ALT=

Here's a little something to attempt to evoke a thread
Since we are going to focus our attention on one issue to test whether 
further work is done in this Working Group we should be clear that this 
is "Article I" of this particular "Accessibility Magna Charta" and that 
the response to it will determine our continuance.  This issue will 
decide whether we are a positive force for making the Web usable or 
mere "Mustang Maoist" dilletantes organized to make meaningless noises 
among ourselves demonstrating how "caring" we are in these matters. This 
is both within this Working Group as well as within WAI and W3C.  The 
"'ALT=' Imperative" deserves our best efforts.

1) We are agreed that the AU-WG of the WAI of the W3C is publishing 
"official" guidelines for authoring tool *COMPLIANCE* with our published 
page authoring guidelines.

2) These guidelines are *REQUIREMENTS* - not "recommendations".

3) Any language used should emphasize this distinction, e.g. we should 
say "alt= *WILL* be used..." rather than "should be..." and there must 
be no ambiguity such as is implied by "there is some argument..." (in 
connection with TITLE=).  Once this is in our guidelines there is no 
argument acknowledged.

4) The fact that "LONGDESC" is "recommended" while "ALT=" is *REQUIRED* 
should be indicated by the prompt presented in connection with the 
inclusion of both attributes (as well as the "TITLE" attribute) in the 
dialogue box that is presented to the author by compliant authoring 
tools whenever IMG is put into a document without ALT=.  The 
justification for this imperative is that ALT= is a "must" for HTML 4, 
not just for the accessibility guidelines.  If W3C's recommended 
standards are to become *STANDARDS*, whether official (as in ISO) or 
just de facto, they must (within the organization's context) be presented 
as faits accompli, not items any longer under discussion.

THE OMISSION OF ALT= IN CONNECTION WITH THE IMG ELEMENT SHOULD HAVE THE 
SAME CONSEQUENCES AS ANY OTHER SYNTAX ERROR!

In short, in connection with our "trial balloon" the importance of this 
first (foremost) issue must be made abundantly clear: ALT= inclusion is 
our "line in the sand" the crossing of which will bring large pins stuck 
in the dolls of the offending  authoring tool authors.

Received on Sunday, 17 May 1998 17:07:36 UTC