Minutes: MathML full meeting 27 Feb, 2025

 Attendees:

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

<https://sandbox.cryptpad.info/code/inner.html?ver=2024.12.0-3#cp-md-0-regrets>
Regrets
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.12.0-3#cp-md-0-action-items>Action
Items
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.12.0-3#cp-md-0-1-announcements-updates-progress-reports>1.
Announcements/Updates/Progress reports

*CONSENSUS* We unanimously agree to send the core spec to CR.

From Paul Libbrecht to everyone: I’ll be starting a
breakout-session-proposal: Read-Aloud-Math
https://cryptpad.fr/pad/#/2/pad/edit/xsuFt0+CdvmfbBXk55eeciaA/
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.12.0-3#cp-md-0-2-a-href-https-github-com-w3c-mathml-issues-181-181-mathml-4-extensions-for-alignment-and-possible-deprecation-of-maligngroup-and-malignmark-a->2.
#181: MathML 4 extensions for alignment and possible deprecation of
(maligngroup/) and (malignmark/) <https://github.com/w3c/mathml/issues/181>

(updates?)

*ACTION* DC will look at it again. He wants to come up with a new proposal.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.12.0-3#cp-md-0-3-a-href-https-github-com-w3c-mathml-issues-385-385-comments-on-chapter-3-a->3.
#385: comments on chapter 3 <https://github.com/w3c/mathml/issues/385>

(updates/new thoughts on how to handle examples?)

No updates.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.12.0-3#cp-md-0-4-a-href-https-github-com-w3c-mathml-core-issues-279-279-core-clarify-the-language-around-deprecated-features-of-mathml-a->4
#279 (core) Clarify the language around deprecated features of MathML
<https://github.com/w3c/mathml-core/issues/279>

(updates)

*action* From the February 24, 2025, core meeting, BK will talk to MDN
about how to name things that are in full, but not in core.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.12.0-3#cp-md-0-agenda>
Agenda
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.12.0-3#cp-md-1-1-announcements-updates-progress-reports>1.
Announcements/Updates/Progress reports

LM suggested that we pass a motion to take the core spec to CR.

BM: Should we pass this before all the core issues have been discussed?

NS: I think most of the issues that we're talking about in CORE are
rendering issues. So there are details about the algorithm, which I doubt
many groups are going to be concerned about.

*CONSENSUS* We unanimously agree to send the core spec to CR.

From Paul Libbrecht to everyone: I’ll be starting a
breakout-session-proposal: Read-Aloud-Math
https://cryptpad.fr/pad/#/2/pad/edit/xsuFt0+CdvmfbBXk55eeciaA/
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.12.0-3#cp-md-1-2-a-href-https-github-com-w3c-mathml-issues-181-181-mathml-4-extensions-for-alignment-and-possible-deprecation-of-maligngroup-and-malignmark-a->2.
#181: MathML 4 extensions for alignment and possible deprecation of
(maligngroup/) and (malignmark/) <https://github.com/w3c/mathml/issues/181>

(updates?)

NS: I think we said It's deprecated. What deprecated means is that a
renderer should not generate it, but of course legacy renderers will not
change, right?

NS: It is not deprecated, it is just not supported in core.

NS: The proposal here is to say it's deprecated.

MuS: The workarounds do not work with MS word. He has a method to make it
work with word.

MuS: I check it out with word, And where it doesn't work with the proposed
workarounds, it does work with the original line mark and aligned groups.
So I have a special Control C, "control shift c", which generates MathML
that Word understands.

DC: One option is to make what you're doing valid and then if you want to
know how it's supposed to work, you look at the MathML3 spec, but we say
it's still valid.

NS: There's still a lot on this that nobody supports except for Word.

From Deyan Ginev to everyone: Curious: and are not documented in MDN at all

NS: You can say it's deprecated. That doesn't change its validity at all in
MathML3, right? This could be an informative note.

*ACTION* DC will look at it again. He wants to come up with a new proposal.

NS: I don't object at all to having a section that says this is This has
been deprecated. The only implementation that uses it is office. And in
that case, it works like this, and you write out a paragraph describing how
it works.

MuS: You know, it'd just be sort of an informative note kind of a thing.

From Deyan Ginev to everyone: Mathpix appears to generate *some* malignmark
and maligngroup elements when OCR-ing two large matrices multiplying.

DG: The mathematical word Mathpix produces contains malign mark and malign
group.

NS: If we have another generator that produces maligngroup and malignmark,
then we cannot get rid of maligngroup and malignmark.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.12.0-3#cp-md-1-3-a-href-https-github-com-w3c-mathml-issues-385-385-comments-on-chapter-3-a->3.
#385: comments on chapter 3 <https://github.com/w3c/mathml/issues/385>

(updates/new thoughts on how to handle examples?)

No updates.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.12.0-3#cp-md-1-4-a-href-https-github-com-w3c-mathml-core-issues-279-279-core-clarify-the-language-around-deprecated-features-of-mathml-a->4
#279 (core) Clarify the language around deprecated features of MathML
<https://github.com/w3c/mathml-core/issues/279>

(updates)

*action* From the February 24, 2025, core meeting, BK will talk to MDN
about how to name things that are in full, but not in core.

NS: You cannot deprecate something that was never there.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.12.0-3#cp-md-0-5-a-href-https-github-com-w3c-mathml-issues-525-525-equation-numbers-a->5
#525: Equation Numbers <https://github.com/w3c/mathml/issues/525>

(update)

DG: I learned CSS Grid last week. I have no idea how well I used it, but it
seems to have worked sufficiently.

DG: One issue I have with the previous example was absolute position.
Absolute position is notoriously limiting; that, once you use it, you
cannot remix whatever you did in other components; or ,as a component,
relative positioning allows you to make things a bit more consistent. I'll
put it in a larger frame like a slide or do something else with it. It's
helpful if the question number is relative to the parents, the container
parent of the equation.

DC: I gave up with absolute positioning because some people said that it
messed up at the baseline. Relative positioning works well with the
baseline.

NS: Can we find a solution that works? If so, is it acceptable? There was
some talk about potentially using an attribute on an mtr.

NS This seems like it's another solution. And how we tag the mtd has to be
decided upon.

DC suggest having a property equation label or no equation label.

DG: But the accessibility questions are a separate and important aspect to
the issue. In fact, I think the main aspect of the issue as originally
posted is how do we get them spoken well? How do we get them navigated to
and navigated between?

NS: It does speak well with MathCAT. I could also search for them.

NS: If you want to have an equation be findable, you put an Mtext ID on it.

NS: MathCAT needs the equation label to be in the first column.

DG: Let us wait and see what the next setup is. We do not want to change
all the label generators to put labels only in the first column.

DC: We cannot tell people that their documents cannot have equation labels.

DG: Do what you are doing now as long as it is open, but not core.

DC: It should be in core.

MoS: People asked for support for these labels and also for the references
which is another topic So because when you use the label you also want to
be able to reference that And so until now we always said yes you can use
some HTML to do that and it's not specific to MathML.

MoS does not know why labels have to be in MathML.

DC: These equation numbers have to be aligned with this MathML table So
they're part of the MathML.

DG: Let me address this. So this is also a subversive element of CSS grid.
Once you can mod an mtable to be a grid, you can clearly mod any HTML to
fit this scheme. And if you lift it out of the mtable, you can still make
it work.

DG: You have two contradictory goals. If you want it done now, it cannot be
the same goal as wanting it to be a standard, because what you're going to
standardize will hurt MathML and LaTeX downstream.

DG: Either we should do a standard, take the time, and include people in
the process that everything is working well, or you just run with the
experiment.

DC wants to make equation label a core property.

BM: We should not restrict the label to being in a specific column or row.

NS: What DG or MoS are saying is that I want to generate the label on the
outside, and I just want to have some separate mechanism that says, this is
a label, and here's what it's attached to.

NS: There are other ways to get labels other than MathML.

NS: The other way to think about it is there could be multiple ways that we
do labels. But if you want to put it in the MathML, here is how you do it.

DC: The screen reader doesn't read the label id, it reads the label.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.12.0-3#cp-md-0-6-a-href-https-github-com-w3c-mathml-issues-527-527-intent-or-anything-else-for-forced-pause-a->6
#527: intent (or anything else) for forced pause
<https://github.com/w3c/mathml/issues/527>

DC: The issue essentially is that if you put the equation label in, most of
the alignments are read quite nicely, but this column separation just isn't
correct.

DC It's not very semantically rich to just say pause here, but it doesn't
make it more understandable when you're listening to it.

DC: the proposal was to have an intent:pause parameter.

NS: So I could imagine adding 3 properties: Pauselong, pausemedium, and
pauseshort.

From Deyan Ginev to everyone: To me

_(pause, pause, pause)

is just another way to say

_(_,_,_)

From Deyan Ginev to everyone: "mspace" would be a concept name that is
easier to accept

NS: People cannot send punctuation to the system like commas and semicolons
because the screen reader would read the punctuation if the user set that
option. The punctuation would not cause pauses.

If the author is not in the loop, the AT knows it has a big amount of white
space.