Minutes: MathML intent meeting, 18 Aug

Reminder: there is no meeting this week due to US Thanksgiving.

============
Attendees:

   -

   David Carlisle
   -

   Sam Dooley
   -

   David Farmer
   -

   Deyan Ginev
   -

   Patrick Ion
   -

   Louis Maher
   -

   Bruce Miller
   -

   Murray Sargent
   -

   Neil Soiffer
   -

   Steve Noble
   -

   Moritz Schubotz
   -

   Cary Supalo
   -

   Bert Bos
   -

   Paul Libbrecht
   -

   Brian Kardell
   -

   Stephen Watt

Regrets:

   - Daniel O'Mahony

Announcements/updates

(Upcoming meetings/holidays)

We will skip next week and meet on December 2. Our last meeting of the year
will be December 16.

NS: would like us to have a plan for intent by the end of the year.
What Has Been Discussed

Bruce's material:

Proposal:

https://mathml-refresh.github.io/discussion-papers/semantics-mini

I think Deyan's demo page (with some options selected) mostly reflects the
proposal:

https://dginev.github.io/tiny-mathml-a11y-demo/

Sam's material:

Proposal:

https://samdooley.github.io/mathml-docs/intent2cmml/intent.html

Demo:

https://samdooley.github.io/mathml-docs/intent2cmml/demo.html

(updates on paste)

Discussions:

Initial merge proposal from David C:

https://github.com/w3c/mathml/issues/252

Deyan's pullout on n-ary intent:

https://github.com/w3c/mathml/issues/253
Deyan is stepping up to the plate this week with his version of the intent
proposal

DG: presented his proposal.

https://prodg.org/talks/encyclopedic-intent#/

PL: What is the difference between a local and a global definition?

dg: A global name has reached a textbook and is accepted where a local term
may change from place to place.

BM: Where is the dividing line between local and global definitions. If
something is not in the dictionary, then you pronounce the term as it
appears in the MathML.

BK (could have been DC): The ARIA group did not want aria-labels used
because they were not interactive.

NS: The ARIA group has not thought much about it. They said the aria-labels
would not solve our accessibility problems.

BK: thought that we should not use aria-labels for non-interactive terms.

There is an upcoming meeting at which time we can discuss this.

NS: The main source of intent language will come from MathML generators.

NS: The authors should not have problems (also called friction) to get
intent into the MathML. That is, it should be easy to enter intent into the
MathML.

MS: There is a double factorial symbol in Unicode.

MS: Does not see why intent is necessary. He does approve of defaults that
will let you generate good speech without putting in extra parameters. He
cannot think that authors will enter intent parameters. They would use
LaTeX grammar and not intent. He suggests adding new LaTeX macros to
disambiguate things. For the time being, we can live with content MathML
and aria-labels without new intent grammar.

DG: You cannot speak content MathML directly. He believes that Intent is
still the easiest way to make things accessible.

SW: He is trying to understand the difference between an annotation element
versus an intent element. He suggests that annotation may be sufficient for
accessibility.

NS: Most MathML is presentation MathML, and that content MathML has not
been used widely.

PL: Nesting elements is difficult. Annotation may not be able to manage
this easily.

DG: Would like to change content MathML to make it easier to generate.

NS: Does anyone else want to put up their ideas after today's discussion?
If so, let NS know.

NS: We will continue this discussion in two weeks.