Minutes: MathML full meeting: 9 June, 2022

 Attendees:

   - Louis Maher
   - Neil Soiffer
   - David Carlisle
   - Cary Supalo
   - Stephen Watt
   - Sam Dooley
   - Deyan Ginev
   - Patrick Ion
   - Murray Sargent
   - Moritz Schubotz
   - Paul Libbrecht
   - Bruce Miller
   - Steve Noble

<https://sandbox.cryptpad.info/code/inner.html?ver=5.0.0-03#cp-md-0-regrets>
Regrets
<https://sandbox.cryptpad.info/code/inner.html?ver=5.0.0-03#cp-md-0-1-announcements-updates-progress-reports>1.
Announcements/Updates/Progress reports

NS: We did get feedback from the ARIA group. They said let MathML do what
MathML wants to do.

DG: Use ARIA as a stopgap, otherwise use MathML.

MOS: Intent is going to be used in Wikipedia.
<https://sandbox.cryptpad.info/code/inner.html?ver=5.0.0-03#cp-md-0-2-a-href-https-w3c-github-io-mathml-docs-mathml-media-types-advance-media-types-to-fpwd-a->2.
Advance media types to FPWD
<https://w3c.github.io/mathml-docs/mathml-media-types/>?

PL: All the comments have been addressed. The references have been updated.
The document is ready to become an FPWD.

NS: Are we ready to go to the FPWD stage?

DC: The MathML spec 4 has not yet gone to the first draft stage.

NS: proposed, and DC seconded, making the Media Type document an FPWD.
Consensus to move it to FPWD.

*ACTION* NS will send BB a note to this effect.

NS: Sections 6.3 and 6.4 need work. He would like them to be smaller.

PL: We could move some of the discussion to notes. We could use folding. We
could try to go for more thinning. He does not think the content should be
thinned.

NS: Some of the examples can be eliminated.

NS: Does the CSS section in chapter 6 need updating to align with core?
<https://sandbox.cryptpad.info/code/inner.html?ver=5.0.0-03#cp-md-0-3-issues-of-the-week>3.
Issues of the week

a) Use character entity reference or character directly in spec
<https://github.com/w3c/mathml/issues/381>.

MUS: It is dramatically easier if you have the Unicode characters there.
You could add a comment if you wanted to show the actual code.

DG: We could have a JavaScript polyfill showing you whatever view you
wanted.

DC: There is a MathML 3 (non-rec) version with checkboxes to do this. We
generally do not put JavaScript code in the spec.

Conclusion: Use characters instead of character entity references.

b) Intent Grammar (recursive heads
<https://github.com/w3c/mathml/issues/380>,

floating point numbers <https://github.com/w3c/mathml/issues/376>,

DC discussed CSS numbers. They allow leading '+' along with 'e' notation.
Do we wish to allow this format?

NS: Do we allow leading decimal points?

NS: What floating point numbers should we allow?

DG: What is this for? There is a benefit in having a small grammar.

DG: Allowing multiple syntaxes for the same thing makes the spec grow, and
thus harder to understand.

DG: The decimal comma is alive and well.

DG What do we achieve by having two ways to write floating-point numbers?

SW: It is nice to allow international radix. If we cannot have a leading
decimal than we cannot have a trailing decimal.

DG: Rust and Javascript object to leading decimals points.

Let CSS figure this out and accept their spec.

SW: Do we expect intent to refer to colors in numeric syntax including
hex-based characters?

NS: No.

SW: Will intent use colors represented by numbers?

NS: We will use the color name and not the number.

BM: Some small complications with the grammar should not bother
programmers.

Most of the time when you read a number, it is in the presentation tree.

NS: sometimes you want to say .12 or 0.12.

??: Should we use ARIA for something like this?

NS: We will not be using aria to read these numbers.

PI: There will be few numbers in intent. Therefore, we can allow 0.12 or
.12; or could insist on 0.12 and even 1.0 without causing much harm.

NS: When NS worked with the Educational Testing Service (ETS), there were
AT issues about how you would speak numbers that were in scientific
notation so as not to bias the answer or make it easier or harder for a
student using AT vs a sighted user. Hence, controlling the speech for
numbers is important in that situation.

NS: NS and DG will settle this.

empty arguments <https://github.com/w3c/mathml/issues/378>).

NS: does not have a position on this issue.

DG: Showed an example of a random number generator function which did not
have arguments.

DC: We use functions that test to see if you have a license to run certain
software. Some of these functions do not have arguments.

NS: It is a small amount of work to handle this.

DG: Just wanted to make sure that there was a reason to accept functions
with empty argument lists.

The committee decided to accept functions with empty argument lists.

Intent: implicit functions allowed in function names
<https://github.com/w3c/mathml/issues/380>

NS: Arg ref let you implicitly do this. BM pointed out that it required
adding an mrow, so explicitly allowing functions in the head would be good
so remediators don't have to change the mathml.

DC: Should we add another level of parens so look ahead is not needed to
determine where the head ends?

NS: I don't like that idea. MS and I are working on intent -- let's see if
we have a problem with the implementation.

Agreement to add complex heads to intent

c) Media Types -- remove appendix and update chapter 6

NS: Get rid of the media type appendix.

d) Use images in appendix F <https://github.com/w3c/mathml/issues/383>?

DC: Is ok with dropping the images in appendix F. SD agrees.

DC will do this.

SN: will work on the accessibility appendix. The result should be in HTML.
DC will help SN put the appendix in HTML format. Our group has a Wikipedia
page describing this process


.

NS: Topics for next week: maction, location and format for known and
unknown intent