does microsecond delays (even
delays of a whole second - the time it took to read this aside) in the
ability to fire scripts cause "harm".  Please Jim, with all due respect,
can you further define "harm" in the context of this conversation - I
think (hope) I have clearly articulated my concerns over potential
"harms" regarding one user group, but I am still unclear on the "harm"
to other user groups you keep referencing.


Received on Thursday, 3 August 2006 02:09:45 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:06:13 UTC