Minutes: MathML Core meeting, 5 Oct, 2020

Attendees:

   -

   Neil Soiffer
   -

   Brian Kardell
   -

   Louis Maher


   -

   David Carlisle


   -

   Murray Sargent
   -

   Bruce Miller
   -

   Patrick Ion


Regrets: Frédéric Wang



Agenda:
https://github.com/mathml-refresh/mathml/issues/8#issuecomment-703384382

The meeting was recorded:
https://benetech.zoom.us/rec/share/trhNzT0Nqh0sIidh-iFxLcx2G-swhoamgHszbwSh1jRdY_OmFMxvSD8SnG_i2gGm.VYwnyWXykgNQn_uV
Passcode: @!Ce15$N  (meeting starts about 15 minutes in).


Charter

Draft: https://mathml-refresh.github.io/charter-drafts/math-2020.html

NS: Please look at and send comments to Neil or Patrick

TPAC Proposals: https://www.w3.org/wiki/TPAC/2020/SessionIdeas

NS: Added our proposal there with Brian's feedback. Suggested a time that
is an hour off "prime time" in the hopes we don't have conflict with other
groups and will draw more people.

The remaining 'needs resolution' issues:NS: The list is getting short...

Applying visual effects #179
<https://github.com/mathml-refresh/mathml/issues/179>Updates?

Nothing new

automatic size adjustment: #225
<https://github.com/mathml-refresh/mathml/issues/225>Updates?

Nothing new

U+002D HYPHEN-MINUS in mo operators: #146
<https://github.com/mathml-refresh/mathml/issues/146> => write a polyfill?
Updates?

Rob is/was going to try prototype (via math-auto text transform) but he’s
on vacation

Add scriptsizemultiplier to core #230
<https://github.com/mathml-refresh/mathml/issues/230>/ #1
<https://github.com/mathml-refresh/mathml/issues/1>Updates

NS: In talking to a few low vision people, the common thing appears to be
to decrease the resolution which causes reflow, etc., to work while
increasing the font size. They don’t like screen magnifiers because of the
work to horizontally scroll.

BK: There is a lot of variance among users - decreasing the screen
resolution is a nice easy coarse-grained switch that has a lot of great
knock on effect that works for a lot of people and a lot of users.  Screen
magnifiers are used by some, for some use cases, or if we get close enough
or, there's lots of things.  I feel like this needs non-trivial research
that we don't have and I'm not sure we are sort of 'ready' to solve this
particular problem.

MS: Can use touchpad or touch to magnify

NS: Could use ctrl+ to increase font on page also. But all these cause
significant reflow. I doubt people are happy increasing the size of a page
just because there is some math in it they have trouble reading.

NS: Brian -- are you saying drop it or move it to level 2?

BK: I  think we should at a minimum think of it as 'at risk' - that can be
a label if it is not in core now, to indicate something that might be
doable for core still but needs research or resolution.  I think the more
time passes though, the less likely it is to practically make it in, and we
should consider moving it to L2.

PI: Is there a problem with implementing it?

NS: I don’t think anyone has said it is hard to do. It requires at least a
private CSS property. Might be harder to be public property and get the CSS
WG to agree.

MS: I don’t think it should be in the markup but it should be changeable
via a user accessibility setting

BK: In order to move forward, we don’t have a current proposal on how to do
this in CSS. I’d like to propose we move it to level 2.

BK: I think it makes sense to add a note to the spec to say that in the
future this may be changeable.

Consensus: let’s move it to level 2 with the idea of adding a note
concerning the details of how changes might be carried out.

Action Item: NS to craft some text and do a PR with Brian and Fred as
reviewers.

Action Item: NS to find how well scriptsizemultiplier is supported by
browsers.
Include mo@accent attribute into MathML Core? #210
<https://github.com/mathml-refresh/mathml/issues/210>Updatesimplement

NS: see the comments at end of #210 which shows what the browsers actually
implement (not all 100% correctly; cf. Safari) (as I said I’d investigate
this);  it seems to me given usage and support in browsers, mo’s need to be
treated as accents, not limits, by defaults.

MS: Office puts the right accents in anyway, with or without an explicit
attribute

NS: So this should be part of Core; and we have to hope it can be
implemented though Fred seems not to find it attractive.  It was a feature
in the Operator Dictionary.

MS: you do need to display the accents correctly

NS: Fred points out the deficiencies of the spec design here where <mo> has
a way of doing under and over anyway and this would be redundant; however,
this feature is well used

BM: did you check both mechanisms work

NS: yes, as you can see from the table I checked the mover part (not the
munder)

DC: the reason for having on <mo> originally was so that using regular
accent chars like “^” work without extra markup.

NS: then we pulled out of the operator dictionary sometime in the last year
and applied to any operator. There still some complications about
embellished operators and CSS not being able to look arbitrarily into a
child; this complicates finding properties in children for CSS purposes.
However the group could not think of a real life instance where accents are
embellished and felt we should just remove the word “embellished” in the
spec for this case so there is no CSS issue and legacy content won’t be
affected.

DC: at present it’s true by default; it’s the cases setting it false that
are then at risk

NS: <<<

DC: accents on  <mover> should remain working if we can arrange it at all!

Consensus: MathML Core should support and @accent property on <mo> [nem con]

(Text in spec needs rechecking)

NS: I don’t know what Igalia’s implementation (branch) does at present.
They haven’t gotten to adding support for accents into Chrome itself yet.

BK: Below is a screenshot from Igalia's fork from last year:


NS: The first line is what legacy content does with the exception of the
‘=’ which was not listed as an accent in the operator dictionary. The
second line is with accent=true on the mover which is basically Fred’s
proposal. The third line is with accent=false on the mover. If an
implementation doesn’t support accent on mo, the first line would match the
third line.

NS: With the exception of ‘=’, the first and second lines are the same, so
the Igalia branch is implementing MathML 3 and not Fred’s proposal for core.

NS: Since the Igalia branch supports mo’s accent, hopefully this means that
it is relatively easy to implement in Chrome.

Review global attributes of HTML and SVG for parity #207
<https://github.com/mathml-refresh/mathml/issues/207>

Action Item for BK: try to get to this and find any elements besides
‘title’.

tab-navigation and links? #152
<https://github.com/mathml-refresh/mathml/issues/152>Move to level 2 and
remove 'needs resolution'?

BK: Move to level 2

User Agent style for href hyperlinks and focusable elements #94
<https://github.com/mathml-refresh/mathml/issues/94>Move to level 2 and
remove 'needs resolution'?

BK: Move to level 2

Next meeting will be in two weeks