Minutes: MathML Full meeting 30 April, 2026

 Attendees:

   - Neil Soiffer
   - Louis Maher
   - David Carlisle
   - Deyan Ginev
   - Murray Sargent
   - Peter Wu
   - Bruce Miller

<https://cryptpad.fr/#cp-md-0-regrets>Regrets
<https://cryptpad.fr/#cp-md-0-action-items>Action Items
<https://cryptpad.fr/#cp-md-0-2-issues-to-review->2. Issues to review:
<https://cryptpad.fr/#cp-md-0-a-can-we-drop-needs-spec-update-from->a. Can
we drop needs-spec update from:
<https://cryptpad.fr/#cp-md-0--a-href-https-github-com-w3c-mathml-issues-1-1-simplification-of-the-mstyle-element-a->#1
Simplification of the mstyle element
<https://github.com/w3c/mathml/issues/1>

NS: We dropped needs-spec update from core issue 1.
<https://cryptpad.fr/#cp-md-0--a-href-https-github-com-w3c-mathml-issues-181-181-mathml-4-extensions-for-alignment-and-possible-deprecation-of-maligngroup-and-malignmark-a->#181
MathML 4 extensions for alignment and possible deprecation of
maligngroup/and malignmark <https://github.com/w3c/mathml/issues/181>

NS: We dropped needs-spec update from core issue 181.
<https://cryptpad.fr/#cp-md-0--a-href-https-github-com-w3c-mathml-issues-385-385-comments-on-chapter-3-a->#385
comments on chapter 3 <https://github.com/w3c/mathml/issues/385>?

NS: We closed core issue 385.
<https://cryptpad.fr/#cp-md-0-b-review-a-href-https-github-com-w3c-mathml-issues-534-534-summary-of-polyfill-needs-a->b.
Review #534: Summary of Polyfill needs
<https://github.com/w3c/mathml/issues/534>

*ACTION:* DC wrote: We need to add rollup or similar JavaScript packaging.

*ACTION:* We need to have a meeting on getting people to write polyfills.
<https://cryptpad.fr/#cp-md-0-3-selection-attributes>3. Selection Attributes

*ACTION:* MuS will create an issue putting selection attributes into MathML
so that you know where the user's selections are.
<https://cryptpad.fr/#cp-md-0-agenda>Agenda
<https://cryptpad.fr/#cp-md-0-1-announcements-updates-progress-reports>1.
Announcements/Updates/Progress reports

LM: Apologies folks: The chat-based links were lost because my method to
transfer information from the Zoom chat to the markdown-formatted CryptPad
minutes is text based. I do not know how to accommodate for this in
real-time. It would be helpful to put web addresses, for links, in the chat.

From Murray Sargent to everyone: Microsoft wants to add selection
attributes to MathML. Here's an early proposal MathML and OMML User
Selection Attributes

   - Math in Office

MuS: Peter Wu (PW) Peter has been doing amazing work With our MathML reader
and writer, and the LaTeX interactions and Word and PowerPoint. He used
Claude a little bit.

PW: Microsoft has a new LaTeX converter.

PW: Microsoft is still generating maligngroup and malignmark. If someone
knows what they should be generating instead, please let PW know.

NS: I believe there is a one-to-one mapping into an mtable when compared
with maligngroup and malignmark.

Pw has asked for pseudo code to see how that would map to MathML core.

DG: You said you have a new LaTeX converter. What did you base the LaTeX
coverage on? So what kind of LaTeX do you cover and how do you choose?

NS: I think DG's question is really about which packages you cover.

PW: Excellent question. I will share some material with you on that.

From petewu to everyone: LaTeX Support in Microsoft 365

From petewu to everyone: LaTeX mhchem Support in Microsoft 365
<https://cryptpad.fr/#cp-md-0-no-meeting-next-week>No meeting next week
<https://cryptpad.fr/#cp-md-0-charter-status>Charter status

NS: I still haven't seen anything where people have pushed the charter
along.

NS: One more group ticked off the box, I think it was security, for the
horizontal review of the charter status.
<https://cryptpad.fr/#cp-md-0-horizontal-review-status-for-cr>Horizontal
review status for CR
<https://cryptpad.fr/#cp-md-0-from-deyan-a-href-https-www-sovereign-tech-news-join-sovereign-tech-standards-network-sovereign-tech-funding-a->From
Deyan: Sovereign Tech funding
<https://www.sovereign.tech/news/join-sovereign-tech-standards-network>

DG: This is a group that will fund people for doing standards work.

From Deyan Ginev to everyone: Sovereign Tech Standards:
https://www.sovereign.tech/programs/standards

From Deyan Ginev to everyone: "W3C is very pleased to announce that the
Sovereign Tech Agency has launched a pilot program to support independent
open source maintainers participation in the work of our groups, including
training, mentoring, and financial support to make sustained engagement
possible."
<https://cryptpad.fr/#cp-md-0-action-item-follow-up->Action Item follow up:

From NS: WCAG intent issue <https://github.com/w3c/wcag/issues/5093> from
last week action items.

NS is trying to get intent into WCAG.
<https://cryptpad.fr/#cp-md-1-2-issues-to-review->2. Issues to review:
<https://cryptpad.fr/#cp-md-1-a-can-we-drop-needs-spec-update-from->a. Can
we drop needs-spec update from:
<https://cryptpad.fr/#cp-md-1--a-href-https-github-com-w3c-mathml-issues-1-1-simplification-of-the-mstyle-element-a->#1
Simplification of the mstyle element
<https://github.com/w3c/mathml/issues/1>,

NS: We dropped needs-spec update from core issue 1.

NS: Core issue 1 needs polyfills.
<https://cryptpad.fr/#cp-md-1--a-href-https-github-com-w3c-mathml-issues-181-181-mathml-4-extensions-for-alignment-and-possible-deprecation-of-maligngroup-and-malignmark-a->#181
MathML 4 extensions for alignment and possible deprecation of maligngroup
and malignmark <https://github.com/w3c/mathml/issues/181>

NS: We dropped needs-spec update from core issue 181.

NS: Maybe we can even get rid of this, which would be great, Peter, if you
solve the problem.

PW: Perhaps we can if someone gives us guidance.

PW: We have not done anything with it yet.

DC: We basically kept it in because Word was still using it. Word was the
only one that was using it.

MuS has examples when maligngroup and malignmark are created within
Microsoft word.

MuS: We would probably keep it around because there are older versions of
Word around.

NS: Yeah, so the polyfill needs to be around. But whether we need to
support it in MathML 4 is a different question.
<https://cryptpad.fr/#cp-md-1--a-href-https-github-com-w3c-mathml-issues-385-385-comments-on-chapter-3-a->#385
comments on chapter 3 <https://github.com/w3c/mathml/issues/385>?

NS: We closed core issue 385.

DC: If there's anything specifically wrong in the CR draft, we should just
open a new issue.
<https://cryptpad.fr/#cp-md-1-b-review-a-href-https-github-com-w3c-mathml-issues-534-534-summary-of-polyfill-needs-a->b.
Review #534: Summary of Polyfill needs
<https://github.com/w3c/mathml/issues/534>

NS: This is a summary of polyfill needs.

*ACTION:* DC wrote: We need to add rollup or similar JavaScript packaging.

Ns: Linebreaking breaks the rollup process.

DC: The rollup process will work if you put the linebreaking last.

NS wanted to encourage people to write polyfills.

NS: Perhaps we can get AI to write polyfills.

NS wants to have a meeting on polyfill writing.

NS: The idea is to be able to take MathML 3, also MathML 4, and have it
work with MathML core.

*ACTION:* We need to have a meeting on getting people to write polyfills.
<https://cryptpad.fr/#cp-md-1-3-selection-attributes>3. Selection Attributes

MuS: Well, we're interested in adding selection attributes to the MathML we
generate so that ATs can figure out where things are in the Braille.

MuS: Maybe this is a MathML 5 thing, but it is something that the
accessibility community is really interested in.

*ACTION:* MuS will create an issue putting selection attributes into MathML
so that you know where the user's selections are.
<https://cryptpad.fr/#cp-md-0-zoom-intent-meeting-summary-for-4-30-2026>Zoom
Intent Meeting Summary For 4/30/2026 <https://cryptpad.fr/#cp-md-0-summary>
Summary

The MathML working group meeting focused on updates to LaTeX converter
improvements, and polyfill development needs. Peter Wu presented
significant progress on a new LaTeX converter achieving 99.5% coverage
based on Math Stack Exchange data, and discussed the implementation of PDF
2.0 compliant MathML in Word and PowerPoint. The group reviewed and updated
several issues related to maligngroup and malignmark extensions and
polyfill requirements, deciding to close some older issues where
specifications had been improved. Murray raised the need for adding
selection attributes to MathML to improve accessibility, which will be
addressed in an upcoming meeting. The group also discussed a new W3C
funding opportunity for independent developers to contribute to working
groups.
<https://cryptpad.fr/#cp-md-0-mathml-import-and-export-improvements>MathML
Import and Export Improvements

The team discussed improvements to MathML import and export functionality,
including a new LaTeX converter available to Microsoft 365 Insider Program
subscribers. They explored mapping malignmark and malign formats to MathML
Core, with Neil suggesting a simple approach using alignment pairs and
requesting pseudocode examples from David. The discussion also touched on a
web polyfill for MathML, with Murray noting its decent performance except
for issues with vertical alignment in complex objects. Deyan asked about
the basis for the new LaTeX converter, though the specific details were not
addressed in the transcript.
<https://cryptpad.fr/#cp-md-0-latex-package-support-methodology>LaTeX
Package Support Methodology

Peter Wu explained the methodology behind choosing which LaTeX packages to
support, detailing their decision to focus on real-world usage found on
Math Stack Exchange. The new converter achieved 99.5% coverage of keywords
from this source, compared to 90% with the old converter and 93.5% after
previous improvements. Peter clarified that while user-defined macros are
supported, they must be self-contained within each equation rather than
defined globally at the document level.
<https://cryptpad.fr/#cp-md-0-latex-conversion-features-discussion>LaTeX
Conversion Features Discussion

The team discussed feedback collection on LaTeX-to-Word/PowerPoint
conversion features and usage patterns. Deyan shared Morris's idea about
standardizing web tech use, particularly regarding MathJax and KaTeX
rendering. The group explored data sources for analyzing LaTeX usage, with
David mentioning package usage data from the LaTeX Accessibility Project
Archive and Peter noting challenges with extracting math-specific commands
from general LaTeX content. Petewu also announced the implementation of
MathML in PDF export for Word and PowerPoint in beta, now compliant with
PDF 2.0 associated files format.
<https://cryptpad.fr/#cp-md-0-w3c-funding-and-mathml-updates>W3C Funding
and MathML Updates

The team discussed skipping next week's meeting due to Neil's absence and
lack of pressing issues. Deyan shared information about a new W3C funding
opportunity from Sovereign Tech Fund for independent developers to
contribute to working groups and standards work, particularly relevant for
groups like MathML. Neil mentioned submitting an issue to WCAG's issues
list regarding MathML intent implementation, and noted that MathML is
currently included in WCAG as part of appropriate semantics for web math
content.
<https://cryptpad.fr/#cp-md-0-maligngroup-and-malignmark-extensions-spec-update-discussion>maligngroup
and malignmark Extensions Spec Update Discussion

The team discussed removing the "needs spec update" tag from several items,
including mstyle simplification and malign FML extensions for alignment and
deprecation. They determined that while the malignGroup and malignmark
polyfills are still needed due to their use in Word, particularly for
aligned equations, the more complex nested implementations are rarely used
in practice. Peter agreed to review the malign extensions once guidance is
provided, and the team noted that the current simple implementation should
be sufficient for real-world usage since Microsoft is the primary generator
of this content.
<https://cryptpad.fr/#cp-md-0-documentation-examples-review-meeting>Documentation
Examples Review Meeting

The team discussed the current state of examples in their documentation,
with David noting there are fewer examples available compared to 2022. They
reviewed specific examples including Menclosed and elementary math, and
determined that images are being used where the polyfill is not perfect.
David concluded that Issue 385 should be closed since general issues about
everything being wrong in the spec are difficult to address, and specific
problems in the CR draft should be handled through separate issues.
<https://cryptpad.fr/#cp-md-0-polyfills-and-accessibility-updates>Polyfills
and Accessibility Updates

The team discussed updates to polyfills, with Neil suggesting using AI to
help create some of these tools. David mentioned having a roll-up package
ready but needing to push it to the repository. Murray raised the topic of
adding selection attributes to MathML for better accessibility, which will
be addressed in a future meeting. The group agreed to focus on polyfills in
their next meeting in two weeks, with Murray creating an issue for the
selection attributes topic.