Re: AU techs / ATAG errata

Yes, it would violate 2.2 - by removing the attribute from any HTML
specification since 1997 it is generating invalid markup. Internal
information held by the tool can be in any format desired - whether it adds a
temporary flag that is not saved as part of the document, (amaya does
something like this for marking various features) or whether it generates
some EARL that it stores and updates locally, publishing it only at the point
where the document is saved, or whether it applies a validity checking
algorithm to find where alternatives are missing.

In general we don't specify how tools should deal with checkpoints (what
their internal processes are) in the guidelines. We could list this as a
technique, but we still need to resolve the issue of which guideline comes
out on top. I agree with Emmanuelle that 3.4 is the important one, and we
should be explicit about that.

cheers

Charles

On Fri, 7 Dec 2001, Jan Richards wrote:

  Charles McCathieNevile wrote:
  >
  > That seems like my proposal 2 below, since I don't believe that there is any
  > way we can effectively force authors to provide meaningful text replacements.
  > (Amaya requires an alt before it will insert an image, and I know that people
  > who really care about accessibility from time to time use asdfasdf as the alt
  > to get on with working).

  It's like 2 except that that message is for the INTERNAL use by the tool
  (basically flagging a problem). If it isn't fixed by the end of the
  editing session, the attribute must be removed. Would removing this
  flagged attribute violate 2.2? i.e. does "generate" refer to creating
  the markup in the tool or saving it? Or both?

  JR: 3.4 could be changed (in an errata) to specify that:
      the only allowable generated text is "place holder text generated by
      [authoring tool name]", which must be replaced by human authored
  text
      by the end of the document editing session.

  >   > Charles McCathieNevile wrote:
  >   > >
  >   > > 2. We state in errata that 2.2 overrides 3.4 and that the checkpoint can be
  >   > > ignored in order to meet 2.2. I don't think this is what we intended either -
  >   > > it would lead to the addition of dummy text just to provide validity, and
  >   > > there is a real need not to do so. If we do go this way we should delete
  >   > > technique T0176






-- 
Charles McCathieNevile    http://www.w3.org/People/Charles  phone: +61 409 134 136
W3C Web Accessibility Initiative     http://www.w3.org/WAI    fax: +1 617 258 5999
Location: 21 Mitchell street FOOTSCRAY Vic 3011, Australia
(or W3C INRIA, Route des Lucioles, BP 93, 06902 Sophia Antipolis Cedex, France)

Received on Friday, 7 December 2001 23:52:11 UTC