Minutes: MathML Full meeting 26 June, 2025

 Attendees:

   - Neil Soiffer
   - Louis Maher
   - David Carlisle
   - Bruce Miller
   - Paul Libbrecht
   - Deyan Ginev
   - Murray Sargent
   - Moritz Schubotz
   - Bert Bos

The Zoom-generated summary of this meeting has been placed at the end of
the minutes under the level-one heading "Zoom Meeting Summary".
<https://cryptpad.fr/#cp-md-0-action-items>Action Items
<https://cryptpad.fr/#cp-md-0-1-announcements-updates-progress-reports>1.
Announcements/Updates/Progress reports

*ACTION:* Brian to post the agenda for the upcoming Core meeting on Monday.
<https://cryptpad.fr/#cp-md-0-2-a-href-https-github-com-w3c-mathml-issues-534-534-summary-of-mathml-polyfill-needs-a->2.
#534: Summary of MathML Polyfill needs
<https://github.com/w3c/mathml/issues/534>

Progress reports Remaining polyfills (volunteers?)
<https://cryptpad.fr/#cp-md-0-3-there-are>3. There are

9 open issues with the "needs spec update" label.
<https://github.com/w3c/mathml/issues?q=is%3Aissue%20state%3Aopen%20label%3A%22need%20specification%20update%22>
Review issues and see who feels like they can contribute

*ACTION:* DC will handle Remove alignmentscope from MathML 4 spec issue 535
<https://github.com/w3c/mathml/issues/535>

*ACTION:* bring up nested math in token elements issue 522
<https://github.com/w3c/mathml/issues/522> in the next meeting.

*ACTION:* DC will handle XSD not allowing (semantics)-element as child of
(apply)-element: issue 498 <https://github.com/w3c/mathml/issues/498>.

*ACTION:* DG: Find out what we agreed to do about Intent Properties:
ordering & references · Issue #449
<https://github.com/w3c/mathml/issues/449>.

*ACTION:* NS: move this to the core meeting Spec should specify what char
to use for accents/lines · Issue #247
<https://github.com/w3c/mathml/issues/247>.
<https://cryptpad.fr/#cp-md-0-agenda>Agenda
<https://cryptpad.fr/#cp-md-1-1-announcements-updates-progress-reports>1.
Announcements/Updates/Progress reports

*ACTION:* Brian to post the agenda for the upcoming Core meeting on Monday.

DC: NVDA 2025.1.2 was released. For the first time, the math in PDF gets
read without having to load development versions of the whole stack. So we
redid all the videos.

DC: We have a macro for TeX that can generate intent.
https://latex3.github.io/tagging-project/documentation/wtpdf/fulldoc

PL: Can MathML use copy and paste? NS says yes in NVDA. NS said copy and
paste works regardless of the source, including PDF.

NS experienced two cases where intent would have helped when working on a
braille translation.

NS received a request for help. That person had a Word document, containing
something like H single bond H, which is just below the threshold that
MathCAT recognizes as chemistry. The document was generated from ChemType,
which is Wiris' chemistry version of MathType.

NS tweaked MathCAT to recognize a chemical equation in this case, but his
change is not a general solution.

NS suggested that the caller ask Wiris to add intent to this expression
identifying it as chemistry. The caller got a generic response from Wiris.

NS asked PL to help in this situation because PL has contacts at Wiris.

NS's second situation where intent would be useful was to recognize when he
was reading a minus sign. In the Polish braille code, it has one set of
dots when it is binary minus and another one when it represents
complement/set difference in set theory.
<https://cryptpad.fr/#cp-md-1-2-a-href-https-github-com-w3c-mathml-issues-534-534-summary-of-mathml-polyfill-needs-a->2.
#534: Summary of MathML Polyfill needs
<https://github.com/w3c/mathml/issues/534>

Progress reports Remaining polyfills (volunteers?)

Concerning Mathvariants: Deprecate/Remove mathvariant attributes Issue #77
<https://github.com/w3c/mathml/issues/77>

MuS: We want to translate Characters to their Unicode code points and get
rid of the math variant, And, so I wrote a polyfill to do that, and I used
a table that I inherited, actually from Noah Dersing, who wrote the
original Unicode MathML program that I started with and have been
developing ever since.

MuS' polyfill converts the math alphanumerics into the corresponding styled
characters.

MuS prefers JavaScript to C++ for this case.

MuS' results agree with MOs' table.

DC: The algorithm should be extendable and understandable, because, for
instance, the Unico TR25 work may restart, and we may get more ranges to
add.

MuS: I wanted to just spend a little more time on the algorithm to put in
the Greek and the Arabic.

NS: I had worked on a mlabeledTR update to take advantage of what the
intent for row label. It got stalled, and I need to push on it to get it
updated.

NS: I had started on a set of polyfills for subscript and superscript
shifts that I'm stuck on. The problem is that I need to measure the height
and depth of the subscript to know how much I need to shift it. I am
getting unreasonable and very different numbers from Firefox and Chrome. I
would like an interface that just gave you these things. Maybe this is an
issue for core.

NS: mtable and maction are left. The selection attribute on maction should
be accepted and should be an easy polyfill for anyone wanting to try and do
a polyfill.

DC to BB: there is a cosmetic release of CSS improvements to the core spec.
But if I were to push that, it would rebuild and republish with the new
system, so I didn't want to change the date during the CR. We could rebuild
after the CR.

NS: I would think you should be allowed to update during CR, because the
whole point is you're getting feedback and making changes based on the
feedback.

DC: W3C policy would never change the CR draft, during the CR.

BB: We can update the CR, indeed, but the automatic publishing system Still
has a working draft hard-coded of its own, The script on GitHub needs to be
updated, but otherwise, it's fine to publish a new CR.

NS: I think that's it for the polyfills. Anybody else have anything for the
polyfills?

MuS: Well, I had one quick question, especially for David. You know, we did
add the ability to distinguish between two kinds of scripts, the round hand
and And the Chancery. Do we want to support that in MathML?

DC: We were trying to deprecate mathvariants so adding one for a script is
not desirable.
<https://cryptpad.fr/#cp-md-1-3-there-are>3. There are

9 open issues with the "needs spec update" label.
<https://github.com/w3c/mathml/issues?q=is%3Aissue%20state%3Aopen%20label%3A%22need%20specification%20update%22>
Review issues and see who feels like they can contribute

*ACTION:* DC will handle Remove alignmentscope from MathML 4 spec: issue 535
<https://github.com/w3c/mathml/issues/535>

*ACTION:* bring up nested math in token elements: issue 522
<https://github.com/w3c/mathml/issues/522> in the next meeting.

*ACTION:* DC will handle XSD not allowing (semantics)-element as child of
(apply)-element: issue 498 <https://github.com/w3c/mathml/issues/498>.

*ACTION:* DG: Find out what we agreed to do about Intent Properties:
ordering & references Issue #449 <https://github.com/w3c/mathml/issues/449>.

NS: If you have the same font in Firefox and Chrome, and they do something
different with the same character, then the core spec should be updated.

*ACTION:* NS: move this to the core meeting Spec should specify what char
to use for accents/lines Issue #247
<https://github.com/w3c/mathml/issues/247>.

DG to NS: Do you have a list of accent characters? NS has a partial list.
<https://cryptpad.fr/#cp-md-0-zoom-meeting-summary>Zoom Meeting Summary

Meeting summary for Math WG (2025-06-26)
<https://cryptpad.fr/#cp-md-0-quick-recap>Quick recap

The meeting covered discussions on math variant characters, MathML
specifications, and accent character display issues across browsers and
fonts. The team addressed implementation challenges, including the
placement of math elements within mtext elements, and agreed to review
previous decisions on script support. They also planned to update
documentation for agreed decisions and scheduled future meetings to
continue addressing these topics and reviewing outstanding issues.
<https://cryptpad.fr/#cp-md-0-next-steps>Next steps

   - Neil to add issue #247 (Specify what character to use for accents) to
   the agenda for the upcoming Core meeting with Brian.
   - Neil to compare accent rendering between Firefox, Chrome, and Safari
   for issue #247.
   - Deyan to write down the agreed-upon direction for issue #449 (Intent
   properties ordering and references).
   - David to fix the bug in issue #498 (XSD not allowing semantics).
   - Neil to review the list of spec update issues before the next meeting
   and identify which ones need special consideration.
   - Brian to post the agenda for the upcoming Core meeting on Monday.

<https://cryptpad.fr/#cp-md-0-summary>Summary

Math Variant Characters Documentation Issues David discussed issues with
math variant characters, noting inconsistencies and difficulties in
documentation due to arbitrary character support across different fonts. He
mentioned working on filling gaps, such as italic digits and Greek
characters, but expressed doubt about the proposal's acceptance. Bert
confirmed that cosmetic CSS changes to the core specification could be made
during the CR period, and David agreed to proceed with minor table
alignment updates. Murray raised a question about supporting round-hand and
chancery scripts in math ML, but David suggested avoiding adding new math
variants and instead focusing on separate solutions. The team agreed to
review the minutes for any prior decisions on script support.
<https://cryptpad.fr/#cp-md-0-mathml-implementation-discussion>MathML
Implementation Discussion

The team discussed MathML specifications and implementation issues,
particularly regarding the placement of math elements within mtext
elements. David explained that while the HTML spec allows math as a child
of mtext, this needs specification work and is not yet approved. Neil
realized he had made an assumption about implementation progress after the
addition of a need specification update. The team agreed to bring this
topic up for discussion at the next meeting, and David mentioned he would
address a separate bug related to XSD conversion. The conversation ended
with a reminder that the team needs to write documentation for previously
agreed decisions, which has been pending for about half a year.
<https://cryptpad.fr/#cp-md-0-accent-character-display-issues>Accent
Character Display Issues

The team discussed issues related to accent characters and their display
across different browsers and fonts. Neil assigned Deyan to review a
CodePen comparison of Firefox and Chrome accent character rendering, which
revealed significant differences in how characters are displayed. The group
agreed that the core specification may need updating, and Neil will add
this topic to the upcoming core meeting agenda. They also confirmed the
next meeting will be held on July 3rd, with plans to continue discussing
accent character standardization and review the full list of issues.

AI-generated content may be inaccurate or misleading. Always check for
accuracy.