[html4all] alt issue raised on HTML issue tracker

[John Foliot]

Steven Faulkner wrote:
> Hi John,
> 
> I am glad the PFWG has responded to the alt issue, but I don't think
> that the tone you are using is useful or productive (even in jest),
> we need to be working with the other members of the working group not
> throwing "salvos".   

Steven,

I agree with your assessment - "salvo" is perhaps not the best word to use
in public - and I must remember to be mindful of this, but in many ways we
*do* need to continue to press at the issues of concern, and with a
concerted and focused attention.  Now that a formal response has been issued
regarding @ALT, I suggest we need to pick up another issue of concern *now*
(as in "when the time is right" - Joshue) and focus our attention on that
issue, no matter which it might be - and there still remains many
outstanding issues.

> 
> In regards to the id/headers issue one thing i picked up from the
> face2face is that j graham and ben have been doing some great work on
> making the tables algorithm work much better so that in many cases
> (note i say many not all) the use of id/headers will not be needed.
> That is not to say that it should be abandoned and from what i
> remember j graham and ben thought along similar lines.    
> one of the things that struck me was that it appears (from my
> understanding of what interrogation features are offered) that screen
> reader vendors have not even implemented the html4 tables algorithm. 

Steven, it is that distinction between "many" and "not all" that is the
issue.  I believe that a strong enough case has been put forward so far to
preserve @headers/@id in complex tables that it should be advanced to the
PFWG.  There was a recent posting regarding empirical need at the
governmental level (which I cannot put my hands on at this moment), and in
the absence of a better solution for complex tables, I prefer even imperfect
A.T. support today vs. a void in the ability to mark up the content
appropriately.  There is no debate that @headers/@id will generally be of
edge-case use/need, but that is not the issue; rather the very fact that
they serve a useful and unique purpose in accommodation should be sufficient
enough to preserve them in HTML5.  I have previously stated that I am quite
happy that @scope is not only being preserved, but advocated within HTML5 -
this is a positive re-enforcement of an accessibility construct that all
developers should be aware of.  But that should not be reason to walk away
from @headers/@id.

And so whether it is the topic of @headers/@id, or of @summary, or
[take-your-pick], I suggest that we take up another of our concerns in a
coordinated effort, and move it forward via the formal process afford us via
W3C WAI PFWG.  And if the colloquial use of "salvo" is counter-productive to
the discussion, then I apologize for any lack of sensitivity on my part.

JF