Minutes: MathML Full meeting 5 Oct, 2023

 Attendees:

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

<https://sandbox.cryptpad.info/code/inner.html?ver=5.5.0-b#cp-md-0-regrets>
Regrets
<https://sandbox.cryptpad.info/code/inner.html?ver=5.5.0-b#cp-md-0-agenda>
Agenda
<https://sandbox.cryptpad.info/code/inner.html?ver=5.5.0-b#cp-md-0-1-announcements-updates-progress-reports>1.
Announcements/Updates/Progress reports

NS asked BB for a charter update.

BB: said that he looked at the charter status, and everything looks very
nice. There was only positive reviews, no request for changes.

BB: There were 23 good reviews.

BB: There is an internal task force which analyzes the reviews. It will
take a few days for them to do the work.

BB felt that they would accept the charter.

BB must make sure that there is a process in place for people to join the
working group. This might take an additional week.
<https://sandbox.cryptpad.info/code/inner.html?ver=5.5.0-b#cp-md-0-dg-renamed-closed-items-as-per-action-item>DG
renamed/closed items as per action item

DG: said that he had no feedback, and this was good.
<https://sandbox.cryptpad.info/code/inner.html?ver=5.5.0-b#cp-md-0-wikipedia-announcement>Wikipedia
announcement

MoS: Wikipedia is using his algorithm on a test branch. He would like us to
look at the results before the algorithm is used on Wikipedia's main branch.

MoS algorithm generates MathML. It will allow authors to add intent.

DC said that BK had a PR for the HTML spec which now points to MathML 4
instead of MathML 3.

DC: will make the MathML 4 spec point to a static version instead of the
ReSpec version so that the spec will load faster.
<https://sandbox.cryptpad.info/code/inner.html?ver=5.5.0-b#cp-md-0-2-interop-submission-suggestions->2.
Interop submission suggestions:
<https://sandbox.cryptpad.info/code/inner.html?ver=5.5.0-b#cp-md-0-ns-action-item-polyfill-problems>NS
action item: polyfill problems

NS: looked at his polyfills and compared their results using the chrome,
Firefox, and Safari Browsers.

NS's first example: It was trying to do left right and center underlining
as if there's a stretchable character here.

NS: I thought that Chrome did not do it right; but I went and looked at the
description of when to stretch things. The spec wasn't clear, so I think we
have a spec problem.

NS: Chrome is stretching the arrow in an mrow with an mspace while the
other systems are not.

NS: So, the spec algorithm doesn't say when to do stretching and when not
to do stretching. This is not an Interop bug.

His second bug involved automatic line-breaking. What happens is that the
implementation for the line breaking turns it into a one column table with
a row for each of the lines. In Firefox, the equal signs are nicely
aligned. In Chrome, the equal signs are not aligned. NS will have to do
some more debugging.

NS's first two issues are not Interop bugs.

NS's third issue concerned named thicknesses. The parameters: thin, medium,
and thick, may not do anything.

NS: In core the only thing that is supported is a percentage and a length
percentage. Actual length isn't supported.

NS: So, I think the img tag has some funkiness where styles aren't fully
supported.

NS: Safari and Firefox do not support CSS.
<https://sandbox.cryptpad.info/code/inner.html?ver=5.5.0-b#cp-md-0-dc-action-items-vertical-spacing-height-problem-1-2-problematic-items-in-some-document-utr-doc->DC
action items: vertical spacing/height problem, 1-2 problematic items in
some document (UTR doc?)
<https://sandbox.cryptpad.info/code/inner.html?ver=5.5.0-b#cp-md-0-pl-clipboard-bugs>PL
clipboard bugs
<https://sandbox.cryptpad.info/code/inner.html?ver=5.5.0-b#cp-md-0-items-from-fred->Items
from Fred?

NS: DG and BK have been having a discussion on some things that are broken
in the web platform tests.

DG: The tests concern width and height. Chrome and Edge passed his tests,
but Firefox and Safari do not support those CSS properties.

DG: If you want to have a stretch capability, support the whole CSS box
sizing.

NS: Anything we submit to interop should have tests.

NS: We may not have to have tests at the time of submission; however, we do
need to have tests shortly after that. If we do not have tests, then our
proposal will be rejected.

DG: The width-height thing seems simple enough that it should not be
rejected. It is useful because you can both think of leveraging this for
stretchy Context and also for reflow eventually. And of course, the more
CSS properties we support, the more we can do proper web development with
CSS.

DG: had another issue about the padding border margin which you know a
little less about but it's also useful of course.

DG: Padding bugs are notoriously bad in CSS and I'm not an expert on the
So, Brian was asking me if I would be open to submit at least one interop
issue.

DG: I would be comfortable writing something about width-heights.

NS: We should be considering the CSS stretching issue and the sizing issue.
They are cross platform issues. We will wait for DC and PL to do some work
on these issues.

DG received an email from BK saying that the interop submission deadline is
today.

DG's issue is the only one that can be written to meet this deadline. It's
Fred's suggestion for the smallest possible CSS piece we can fix.

*CONSENSUS* DG will represent our group for the Interop submission.

DG may email his write-up to the members list for rapid proof reading.

DG will add text to his write-up saying that our group supports this issue.
<https://sandbox.cryptpad.info/code/inner.html?ver=5.5.0-b#cp-md-0-3-spec-changes-for>3.
Spec changes for

intent <https://www.w3.org/TR/mathml4/#mixing_intent> (what needs changing?
what needs to be added?)

NS: has removed omission, which describes blank spaces to be filled, and
has replaced it with the word blank.

DC wants to work on the properties tables.

NS: says we should add some more properties and say that our table is not
complete.

NS: these additional properties would give people an idea of what these
other properties are because these prefix, postfix, and silent functions
are in some sense very different than unit and matrix and so forth.

NS: MathCAT can change. MathCAT does not control what we use.

*ACTION* NS should send DC the latest MathCAT.

DG: One item that needs a specification update is the one for Math Color
and CSS that explicitly explains that the CSS colors will not be used for
accessibility.

DG and DC said the spec grammar is good.

NS: Next week we'll Hopefully dig into intent, and resolve some of these
issues.