Minutes: MathML Full Meeting, 4 Dec, 2025

 Attendees:

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

<https://cryptpad.fr/#cp-md-0-regrets>Regrets
<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
<https://cryptpad.fr/#cp-md-0-spec-updates-some-parts-overlaps-with-next-two-issues->Spec
updates (some parts overlaps with next two issues)

*CONSENSUS, ACTION:* NS proposes that we add <a> to MathML full on the
presumption that it is actually implemented in MathML core. DC will make
PR's for MathML core and MathML full to do this.
<https://cryptpad.fr/#cp-md-0-2-small-things-from-a-href-https-github-com-w3c-mathml-issues-385-385-comments-on-chapter-3-a-that-require-editorial-spec-updates>2.
Small things from #385 Comments on chapter 3
<https://github.com/w3c/mathml/issues/385> that require editorial spec
updates
<https://cryptpad.fr/#cp-md-0-wide-examples-needing-scrollbar-a-href-https-w3c-github-io-mathml-open-presm_mo_linebreaks-https-w3c-github-io-mathml-open-presm_mo_linebreaks-a->Wide
examples needing scrollbar
https://w3c.github.io/mathml/?open#presm_mo_linebreaks

(affects several sample presentations in chapter4 as well)

NS: You need a scroll bar because certain lines are too long to be seen on
one frame.

*ACTION:* In the spec, DC will break up any lengthy line, which cannot be
seen in a single frame, into a few short lines which can be seen on a
single frame.
<https://cryptpad.fr/#cp-md-0-3-a-href-https-github-com-w3c-mathml-issues-1-issuecomment-3518606914-1-simplification-of-the-mstyle-element-rename-decimalpoint-attr-a->3.
#1 Simplification of the mstyle element (rename decimalpoint attr?)
<https://github.com/w3c/mathml/issues/1#issuecomment-3518606914>

*ACTION:* NS: The name "decimalpoint" is the name of an attribute
"decimalpoint", and the name of its value. NS suggests leaving the name of
the attribute as "decimalpoint" and naming its value as "decimalcharacter".
NS will make these changes in the spec.
<https://cryptpad.fr/#cp-md-0-5-a-href-https-github-com-w3c-mathml-issues-q-is-3aissue-20state-3aopen-20sort-3aupdated-desc-20label-3a-22need-20specification-20update-22-other-remaining-quot-needs-spec-updates-quot-issues-a->5.
Other remaining "Needs spec updates" issues
<https://github.com/w3c/mathml/issues?q=is%3Aissue%20state%3Aopen%20sort%3Aupdated-desc%20label%3A%22need%20specification%20update%22>
.

NS would like to have the spec ready to be considered for CR by next
Thursday.

NS, DC, and DG have items that need attention on this list.
<https://cryptpad.fr/#cp-md-0-agenda>Agenda
<https://cryptpad.fr/#cp-md-1-1-announcements-updates-progress-reports>1.
Announcements/Updates/Progress reports
<https://cryptpad.fr/#cp-md-0--a-href-https-mathfonts-github-io-mathfonts-site-a->MathFonts
site <https://mathfonts.github.io/>

DC: This site helps people develop fonts.

From Moritz Schubotz to everyone: Do you have control which LaTeX macros
are used?

From Moritz Schubotz to everyone: ok thank you. it turned out that some
symbols heavily depend on the latex macro, so it’s good that you can see
the MathML

NS: This can be used to test fonts.

DC: When you see things side-by-side, things show up.

DG: Do these fonts have people you can contact?

DC: David shared that he maintains records of the fonts' sources and
licenses, though he does not directly expose this information on the web
view due to licensing restrictions.

DC: You can see that some things do not stretch.

From Moritz Schubotz to everyone: \Stigma looks completely different
<https://cryptpad.fr/#cp-md-0-update-from-moritz-on-open-list-prompted-by-some-discussion-on-adding-bra-ket-notation->Update
from Moritz on Open List (prompted by some discussion on adding bra ket
notation)

MoS: There has not been much community involvement up to now.

MoS: We have built an initial user interface where people could contribute.

NS: I've suggested that they come up with some names for the BraKet stuff,
and I think it's still unresolved.
<https://cryptpad.fr/#cp-md-1-spec-updates-some-parts-overlaps-with-next-two-issues->Spec
updates (some parts overlaps with next two issues)

NS: There has been some improvements to the spec.

DC: There has been work simplifying the mstyle element.

DC: The things that were on mstyle are now allowed on presentation elements.

DC: There has been work done on the mtext element.

DC: We should decide what to do about the A element.

DC: If it actually gets links implemented in browsers, it's very popular.

NS: Because we want to be a superset of core, we do need to add an A
element.

*CONSENSUS, ACTION:* NS proposes that we add A to MathML fully on the
presumption that it is actually implemented in MathML core. DC will make
PR's for MathML core and MathML full to do this.

MuS: I'm just saying we shouldn't recommend that we switch all your mrows
to A's.

DC: The specs will not suggest that.

MuS: Many things are happening at Microsoft in that desktop applications
are increasingly able to handle MathML.

MuS: Currently the desktop applications cannot handle intents, but MuS is
pushing for that.

NS: Narrator is starting to use MathCAT.

DC: LaTeX can generate MathML.

PL: Teachers want students to use JAWS, and to use cut and paste to
generate formulas.

NS: When I worked for Design Science, It was very clear that the world was
split between people who wanted to use LaTeX, and people who wanted to use
Word, and you were divided into communities because everybody else in your
community used one or the other.

NS: In the math world, most communities use LaTeX. However some, like
Operations research use Word (at least that's my memory).

DC: You can now read tagged pdf files in Firefox using MathCAT. Still buggy
though.

NS: This works both in NVDA and JAWS.
<https://cryptpad.fr/#cp-md-1-2-small-things-from-a-href-https-github-com-w3c-mathml-issues-385-385-comments-on-chapter-3-a-that-require-editorial-spec-updates>2.
Small things from #385 Comments on chapter 3
<https://github.com/w3c/mathml/issues/385> that require editorial spec
updates

NS: Is this going to hold up CR?
<https://cryptpad.fr/#cp-md-1-wide-examples-needing-scrollbar-a-href-https-w3c-github-io-mathml-open-presm_mo_linebreaks-https-w3c-github-io-mathml-open-presm_mo_linebreaks-a->Wide
examples needing scrollbar
https://w3c.github.io/mathml/?open#presm_mo_linebreaks

(affects several sample presentations in chapter4 as well)

NS: You need a scroll bar because certain lines are too long to be seen on
one frame.

*ACTION:* In the spec, DC will break up any lengthy line, which cannot be
seen in a single frame, into a few short lines which can be seen on a
single frame.
<https://cryptpad.fr/#cp-md-0-remaining-numeric-character-ref-lt-mmultiscripts-gt-lt-mo-gt-amp-x0644-lt-mo-gt-must-have-missed-my-global-update-for-some-reason->Remaining
Numeric Character ref <mmultiscripts><mo>&#x0644;</mo> must have missed my
global update for some reason.

DC: There is no problem with it. It does not need fixing.
<https://cryptpad.fr/#cp-md-1-3-a-href-https-github-com-w3c-mathml-issues-1-issuecomment-3518606914-1-simplification-of-the-mstyle-element-rename-decimalpoint-attr-a->3.
#1 Simplification of the mstyle element (rename decimalpoint attr?)
<https://github.com/w3c/mathml/issues/1#issuecomment-3518606914>

*ACTION:* NS: The name "decimalpoint" is the name of an attribute
"decimalpoint", and the name of its value. NS suggests leaving the name of
the attribute as "decimalpoint" and naming its value as "decimalcharacter".
NS will make these changes in the spec.
<https://cryptpad.fr/#cp-md-0-4-a-href-https-github-com-w3c-mathml-issues-247-247-spec-should-specify-what-char-to-use-for-accents-lines-a->4.
#247: Spec should specify what char to use for accents/lines
<https://github.com/w3c/mathml/issues/247>.

NS: What character should be used for the accent mark?

NS: There is no way to do this. a character may not work across all
browsers.

BM: We should avoid using combining characters.

MuS: The reason that the Microsoft software went with the combining marks
is because we're using Unicode Math, and this problem just doesn't emerge
with Unicode Math, it just works.

DG: This problem looks extremely polyfillable. We just write the correct
characters if we know them. The polyfill will be a living document on what
works.
<https://cryptpad.fr/#cp-md-1-5-a-href-https-github-com-w3c-mathml-issues-q-is-3aissue-20state-3aopen-20sort-3aupdated-desc-20label-3a-22need-20specification-20update-22-other-remaining-quot-needs-spec-updates-quot-issues-a->5.
Other remaining "Needs spec updates" issues
<https://github.com/w3c/mathml/issues?q=is%3Aissue%20state%3Aopen%20sort%3Aupdated-desc%20label%3A%22need%20specification%20update%22>
.

NS: We hopefully have these issues in hand.

NS, DC, and DG have items that need attention on this list.

NS would like to have the spec ready to be considered for CR by next
Thursday. DC has many work-related meetings, and many issues that need
work. He said he would not be able to complete all of his work by next
Thursday.

MuS was working with a tilde, and it stretched. It seems to be a Chromium
bug that it isn't stretching.
<https://cryptpad.fr/#cp-md-0-zoom-meeting-summary-12-04-2025>Zoom Meeting
Summary 12/04/2025 <https://cryptpad.fr/#cp-md-0-summary>Summary

The team addressed technical issues with agenda links and discussed a new
font display site that allows users to test and compare mathematical
expression rendering across different fonts. They reviewed ongoing projects
including the Bra Ket notation initiative and discussed changes to MathML
specifications, particularly focusing on the addition of the A element. The
conversation ended with discussions about accessibility solutions for
mathematics, including LaTeX and MathML support, and addressed various
technical issues related to character rendering and specifications. David
mentioned he would miss the next meeting due to an in-person company
gathering, and the team briefly discussed technical problems with PDF files
that Neil and others were experiencing, which they were hoping to resolve
by getting a new machine.
<https://cryptpad.fr/#cp-md-0-math-font-display-site-overview>Math Font
Display Site Overview

David presented a font display site that showcases over 30 OpenType fonts
with math support, allowing users to test and compare font rendering for
mathematical expressions. The site, which can be accessed at (
https://mathfonts.github.io/), includes features such as drag-and-drop
functionality, the ability to edit expressions in ASCII Math, MathML, or
LaTeX, and the option to disable fonts for comparison. David noted that the
site has already led to bug fixes in several fonts, including Asana Math,
Nagara GK, and Pensander. Paul inquired about the site's cookie handling,
to which David explained that cookies are only saved on demand when users
choose to save the state. Deyan asked about contacting font developers, and
David shared that he maintains records of the fonts' sources and licenses,
though he does not directly expose this information on the web view due to
licensing restrictions.
<https://cryptpad.fr/#cp-md-0-font-licensing-and-project-updates>Font
Licensing and Project Updates

David discussed his efforts to document the licensing of fonts used in the
project, including creating links to their sources and ensuring compliance
with open licenses. Moritz provided an update on the Bra Ket notation
project, noting limited community involvement and the development of an
initial user interface for contributing, but highlighted the need for
further collaboration to advance the project. Neil acknowledged the
challenges in attracting contributors and suggested that interest might
increase once users encounter specific problems. The conversation ended
with a brief mention of improvements to specifications and pull requests,
though no further details were provided.
<https://cryptpad.fr/#cp-md-0-mathml-a-element-specification-update>MathML
A Element Specification Update

The team discussed changes to MathML specifications, focusing on the
addition of the A element to both MathML full and core specifications.
David proposed creating an A element identical to MROW but allowing href,
which was agreed upon by the group. Moritz expressed concerns about
potential compatibility issues, but David reassured him that A elements
would behave similarly to MROWs. The team also briefly touched on recent
merges and fixes to the specifications. Murray mentioned ongoing
developments in Microsoft desktop apps, including improvements in math
handling and potential interest in intents.
<https://cryptpad.fr/#cp-md-0-math-accessibility-latex-and-mathml>Math
Accessibility: LaTeX and MathML

The group discussed accessibility solutions for mathematics, particularly
focusing on LaTeX and MathML support. Paul shared insights from a workshop
on visual impairment and mathematics, where LaTeX was identified as the
preferred solution for blind users due to its compatibility with JAWS and
copy-paste functionality in Word. David mentioned that Firefox's PDF.js now
supports MathML, allowing accessible reading of mathematical documents
across different screen readers. The discussion concluded with a brief
mention of formatting issues in examples that need to be addressed before
moving forward.
<https://cryptpad.fr/#cp-md-0-fixing-wide-examples-issue>Fixing Wide
Examples Issue

David agreed to fix the issue with wide examples causing horizontal scroll
bars, despite the team's consensus that it wasn't a critical problem. He
explained that the width issues were caused by invisible operators having
comments that made them non-zero width, and he would search for occurrences
of the hex number 2061 to make the necessary fixes.
<https://cryptpad.fr/#cp-md-0-mathematical-characters-specification-updates>Mathematical
Characters Specification Updates

The team discussed several issues related to the specification and
implementation of mathematical characters and attributes. They agreed to
change the name of the "decimalpoint" attribute value to "decimal
character" and to deprecate the old name, with Neil taking responsibility
for making this change in the spec. The group also addressed concerns about
the reliability of combining characters across different browsers, with
David suggesting a polyfill approach to handle character-specific issues.
Neil set a goal to have most remaining PRs completed by the next meeting,
allowing for a potential CR vote at the final meeting.