Minutes of the Open A11y Expert Handlers SIG Call 2008/03/31
- 1 Minutes of the Open A11y Expert Handlers SIG Call 2008/03/31
- 2 Minutes of Expert Handlers Conference Call 2008/03/31
- Neil Soiffer (NS/chair)
- Pete Brunet (PB)
- Gregory J. Rosmaita (GJR, scribe)
- Janina Sajka (JS)
- regrets: none logged
Approval of Last Meeting's Minutes
- Minutes from 2008/03/17 Expert Handlers Conference Call
- Scribe's Note: Attendees at the 17 March 2008 call are asked to review the minutes carefully and to provide additional information or clarification where and if needed
Minutes of Expert Handlers Conference Call 2008/03/31
Topic 1: Continuing the conversation from the 17 March 2008 Handlers call
- For Reference
PB: if application has a built-in navigational mechanism to get to math structure and editor has concept of context and type of structure would know to go to next term -- couldn't it send that MathML segment to the EH, because EH knows how to form it (add to DOM)
NS: if skip term what is next chunk sent? [gives example too complex for this scribe]
PB: when move, what are moving by?
JS: should be irrelevant -- granularity settings right side or left side more granular to more granular to more granual
NS: wandering around tree difficult and wasn't well received; if particular node simple, don't need to wander but just navigate tree and when get to node say "here is the bit you wanted to insert"; wandering around the tree complicated for people to get; would rather walk text string
NS: (example given on PB's request): fraction move over "end of expression"; go down into it would announce "numerator, one of 2 terms" if walk down another level will say "a squared" then "plus" then "b squared" then "bing" to move out of numrator, then move to denominator, walk tree and announce "C squared" walk out denomenator, walk out fraction (would say "fraction") then "end of expression" totally out -- a LOT of steps
GJR: this is where increase/decrease in granualarity of successive where am i queries enters...
NS: so many steps
JS: reminds me of VI -- editor mode and text mode; semantic trap "up" and "down" in complex expression
NS: didn't enter mode -- used control plus arrow to move through tree to move through text, use arrow to read
JS: if not more complicated not more complexity? don't they need for editing?
NS: didn't have an editor -- simply reading and navigating; editing stuff would automatically go into numerator and denomenator -- system for which done, had "smooth" editor: if typed and deleted character, you ended up back where you were (as you should); one key took one up or down (echoed keystroke action) -- only had one user use and project died after i left
PB: how does AT know
GJR: XML Events 2 (implements - can point to local library/script or to namespaced URI
NS: another editor "Chatty Infty" -- self voicing like the application i described and built for mathematica above
NS: think better if AT gets document
PB: AT gives you braille support
NS: some things AT good at: braille one of them; speech another;
Topic 2: Solicited and Unsolicited Feedback
Topic postponed to 2008-04-06 meeting.'
- Item 1.1: comments received so far via accessibility-rfc (at) a11y (dot) org have been logged to the UUC1 Scratch Pad's Reviews & Comments Section
- item 1.2: should each comment be replied-to? if so, by whom?
- boilerplate answer? what should it contain?
Topic 3: Web Site Update
- new Expert Handlers Resource: Semantic Web Activities & Initiatives of Interest to Expert Handlers
- adjourned: h UTC
- Agenda for 2008/04/06 Expert Handlers Call