Minutes: MathML Intent meeting, 28 Jan

Attendees:

Neil Soiffer

Bruce Miller

David Carlisle

Deyan Ginev

David Farmer

Sam Dooley

Murray Sargent

Charles LaPierre

Louis Maher

Laurence Zaysser

Patrick Ion

Steve Noble

The meeting was recorded:

https://us02https://us02web.zoom.us/rec/share/Y_XfeY7gHcVqYwmDQuBC.cA1rvKQs4Fd7ppI7
<https://us02web.zoom.us/rec/share/Y_XfeY7gHcVvGRWatGtVkQHBKX2_FeAFFp0YSnUuyJ97UQTGjbWwgITqYwmDQuBC.cA1rvKQs4Fd7ppI7>

 Passcode: 3QKJV*4m



Thanks to Louis for taking notes.

Agenda:

1. Any followup to Neil's presentation on AT needs

https://mathml-refresh.github.io/mathml/docs/a11y-needs.html

NS: It will be until mid March before the charter will be approved.  Must
change delivery dates in the charter.

We re-introduced ourselves to Laurence.

BM: Was surprised that AT does not use Intent.

DG: What do you do when you can’t figure out what is going on? You can’t
list all the math notations. DG has a catalog of almost all the math
symbols.

BM: What is the minimal amount that you need to deal with accessibility?


2. Deyan's latest demo showing how level 3 math can be spoken (doing screen
share)

DG: Listing all math symbols is a large task. It involves thousands of
symbols.

DG: We need a mechanism for AT to use symbols not on the list.

DG: Looked at wikipedia examples to find unusual math examples that are
difficult to speak.

DG: Has a method for users to add their symbols to the symbol catalog.

DG: His demo used a generic speech engine.  It was not specialized for
mathematics.

NS: There are about 30 entries with speech. Is this because you haven’t
gathered speech for all 1000 entries or because only 30 seem to have
specialized speech.

DG: The former.

DG: The authors are not required to develop code to speak their new
notation.

DG: The authors can enter speech hints.

LM: Do users provide their own speech?

DG: Interesting idea. It could go either way. Depends on what the system
implements.

DF: Sometimes you must say begin quantity and end quantity for an accurate
description.

BM: Vendors might want to supply their own proprietary symbol lists.

NS: There should not be much lag in this system.  More than 0.2 seconds
delay would be bad.

BM: What kind of status would the spreadsheet have?

NS: It’s for us to decide. I don’t think it should be required that AT uses
it. Maybe it is a note with a recommendation that it be consulted.

BM: Need to worry about it being hacked

DG: We could restrict authoring access.

NS: Vendors might want to download and vet and distribute with their
product if they don’t trust the spreadsheet.

??: Might be faster access if it is local.

3. Plans and deadlines for the year

NS: We have many deliverables including MathML specification 4 and test
cases. NS: How do we add tests for the accessibility features. They involve
both authoring and reading.

NS Here’s the timeline we have in the charter:
Timeline

These dates are not meant as deadlines; work may need to continue after
these dates to ensure the quality of the deliverables.

   -

   February 2021: First teleconference
   -

   October 2021: First (virtual?) face-to-face meeting (TPAC)
   -

   March 2021: FPWD, MathML Core Level 1
   -

   June 2021: FPWD, MathML 4
   -

   November 2021: MathML Core Test Suites
   -

   November 2021: Implementation Report
   -

   November 2021: Candidate Recommendation, MathML Core Level 1
   -

   January 2022: Sample code that adds annotations making defaults explicit
   -

   February 2022: Code that converts Content MathML to annotated
   Presentation MathML
   -

   March 2022: FPWD, MathML Core Level 2
   -

   May 2022: Sample code that uses annotated Presentation MathML to create
   Speech and/or Content MathML
   -

   June 2022: Prototype TeX-to-annotated MathML implementation
   -

   October 2022: Living catalog established for extended Presentation
   annotations
   -

   November 2022: Candidate Recommendation, MathML 4


NS: We need to get started on creating a MathML 4 spec that aligns with
core. Volunteers?

DC: Offered to update MathML 4.

NS: I plan to make use of the ‘intent’ in my MathPlayer rewrite.

DG: We are ahead of schedule on the prototype TeX-to-annotated MathML
implementation

DG and BM: will talk in a couple of weeks on their LatexML system that can
generate ‘intent’ values