Minutes: MathML Full Meeting, 14 May, 2026

 Attendees:

   - Neil Soiffer
   - Louis Maher
   - David Carlisle
   - Deyan Ginev
   - Bruce Miller
   - Murray Sargent
   - Paul Libbrecht

<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-progress-on-a-href-https-github-com-w3c-mathml-issues-534-534-summary-of-polyfill-needs-a->2.
Progress on #534: Summary of Polyfill needs
<https://github.com/w3c/mathml/issues/534>

*ACTION:* NS was hoping someone else would work on maction.

*ACTION:* NS asked people to look at his polyfills and look for errors.

*ACTION:* NS would like to have people add more examples to the polyfill
test pages

*ACTION:* PL will check DC's PR polyfill roll-up.
<https://cryptpad.fr/#cp-md-0-agenda>Agenda
<https://cryptpad.fr/#cp-md-0-1-announcements-updates-progress-reports>1.
Announcements/Updates/Progress reports
<https://cryptpad.fr/#cp-md-0-charter-status>Charter status

We had no new information on the charter status.
<https://cryptpad.fr/#cp-md-0-horizontal-review-status-for-cr>Horizontal
review status for CR

We had no new information on the Horizontal review status for CR.
<https://cryptpad.fr/#cp-md-1-2-progress-on-a-href-https-github-com-w3c-mathml-issues-534-534-summary-of-polyfill-needs-a->2.
Progress on #534: Summary of Polyfill needs
<https://github.com/w3c/mathml/issues/534>

NS has checked several items off on this issue.

NS used AI to help generate polyfills.

NS and MuS had a discussion on mathvariants and the decided to check it off
as well.

NS: mstyle is complicated because you cannot just push it onto an mro, but
it is doable.

NS said that he was not sure how to handle maction.

*ACTION:* NS was hoping someone else would work on maction.

NS: In general there are so many tests that he cannot check them all when
he makes a change.

ACTION: NS asked people to look at his polyfills and look for errors.

NS: Converted the elementary math polyfill from using an HTML table to
using a div with CSS grid and flex layout. He hoped that he could attach
that to an mrow and avoid having to articially add mtext/div, but it didn't
work. He spotted one style of long division that has a bug in the example.
It was there before the conversion to grid/flex box layout.

We discussed menclose and decided it could be checked.

*ACTION:* NS would like to have people add more examples.

DC did a roll-up.

*ACTION:* PL will check DC's PR roll-up.
<https://cryptpad.fr/#cp-md-0-3-mus-created-issue-a-href-https-github-com-w3c-mathml-issues-575-575-need-to-add-user-selection-attributes-a->3.
MuS created issue #575: Need to add user selection attributes
<https://github.com/w3c/mathml/issues/575>

Some assistive technology (AT) programs use MathML or OMML as conduits for
generating math speech and braille from math-enabled apps. In addition,
they would like to use these formats for editing math text as well as
speaking it and displaying it on refreshable braille displays. To this end,
the formats need selection attributes that identify where the user
selection is or should be within the MathML or OMML. These selection
attributes are intended for accessibility purposes; MathML/OMML in copy,
paste, and files ordinarily wouldn’t contain such attributes. The post MathML
and OMML User Selection Attributes
<https://devblogs.microsoft.com/math-in-office/mathml-and-omml-user-selection-attributes/>
describes this in some detail. The applet UnicodeMathML
<https://murrayiii.github.io/UnicodeMathML/playground/> implements the
selection attributes with slightly different names: "selanchor" and
"selfocus", which correspond to the selection object in the browser DOM. If
the selection is degenerate, i.e., an insertion point, only the selanchor
attribute is written.

If you have a nondegenerate solution, then you have selected something like
a word or paragraph and you need to know where its ends are.

MuS said that math accessibility will allow you to read what has already
been created, and will allow you to create new material.

Mus: When you write a new symbol, you need to know where your insertion
point goes. You also need to know which end is active so that when you
right shift, which point moves. Sometimes this is called the focus. The
other end is the anchor end.

NS: Should the selection be a global attribute?

NS: MuS could make a note out of this. It is a little late in the MathML
shipment schedule.
<https://cryptpad.fr/#cp-md-0-zoom-summary-for-the-intent-meeting-of-5-14-2026>Zoom
Summary For The Intent Meeting Of 5/14/2026
<https://cryptpad.fr/#cp-md-0-summary>Summary

The meeting focused on reviewing progress on MathML polyfills and
discussing a new proposal for selection attributes in MathML. Neil
presented updates on various polyfills, including M-table, M-enclosed, and
math variance implementations, noting that most were complete with some
remaining bugs to address. Murray then introduced a proposal for selection
attributes in MathML to support braille display editing, which sparked
discussion about implementation approaches and concerns about compatibility
with existing MathML specifications. The group debated whether selection
information should be encoded in MathML markup versus using IDs and
offsets, with David expressing concerns about the timing given the need for
horizontal review and potential CR approval.
<https://cryptpad.fr/#cp-md-0-polyfill-progress-and-implementation-updates>Polyfill
Progress and Implementation Updates

Neil led a discussion on the progress of polyfill needs, highlighting
completed tasks including the m table and mtr, and mtd attributes. The team
discussed the status of math variance, with Murray confirming that Cambria
Math had not been updated for Chancery/RoundHand and David suggesting that
the polyfill could simply include the variance letter without worrying
about font compatibility issues. The remaining items included mstyle
adjustments, which Neil noted were more complex due to dependencies on
mfrac, and maction, which Neil indicated would likely be handled by someone
else due to AI resource constraints.
<https://cryptpad.fr/#cp-md-0-grid-layout-implementation-challenges>Grid
Layout Implementation Challenges

Neil discussed issues with implementing a grid layout and HTML tables,
noting a bug where one example shifts an extra column compared to others
despite using identical encoding. He explained that he switched from using
HTML tables to divs, but encountered challenges with mtables and
transforms, particularly regarding circular elements. Neil also mentioned
improvements to testing and noted that while some features like arrows are
optional in the spec, the current implementation produces cleaner results
without garbage output.
<https://cryptpad.fr/#cp-md-0-mtable-polyfill-development-update>mtable
Polyfill Development Update

Neil discussed his work on creating a polyfill for mtable using AI
assistance, explaining that while the AI wrote the code, he had to guide it
at various points and he didn't learn much about grid and flexbox layouts
in the process. He demonstrated the polyfill's functionality, including
baseline alignment, column and row sizing, and label handling, while noting
some issues with color and frame display that needed further investigation.
Neil also highlighted the polyfill's ability to detect and handle
unsupported features, specifically mentioning its smart handling of labeled
rows in Firefox.
<https://cryptpad.fr/#cp-md-0-semantic-changes-and-polyfills-updates>Semantic
Changes and Polyfills Updates

Neil presented updates on semantic changes and polyfills, particularly
focusing on mtable implementations and MathML modifications. He requested
team members, especially David and others, to review the newer polyfills,
particularly the mtable and menclosed implementations, for any issues or
needed test cases. David discussed the process of creating a roll-up for
polyfills and clarified how to handle JavaScript compilation. Paul
confirmed he would review the pull request and make necessary comments. The
conversation ended with a brief mention of moving on to Murray's new issue
(575), though details about that were not provided in the transcript.
<https://cryptpad.fr/#cp-md-0-mathml-selection-attributes-proposal>MathML
Selection Attributes Proposal

Murray presented a proposal for implementing selection attributes in MathML
to support braille display editing, specifically focusing on anchor and
focus offsets for text selection. The discussion centered on whether these
attributes should be part of the MathML markup versus using IDs and
offsets, with Murray advocating for a single attribute approach while Neil
and David raised concerns about markup consistency and application-specific
implementation. The group acknowledged that while this needs to be
implemented soon for Microsoft's product, it may need to start as a
Microsoft-specific attribute before being standardized, with plans to
continue the discussion in next week's meeting.