Minutes_ MathML Core Meeting_ 27 April 2026

04/27/2026 MathML Core Meeting-core#322<https://github.com/w3c/mathml-core/issues/322>
Attendees

  *   Brian Kardell
  *   Louis Maher
  *   Eri Pazos Perez
  *   Deyan Ginev
  *   Neil Soiffer
  *   Bruce Miller
  *   David Carlisle
Regrets
Action Items
3. Updates on implementation progress - Look into making combining accents work in Chrome: core issue 311<https://github.com/w3c/mathml-core/issues/311>
EP: Some news on #311: I made a rough test and a working prototype for Chromium, I can show it in the meeting, so at least we know it's possible. Next steps would be actually creating the equivalence table between different accents and decide which ones are the "preferred" ones, and check if a mapping like this is desirable.
ACTION: NS an DC will go through the accent marks to see which should be the target of the mapping.
4. Discussion of MathML Core issue 320: Browser Status?<https://github.com/w3c/mathml-core/issues/320>
and related tests

  *   #203: Adjusted height & depth of elements moved via voffset<https://github.com/w3c/mathml-core/issues/203>
ACTION: EP will check to see if item 203<https://github.com/w3c/mathml-core/issues/203> can be closed because height and depth are supported in all the browsers, ideally provide a wpt link showing we have tests and it can be closed. This is linked to: Ink ascent/descent: core issue 78<https://github.com/w3c/mathml-core/issues/78>.

  *   #271 Add calligraphic mathvariant support #271<https://github.com/w3c/mathml-core/issues/271>
ACTION: DC will check variant selectors in browsers with current fonts and report back on core issue #271<https://github.com/w3c/mathml-core/issues/271>.
ACTION: NS will ask Ron Kok if he would move his Browser Rendering Status Report<https://temml.org/docs/en/mathml-status>, discussed in core issue 320<https://github.com/w3c/mathml-core/issues/320> into W3C space so that we can work on it.
5 Follow-up on "new issues" from last time
User Agent style for href hyperlinks and focusable elements #128<https://github.com/w3c/mathml-core/issues/128>
ACTION: BK to check the status of links in MathML and their styling. Make sure we have tests to see if we need edits to the sspec UA style sheets.
Add a template for reporting issue #318<https://github.com/w3c/mathml-core/issues/318>
ACTION: DC will make a template.
Document why mtable sets displaystely to false #314<https://github.com/w3c/mathml-core/issues/314>
ACTION: DC will make a PR to the spec to explain why mtable sets displaystely to false.
Agenda Items
1. Announcements/Updates
NS: MathCAT is being added to Microsoft Narrator.
BK has talked to Microsoft about joining our group.
NS: Microsoft's pdf support includes MathML support.
NS: pdf does not work with Firefox.
2. Charter<https://w3c.github.io/mathml-docs/charter-2026.html> discussion
NS: The charter has been reviewed by two of the five horizontal groups. Thus far, there have not been any issues.
3. Updates on implementation progress - Look into making combining accents work in Chrome: core issue 311<https://github.com/w3c/mathml-core/issues/311>
EP: Some news on #311: I made a rough test and a working prototype for Chromium, I can show it in the meeting, so at least we know it is possible. Next step would be actually creating the equivalence table between different accents and deciding which ones are the "preferred ones and check if a mapping like this is desirable.
EP was unsure if Chromium would work, but there is a working patch now. Will post on the issue after internal review.
EP implemented a table of combining characters that could work as accent marks.
NS and DC will work on the table of characters that map to an accent mark.
ACTION: NS an DC will go through the accent marks to see which should be the target of the mapping.
4. Discussion of MathML Core issue 320: Browser Status?<https://github.com/w3c/mathml-core/issues/320>
and related tests

  *   #135: Add detailed rules for the ssty OpenType feature<https://github.com/w3c/mathml-core/issues/135>
  *   #203: Adjusted height & depth of elements moved via voffset<https://github.com/w3c/mathml-core/issues/203>
This is linked to: Ink ascent/descent: core issue 78<https://github.com/w3c/mathml-core/issues/78>.
ACTION: EP will check to see if item 203<https://github.com/w3c/mathml-core/issues/203> can be closed because height and depth are supported in all the browsers, ideally provide a wpt link showing we have tests and it can be closed.
Ron Kok wrote: Yes, height and depth are currently supported. That is insufficient to resolve this issue. The need is for a way to increment height or depth. Remember, the MathML 2.0 increment syntax is not supported by MathML Core. An expression such as depth=+0.5em does not increment the height by 0.5em.

  *   #271 Add calligraphic mathvariant support #271<https://github.com/w3c/mathml-core/issues/271>
DC: I need to check which fonts have variant selector support.
ACTION: DC will check variant selectors in browsers with current fonts and report back on core issue #271<https://github.com/w3c/mathml-core/issues/271>.
ACTION: NS will ask Ron Kok if he would move his Browser Rendering Status Report<https://temml.org/docs/en/mathml-status>, discussed in core issue 320<https://github.com/w3c/mathml-core/issues/320> into W3C space so that we can work on it.
5 Follow-up on "new issues" from last time
User Agent style for href hyperlinks and focusable elements #128<https://github.com/w3c/mathml-core/issues/128>
BK: We need to decide if we are putting default styles in, and what They look like.
BK: We should recheck what they are doing now, and whether they are in a namespace, style sheet, or whether they are just globally applied.
ACTION: BK to check the status of links in MathML and their styling. Make sure we have tests to see if we need edits to the sspec UA style sheets.
Add a template for reporting issue #318<https://github.com/w3c/mathml-core/issues/318>
BK: We need to make an issue template. We tell DC what should be in the problem report and he will make a template.
BK: So, there is a thing in GitHub or GitLab where you can put an issue template.
BK likes a template that asks a couple of questions.
NS said that browser and font information should be included in the template. A test case should be part of the problem report.
ACTION: DC will make a template.
Problems with "Root with index" #317<https://github.com/w3c/mathml-core/issues/317>
Related to core issue 78 Ink ascent/descent<https://github.com/w3c/mathml-core/issues/78>
DC: If you have not decided how big you will make this radical, you cannot really decide where to put the index.
Unclear how to use Operators_fence special table #315<https://github.com/w3c/mathml-core/issues/315>
Document why mtable sets displaystely to false #314<https://github.com/w3c/mathml-core/issues/314>
ACTION: DC will make a PR to the spec to explain why mtable sets displaystely to false.
Zoom Core Meeting Summary 4/27/2026
Summary
The MathML Working Group held a meeting to discuss various implementation issues and updates. The group reviewed progress on Issue 311, where Eri presented a solution for unifying accent characters across different browsers by creating a mapping table. They discussed the status of MathML Core 320 browser support and related tests, including issues with variant selectors and radical signs. The team also addressed the need to update user agent style sheets for links in MathML and considered creating an issue template for better bug reports. David agreed to check variant selector support in current fonts and to document the style table settings for mtable in the specification.
MathML Collaboration Updates Meeting
The meeting began with participants joining and brief discussions about agenda items and technical issues. Neil shared updates on Microsoft's work with MathML, including improvements to PDF support and ongoing security reviews. The group discussed potential collaboration with Microsoft to promote MathML accessibility, though there were some concerns about legal considerations due to Microsoft's membership status. The conversation ended with plans to discuss implementation updates, specifically regarding Issue 311, though the detailed discussion was not captured in the transcript.
Chromium Accent Characters Unification
The team discussed Eri's work on unifying accent characters in Chromium, which is now possible through a mapping table approach. Neil and David agreed to review and decide on the target characters for the mapping, particularly focusing on fonts like Cambria Math, STIX, and Latin Modern that are difficult to change. Eri will review the code with Fred and post a draft version of the patch for pre-review to ensure there are no concerns before further development.
MathML Variant Selectors Browser Support
The team discussed browser support for MathML variant selectors, specifically focusing on issue 271. David agreed to check which fonts currently support variant selectors and test their compatibility with browsers. The discussion revealed that while most fonts don't support variant selectors, some newer versions might, and David will investigate this further by adding tests to the math fonts page.
MathML Issue Resolution Discussion
The team discussed several MathML issues, focusing on closing resolved items and addressing ongoing problems. Eri agreed to check if issue 203 about height and depth support could be closed, as CSS support has been implemented across browsers. Neil will ask Ron Kok if he would move his Browser Rendering Status Report into W3C space so that we can work on it. The conversation ended with a brief mention of issue 128 regarding user agent style for hyperlinks, though the discussion was cut off in the transcript.
MathML Links Styling Updates
The team discussed updates needed for MathML links styling, particularly focusing on WebKit implementation differences and the need to verify current status across browsers. Brian identified the need for specification updates and tests, while Neil questioned whether default styles were required since links should work similarly to standard A elements. The team agreed that Brian would take action to check the status of links in MathML, verify focus visible tests, and assess whether edits to the spec or user agent style sheets are needed.
Issue Template Development Discussion
The team discussed creating an issue template for problem reports, with David agreeing to develop a lightweight template that includes essential information like browser details, font information, and test cases. They noted that most bug reports involve contributed packages outside their control, so they typically direct users to contact the original package maintainers. The team also briefly touched on action items from issue 317 regarding root width index problems, which were added in the previous meeting and need team review and comments.
GitHub Issue Linking and Updates
The team discussed linking related GitHub issues and decided to link issues 78 and 314. David agreed to create a pull request to clarify documentation about mtable display style.