Minutes: MathML Full Meeting, 14 November, 2024

 Attendees:

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

<https://sandbox.cryptpad.info/code/inner.html?ver=2024.9.0-6#cp-md-0-regrets>
Regrets

   - Moritz Schubotz

<https://sandbox.cryptpad.info/code/inner.html?ver=2024.9.0-6#cp-md-0-action-items>Action
Items

*ACTION* NS: Next week we will consider what needs to be done to get the
spec out.

*ACTION* NS: The "Draft Changes" item needs more work.

*ACTION* DC: #120: Remove/deprecate/simplify the ms element
<https://github.com/w3c/mathml/issues/120> needs more work.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.9.0-6#cp-md-0-498->
498:

XSD not allowing -element as child of -element
<https://github.com/w3c/mathml/issues/498> can we close?

*ACTION* DC No. This is a bug. I will work on it.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.9.0-6#cp-md-0--482->
#482:

Intent for large operators <https://github.com/w3c/mathml/issues/482>
(status?)

NS: So, the idea was to gather together all the large ops, so they didn't
need to be individually listed.

*ACTION* DC: will look into issue 482.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.9.0-6#cp-md-0--a-href-https-github-com-w3c-mathml-issues-388-388-scriptlevel-and-displaystyle-a->#388:
scriptlevel and displaystyle <https://github.com/w3c/mathml/issues/388>

NS: Can we make scriptlevel a global attribute?

Consensus: yes.

*ACTION* NS: Issue 388 calls for a spec change.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.9.0-6#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 and
<https://github.com/w3c/mathml/issues/181>

ACTION: MuS add to 181 an example of what Word is generating.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.9.0-6#cp-md-0--a-href-https-github-com-w3c-mathml-issues-180-180-decimalpoint-can-go-away-if-malignmark-goes-away-a->#180
(decimalpoint) can go away if malignmark goes away
<https://github.com/w3c/mathml/issues/180>

ACTION: NS: Deal with issue 180. Remove decimalpoint as appropriate.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.9.0-6#cp-md-0--103->
#103:

Extend mpadded attribute syntax to be closer to MathML3
<https://github.com/w3c/mathml/issues/103> (this is actually about making
Full closer to Core)

*CONSENSUS* DC: The meeting agreed to align full with core for issue [103] (
https://github.com/w3c/mathml/issues/103).
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.9.0-6#cp-md-0--493->
#493:

Clarify property use as descriptions
<https://github.com/w3c/mathml/issues/493> (about aria-describedby)

*ACTION* DG: Should add something to appendix c on aria label, and
everywhere else it makes sense to do so.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.9.0-6#cp-md-0-agenda>
Agenda
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.9.0-6#cp-md-0-1-announcements-updates-progress-reports>1.
Announcements/Updates/Progress reports

*ACTION* NS: Next week we will consider what needs to be done to get the
spec out.

   - Draft changes So there have been a few draft changes thanks to DG and
   PL.

*ACTION* NS: The "Draft Changes" item needs more work.

   - ms update #120: Remove/deprecate/simplify the ms element
   <https://github.com/w3c/mathml/issues/120>

DC writes: Just a note that the current draft description of is broken due
to the issue as the rendering uses and loses the quotes
https://w3c.github.io/mathml/#contm_cs_sec we could use instead but for now
just leaving a marker here so that doesn't get left pending resolution
here, It still seems that it should be possible to use the above CSS in the
user agent stylesheet and overcome accessibility issues somehow?

*ACTION* DC: #120: Remove/deprecate/simplify the ms element
<https://github.com/w3c/mathml/issues/120> needs more work.

DC: A lot of work is going on with MathML and PDF.

PL: I remember that copy and paste was not completely excluded as a target
of this MathML in PDF.

PL: If you don't have tagged PDF and you copy them out, you just get
garbage.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.9.0-6#cp-md-0-2-mathml-4-issues->2.
MathML 4 issues:
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.9.0-6#cp-md-0--498->
#498:

XSD not allowing -element as child of -element
<https://github.com/w3c/mathml/issues/498> can we close?

*ACTION* DC No. This is a bug. I will work on it.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.9.0-6#cp-md-0--497->
#497:

open concept list issue <https://github.com/w3c/mathml/issues/497> can we
close?

DC writes: The basic structure is now decided, closing here, individual
issues may be opened. Closed 497.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.9.0-6#cp-md-1--482->
#482:

Intent for large operators <https://github.com/w3c/mathml/issues/482>
(status?)

NS: So, the idea was to gather together all the large ops, so they didn't
need to be individually listed.

*ACTION* DC: will look into issue 482.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.9.0-6#cp-md-1--a-href-https-github-com-w3c-mathml-issues-388-388-scriptlevel-and-displaystyle-a->#388:
scriptlevel and displaystyle <https://github.com/w3c/mathml/issues/388>

NS: Can we make scriptlevel a global attribute?

NS: I think that core makes them global. I don't think there's anything
controversial about this.

*ACTION* NS: Issue 388 calls for a spec change.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.9.0-6#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 and
<https://github.com/w3c/mathml/issues/181>

NS to muS: Can you generate an example and put it into this issue?

NS: We can remove the decimal point from the spec.

ACTION: MuS add to issue 181 an example of what Word is generating.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.9.0-6#cp-md-1--a-href-https-github-com-w3c-mathml-issues-180-180-decimalpoint-can-go-away-if-malignmark-goes-away-a->#180
(decimalpoint) can go away if malignmark goes away
<https://github.com/w3c/mathml/issues/180>

NS: Drop decimal point from the spec. Someone needs to check this.

PL: The decimal point should not be removed because long division is not
fully supported.

NS: I have a polyfill that supports long division. It really heavily leans
on table alignments of various sorts and spacings, and so on, but it
doesn't make use of Malignmark, or anything like that, because they're
obviously not supported.

ACTION: NS: Deal with issue 180. Remove decimal point as appropriate.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.9.0-6#cp-md-1--103->
#103:

Extend mpadded attribute syntax to be closer to MathML3
<https://github.com/w3c/mathml/issues/103> (this is actually about making
Full closer to Core)

NS: It started out being a core issue. So, the issue is, it would be really
nice to be able to say width equals plus 3 m and have it incrementally add
3 m To the width, but you can't do that in core.

NS: So plus 3 m just means It's a positive width of 3 m in core.

NS: The full specification says 3 m adds to the width.

BM: It would be nice to have this functionality of being able to use +3 m
to add to the width. We do not want the core and full spec to disagree.

NS: There is a note that in CSS, you can do width equals calc, a hundred
percent plus 3 m.

NS: So, it's apparently possible. Of course this doesn't work in the other
browsers at the moment, but hopefully we get some more CSS alignment.

DC: I was suggesting that be actually allowed in the MathML attribute
directly in the MathML syntax for length, or just that you could override
it with Css externally.

Ns: We would be out of sync with core if we allowed this.

NS: So, it sounds like we can do the alignment and not lose anything,
pending implementations, catching up with the spec.

*CONSENSUS* DC:: The meeting agreed to align full of core for issue [103] (
https://github.com/w3c/mathml/issues/103).

NS: So, width equals that calc thing which you could copy from above
somewhere is supported. So, we really don't lose functionality.
<https://sandbox.cryptpad.info/code/inner.html?ver=2024.9.0-6#cp-md-1--493->
#493:

Clarify property use as descriptions
<https://github.com/w3c/mathml/issues/493> (about aria-describedby)

NS: At some point intent could give more information about a variable
through a property, and this could be shown to the reader with a popup.

NS: We have moved away from doing this with a property and we have moved
towards aria-describedby.

NS: This is problematic with MathCAT which only gets MathML, and does not
get information from something outside of MathML.

DG: There was a method of using aria description in MathML with a hack so
that they trick screen readers. I think it was Chromevox. And there is a
keyboard shortcut that you can make it speak the description optionally.

DG: Aria description is in spec 1.3 which is not fully implemented. If you
hit a shortcut, you get the extra information. There should be a way to
introduce it accessibly.

MuS wants to round trip differential d. Cannot presently tell Chromium how
to display a differential d, say, as math italic d. Need some way to do
this, E.g., intent: 'differential d'

NS: A renderer or an intelligent AT could take account of this intent. It
did start as a property-based matter.

DG: One way of structuring it is the fallback behavior for properties which
is not specified at the moment. We have a fallback behavior for concepts,
and by that I mean, an unknown concept has a default recommendation to be
read as a literal. For properties, we could make a very similar
recommendation, but instead of reading it, we could use words such as make
it available as a secondary piece of information.

NS: Is in favor of aria description.

DC: The option is probably to use aria description because that's defined
by somebody else.

DC: We could have a core property that said description.

NS: You cannot give it a value.

NS: Is in favor of aria description.

DC: So, we say intent is similar to aria label, and we say, don't use aria
label. So, if we wanted to use whatever it's called this the one that is
the URL description it would mean adding it to core.

NS: We should let the aria group know if we use aria description to see if
they agree with us.

BM: Is there any hope that the things that break aria are being fixed?

NS: So, they are probably being fixed in 1.3. I have no idea where it is
right now. They have introduced aria Braille as something to go alongside a
label so that you can give the braille for something.

NS: Aria label does not make much sense in MathML.

NS: The label is meant to be automatically spoken when you reach it. This
is problematic for math because it overrides everything inside it.

NS: We could make these changes in appendix /c on accessibility.

DG would accept aria description.

*ACTION* DG: Should add something to appendix c on aria label, and
everywhere else it makes sense to do so.

NS: Next week we will see what needs to be done to get the spec out.