Minutes: MathML Full meeting 2 June, 2022

 Attendees:

   - Sam Dooley
   - Bert Bos
   - Steve Noble
   - Louis Maher
   - David Farmer
   - Patrick Ion
   - Stephen Watt
   - Murray Sargent
   - Paul Libbrecht

<https://sandbox.cryptpad.info/code/inner.html?ver=5.0.0-02#cp-md-0-regrets>
Regrets

   - Deyan Ginev
   - Cary Supalo
   - David Carlisle

<https://sandbox.cryptpad.info/code/inner.html?ver=5.0.0-02#cp-md-0-1-announcements-updates-progress-reports>1.
Announcements/Updates/Progress reports

NS: ARIA WG is has our issue (https://github.com/w3c/aria/issues/1723)
about ARIA and MathML on their agenda today

NS: Still some discussion about intent grammar (what is a number
https://github.com/w3c/mathml/issues/376, functions with no args
https://github.com/w3c/mathml/issues/378)
<https://sandbox.cryptpad.info/code/inner.html?ver=5.0.0-02#cp-md-0-1-1-deprecating-maction>1.1
Deprecating maction

PL: He requested comments on deprecating maction. He made this request on
the www-math list. He did not receive any responses.

*ACTION:* PL will request comments deprecating maction on the MathJax list.

MUS: There is more than just the web to support. How can input be handled
without maction?

NS: Can't you do this with attributes?

MUS: is "data-" in the spec?

NS: You can always add it. However, it is private to the author.

MUS: If you are doing an interactive thing, you might need maction.

MUS: wrote a blog post about this:
https://docs.microsoft.com/en-us/archive/blogs/murrays/editing-math-using-mathml-for-speech.
This post mentions some possible uses of maction for editing MathML using
speech.

MUS: should respond to the mailing list on how he wants to use maction.

NS: wants someone to comb the web to see if maction was ever used.

NS: Is there someone who has a searchable version of the web?

MUS: Someone has studied the frequency of how often certain Emojis Are used.

It might be worth asking if Google can provide a searchable web.

*ACTION* PL will ask for comments on deprecating maction on the MathJax
list, and MUS will post his input needs using maction and attributes.
<https://sandbox.cryptpad.info/code/inner.html?ver=5.0.0-02#cp-md-0-2-continue-reviewing-the-a-href-https-w3c-github-io-mathml-mathml-4-doc-a-and-issues>2.
Continue reviewing the MathML 4 doc <https://w3c.github.io/mathml/> and
issues

SD: rewrote section 4.1

SW: asked for a discussion on the issues surrounding intent.

NS: DG is a major participant in this discussion.

NS: We might get back to the intent discussion in two weeks, depending on
where we are in spec writing.

BM: We should leave the intent discussion parked until we totally
understand what intent intends.

NS: Chapter 4 is in good shape. Chapter 3 needs work. The spec is almost
ready for editorial work.

NS It is OK if the first public working draft does not have all issues
settled.

SD: showed his rewrite of the beginning of the content chapter. People are
welcome to comment on SD's work.

SD: will delete the paragraph about coverage that NS did not like.

SD: will review the rewritten appendices F and G.

SD: is not satisfied with the titles of sections 4.2 and 4.3. He wants the
titles to do a better job of describing what those sections cover.

NS: asked people to review chapter 4, which is the content chapter.

DC: did a lot of work on chapter 4.
<https://sandbox.cryptpad.info/code/inner.html?ver=5.0.0-02#cp-md-0-3-media-types>3.
Media types

   - Proposed rec: https://w3c.github.io/mathml-docs/mathml-media-types/
   - Issues: https://github.com/w3c/mathml-docs/issues/38
   - ]Current preview](https://w3c.github.io/mathml-docs/mathml-media-types/
   )

PL: Thanked DG for his media type comments.

NS: I just clicked through to RFC3023 and sees that it only has
Application/mathml+xml. This doc is 2001. Is there a better/more up-to-date
reference?

DC had written: 3023 is just the general foo+xml scheme, it does not get
updated for specific instances of xml media types. The currently registered
up to date reference for the 3 MathML types is the appendix in MathML3 that
this would replace. There is no corresponding +html scheme, you'd use
text/html to get html parsed 3023 MathML

NS: The rec points to the 2002 standard . This has been superseded but the
new document has not been fully approved. The 2002 standard is also not
fully approved.

NS: The document says that MathML will be added in the future. I think it
would be a good idea to include the IANA ref/link so that people can easily
find that the registration has been completed.

IANA registry:
https://www.iana.org/assignments/media-types/media-types.xhtml (currently
uses MathML3 as specification)

PL and NS came to agreement on the references.

SW: The 3023 reference is ancient. He wondered how other specs handled this
situation.

NS: looked into the SVG documents to see if they had this ancient reference
issue. He did not find anything in the short time he had during the meeting.

PL: We can reference 2033 and any follow-up references.

PL: will continue to work on the Math Type document. We will consider
voting on it next week.

NS: Does anyone expect to have objections to voting on it next week? No one
responded.
<https://sandbox.cryptpad.info/code/inner.html?ver=5.0.0-02#cp-md-0-4-common-core-attributes-and-what-they-do-on-non-core-elements>4.
Common core attributes and what they do on non-core elements

See issue
<https://sandbox.cryptpad.info/code/(https://github.com/w3c/mathml/issues/361>
.

We have common attributes on core. What would we do about non-core
documents?

NS: What do we say about attributes in non-browser settings? Do they apply
to presentation MathML?

SW: Do we want core to be a subset of MathML 4? We still want MathML to 3
to be valid. Every core document should be valid MathML 4.

NS: event handlers may not make sense in non-browser applications. We
should say that the behavior is implementation defined. Core never says
that it is only for browsers.

BM: MathML 4 is a superset of core. There are things in MathML 4 which are
not accepted in MathML 3.

SW: If we do not know if an attribute should be accepted in full MathML,
perhaps the attribute is not designed properly.

SW: We should have a comment saying that some non-normative things are
there for completeness although they may not be used.

BM: I'm in favor of MathML 4 being a proper superset of Core. The group
seems to have agreed with this.

NS: The full MathML spec is a superset of core. we should add a paragraph
saying non browser renderers are free to do what they need to do.

*ACTION** NS: people should review chapter 4 which is the content chapter.