- From: Al Gilman <asgilman@iamdigex.net>
- Date: Mon, 21 Jun 1999 08:23:52 -0400
- To: WAI UA group <w3c-wai-ua@w3.org>
At 11:23 PM 6/20/99 -0400, Charles McCathieNevile wrote: >I'm not sure where this suggestion fits. Probably as a technique for >orientation or for structure navigation. This would have been a good point to coordinate with the UA group earlier. It is possible GL made a mistake based on inadequate examination of the UA issues. [mistake in not asking for TITLE on all occurrences] There is a technique here but it is in the class of dictionary services or on-the-fly schema extension, not the areas you describe. This is enriching the schema in the area of atomic (leaf-level) components, not structural forms. An illustrative example could take the form: <ABBR title="International Standards Organisation" class="definition" cite="http://www.iso.org/">ISO</ABBR> where <... class="definition"> is a flag indicating that other uses of ABBR with the _same content_ should inherit the other attributes as used here. Asking the User Agent to parse the element content below the markup level and as here to compare content between different nodes in the parse tree is work that the User Agent would like to avoid doing as a general rule (severity understated). If we don't flag the annotated instance with something like <... class="definition"> the User Agent is left to implement a rule that "ABBR and ACRONYM instances that share content share attributes" and that may be more checking that we can negotiate without a better general agreement for multilevel partial understanding of the document content than what we have at the moment. Al > >Charles McCN >---------- Forwarded message ---------- >Date: Fri, 18 Jun 1999 15:58:04 -0400 >From: Guy M. Fisher <guy@squeakywheel.org> >To: 'Chuck Letourneau' <cpl@starlingweb.com> >Cc: w3c-wai-ig@w3.org >Subject: RE: Titles for <ABBR> & <ACRONYM> >Resent-Date: Fri, 18 Jun 1999 16:15:21 -0400 (EDT) >Resent-From: w3c-wai-ig@w3.org > >Thank you for the explanation of the thinking behind the <ABBR> >expansion checkpoint. > >Given the decision to go with the "minimum" accommodation, maybe it >would make sense to recommend that user agents cache the <title> >expansions of <ABBR> & <ACRONYM> elements? > >Thanks again. > > >Guy M. Fisher > >Cleveland, Ohio >guy@squeakywheel.org >
Received on Monday, 21 June 1999 08:18:16 UTC