MathOnWeb CG -- meeting minutes 2018-08-16

Hi everyone,

Here are the minutes from the meeting on August 16.

Best,
Peter.

# MathOnWeb CG meeting 2018-08-16

* Present: Arno, Volker, Kevin, Peter, Neil
  * Regrets: Joanie, Charles
* Peter: recap a11y TF
  * [intial call about AAM, second call with good discussion about
dictionaries vs heuristics, UX]
  * Arno: so what canwe prepare for TPAC?
  * Volker: my main point is that the dictionary approach has been tried
repeatedly and failed repeatedly
    * it seems rational to look for a different approach instead of
repeating the same mistake
* Arno: what else?
  * Peter: I'd still like to talk about making aria-tree or grid walking
easier, getting things to work for the lightweight walker we build at the
workshop
  * Peter: otherwise, I think the TF is on a good track
* Volker: what else could we bring to the ARIA at TPAC?
  * getting rid of JS for grid/tree walking would be nice
  * maybe exposing the relevant data directly to AT?
  * Volker: we should also look at aria tree and grid before that ;-)
* Arno: so we should have a goal of having fairly concrete items?
  * => yes
  * Peter: there's the "math" AAM, the walking problem, braille
annotations/inputs/liveregions
  * Volker: can we start a page? Google doc?
    * => yes
    * Peter: use
https://github.com/w3c/mathonwebpages/wiki/MathOnWeb-CG-@-TPAC-2018 ?
* Peter: CSS task force needs to get back on track - let's do a similar
page?
  * => yes
* [some discussion about upcoming releases]
* Arno: with ARIA I'd like to bring up "Braille mixing" in strings
  * would like to be able to just mix and let AT sort out what's braille
and what's speech
  * Volker: maybe AOM will help here, too?
  * Arno: the trouble I have with braille-specific labels etc is that you
might lose one or the other
    * I'd rather keep them together
    * Volker: yes
  * Peter: any chance they cannot be brought together? (Due to differences
in voicing and Braille)
    * Arno: seems unlikely
    * Volker: but if so, it's another argument for separate streams
  * Neil: there's an issue around focus on Braille
    * also Braille is verbose, multiline braille displays not really on the
market / very expensive
  * Arno: I think one should be able to mix speech and braille as Unicode
    * if there's both in a string, AT should split them up and send them
each way
    * else AT does its thing
  * Neil: has anyone talked to AT people?
    * Volker: brought this up with NVDA at AIM and they had a technical
reasons
  * Arno: talked to Joanie, similar response

* ACTIONS: start work on wiki pages for TPAC, collecting resources for CSS
and APA/ARIA meetings