Minutes: MathML general/semantics meeting 12/11/19

Agenda:
https://github.com/mathml-refresh/mathml/issues/91#issuecomment-550460421

Meeting was recorded:
https://benetech.zoom.us/recording/share/XP5Bsgesnlompm-iZGIyYqdjrXRJa2fOgKAmbk715PiwIumekTzi
MwAttendees:

Neil Soiffer

David Farmer

Charles LaPierre

Sam Dooley

David Carlise

Regrets: Steve Noble
Progress on Core:

   -

   Fred’s talk
   <https://people.igalia.com/fwang/mathml-in-browsers-web-engines-hackfest-2019/slides>
   at Igalia hackfest Chromium
   -

   first patches landed
   -

   BlinkOn 11 - two MathML talks (brian, fred) will be livestreamed/recorded
   -

   TAG Review process #438
   -

   MathML conference next year?
   NS: probably at Igalia’s HQ in A Coruna, Spain. No idea when or format.

CL: Brian Kardell showed Igalia’s MathML at TPAC. It constantly refreshed a
bunch of equations. It was really fast.

CL: I was a meeting at Microsoft and at a Daisy board meeting where
Microsoft was present. Edge is moving to a Chromium-based implementation.
They are watching what Igalia is doing and it will be easy to pull that
work in when it happens. No guarantee that they will take it though,
although they were very impressed with what they have done to date.

There wasn’t much note taking for the main part of the meeting. Those
interested in details will need to listen to the recording. The main points
were:

SD: for the sake of argument, let’s call the attr “mathrole”. Should
“mathrole” be informative or normative?

SD: If one or the other, does that limit its use

SD:  We need to know how the information is used.

NS: That’s still a bit nebulous. Export to computation systems, to things
like DF’s PreText, for speech, for braille.

NS: The values probably needs to be open-ended


DF: Discussion of lines 474 and 475 in SD doc (see agenda) about imaginary
units. They have the same semantics. The problem comes up with “i”, which
can be a summation index or it can be an imaginary number.

What seems to be missing from SD’s chart is a column that says what the
semantic is for the entries, and there may be many different semantics for
some things. For most, there is a default semantics and there aren’t any
other meanings (in non-obscure math). The default meanings need to be
listed somewhere, but most are obvious (“sin” is “sine”). The part we need
to focus on is those where the symbol has several meanings.

There was a lot of discussion between SD and the rest of the group of the
necessity of some columns in the table.

DF: I’ve gone through several textbooks now looking for areas where there
are ambiguities. Sometimes the subject tells you a lot. Sometimes not. In
combinatorics, “×” can mean different things in the same sentence. For
example: “The 8×8 checkerboard has 8×8 squares”. (the 8 by 8 checkerboard
has 8 times 8 squares). Another meaning is cartesian product. It was split
almost equally among all the meanings. You needed natural language
understanding to distinguish many of these meanings. I’ve gone through an
algebra textbook, pre-calculus, a couple of combinatorics books, calculus,
and I’m working on probability.

Discussion of how to format the data involving semantics, symbol used, and
subject area. Put into a spreadsheet and could sort by column?

DF: I’m working on

https://docs.google.com/document/d/1cZnff5_fi_ucNyZ1ex2msmJLE55FAZD-QInkLYe8xiE/edit

Next meeting is next week.