Minutes: MathML intent meeting, 23 Sept

 Attendees:

   - David Carlisle
   - Sam Dooley
   - Mohannad Khasawneh
   - Louis Maher
   - Bruce Miller
   - Murray Sargent
   - Neil Soiffer
   - Steve Noble
   - Cary Supalo
   - Bert Bos
   - Stephen Watt

Regrets:

   - David Farmer
   - Patrick Ion
   - Paul Libbrecht
   - Daniel O'Mahony
   - Moritz Schubotz

Announcements/updates

NS: I added our request for a breakout session to the TPAC wiki page (
https://www.w3.org/wiki/TPAC/2021/SessionIdeas#From_MathML_to_AT)
Gap Analysis Doc

The document is at (
https://docs.google.com/document/d/1Pzsf2dqTkbYXFmoKCLzvtrNERNmtF2MHX1WtugB3PfA/edit#heading=h.k21427jotzf1
).

SD: The document has become a discussion on vocabularies. He wants a
section on what to say about vocabularies.

NS: Does not think that there is a need for a vocabulary discussion.

SD: There are some things that are known and some things that are unknown.
There should be a high-level discussion. There are math symbols that are
not in the standard.

BM: Agrees with SD:. Steer away from pragmatic and stick with csymbol.

*ACTION* SD: Will write a one paragraph description of this.

SD: Math is an opened ended field where people are creating vocabularies
all the time. There are common vocabularies.

SW: Likes to say that content MathML has predefined symbols. There are ways
to add symbols to it. The gap document may conflate semantics with
accessibility. Semantics might describe the math without knowing how to
pronounce it. The needs of accessibility overlap with semantics but are not
solved by semantics.

SD: Semantics is an overloaded term.

SW: Wants this thought to be up front in the document.

SW: Sometimes the word semantics may not be adding extra meaning and may
not be needed in some of the sentences of the gap document.

SW: The first section should have three sentences with these ideas in mind.

BM: agrees this should be discussed at the beginning. How do you speak
math? Sometimes the best method is to read what is there without
interpretation. Other times it is better to interpret the math.

*ACTION* SD: will write a stub vocabulary section. He will also look to see
if the word semantic is used clearly.

NS: wants to move the document into W3C format.

*ACTION* SW: will take a stab at the goals section.

SW: WE should say something about insufficiency without saying what should
be deprecated.

WS: We should say what is there and why it may be insufficient. Also, we
should say what needs to be added.

NS: would like to have this done for each section.

NS: Is there any move towards improving speech generation from CSS?

BB: There is no activity to generate speech with CSS.

NS: Says Brian is unhappy with efforts to get speech on the web.

NS: Would like the parallel section shrunk a little.

BM: The parallel section is missing what it does and does not address.

NS: This must be done by next Tuesday September 28.

BM: There are two approaches. say this is how this expression should be
pronounced and enforce this through presentation MathML. Or, say that the
presentation MathML disambiguates the expression, and you write how it
should be spoken.

NS: Should AT be speaking what is there instead of trying to interpret the
math? Neil believes AT should interpret an expression.

NS: Process wise, after next Tuesday September 28, the gap document needs
three editors.

*ACTION* DC: can help with editing.

NS: will do it if no one else volunteers. The breakout session schedule
will be generated on October 8. The gap document should be published on
October 9.

NS: Can the working group create notes somewhere without asking for
permission?

*ACTION** BB: will make a new directory for the WG to store new documents.
The directory should be called MathML-Docs.