Minutes: MathML Core meeting Nov 4, 2019

MathML Core Meeting of Nov 4, 2019



Attendees:

   -

   Fred Wang
   -

   Brian Kardell
   -

   Neil Soiffer
   -

   David Carisle
   -

   David Farmer


Agenda:
https://github.com/mathml-refresh/mathml/issues/8#issuecomment-541619307

The meeting was recorded:
https://benetech.zoom.us/recording/share/LhTjXs3lKsGI1l6H2qmnSXK6joeH7qOvHR4esAmSA1KwIumekTziMw
Report from Igalia hackfest (Fred's talk
<https://people.igalia.com/fwang/mathml-in-browsers-web-engines-hackfest-2019/slides>
)

   -

   FW: Slides in link above. Talk is available on YouTube
   <https://www.youtube.com/watch?time_continue=79&v=Q8Z1D2i61j8>
   -

   FW: One point mentioned in the talk was that for Chromium a write up for
   “intent to ship” was done.
   -

   FW: Have started upstreaming. Sent two patches already: 1 just for build
   system stuff, another to enable the web platform tests to be verified/run
   for chrome.  Probably our next one will be to update the DOM stuff as it
   seems uncontroversial and easily independent from the rest.
   -

   Good feedback so far.
   -

   Other things at the hackfest:
   -

      Neil working on some tests
      -

      No Windows build of Igalia work
      -

      Good conversations about CSS with fantasai

Brian/Neil discussion about exploring elementary math

   -

   Good conversations about deeper details of some issues like whether
   every element really should be able to act as a link, and the implications
   of this (currently things that are links cannot have a Shadow DOM, but it
   took some investigation to verify that this was a conscious choice and
   probably unlikely to change)
   -

   FW: Added a new run time flag to disable some features and added some
   stats
   -

   Rob worked on design doc (now done)
   -

   Brian talked with people about details about links and security concerns
   -

   There was some discussion on when the CG should move to a working group.
   The general feeling was that we aren’t ready to do that yet. If the TAG
   group feels it should be done sooner rather than later, the consensus is
   that we should do it sooner.

MathML Conference(s)/Event(s)?

   -

   FW: Have budget for holding a conference at Igalia’s headquarters (A
   Coruna, Spain).
   -

   FW: Not sure what the time frame is yet.
   -

   FW: Not sure the format. Maybe talks. Maybe some hacking, some spec
   writing, …


TAG Review process

   -

   BK: Brian filed preliminary spec review
   <https://github.com/w3ctag/design-reviews/issues/438> with TAG.
   -

   BK: Fred and Rob helped with the initial draft.
   -

   BK: Plan is that there is a face-to-face in December to discuss MathML
   there.
   -

   BK: Needed as part of intent to implement for Chrome.



Other conferences:WebKit contrib meeting
<https://trac.webkit.org/wiki/IgaliaFocusandGoals2020>:

FW: Rob attended and gave a presentation that included MathML. One question
<https://trac.webkit.org/wiki/IgaliaFocusandGoals2020#Questions> was “are
you reviving the MathML spec?”. Rob answered “yes”.

BlinkOn 11: next
<https://groups.google.com/a/chromium.org/forum/#!msg/blink-dev/LHsv4eqJ80Q/VRzGd0Y8BAAJ>


FW: Fred will be there in two weeks and will present MathML and in
particular Chromium aspects. Brian will be there also, presenting a
lightning talk on MathML Core (maybe)


Changes to spec to support CSS proposals

Discussion of names. “cramped” is not intuitive to non-math people.

FW: it is used in TeX and in OpenType and they may or may not be the same
thing. And they may or may not be the same as how the spec uses it, which
comes from Firefox.

BK: I don’t care about the name itself, just make sure it is documented.

FW: In OpenType it is “superscriptshift”.

NS: probably best to deal with in issue tracker to list the various
definitions. Then choose one or define our own.

BK: Either be precise or just use the vague English definition (of
handwriting) ‘with characters written small and crowded together (of a
style of writing) hard to understand)’
Remove/simplify ms? #120
<https://github.com/mathml-refresh/mathml/issues/120> #126
<https://github.com/mathml-refresh/mathml/issues/126> (only implemented in
Gecko)

NS: used for computer systems because they use different delimiters

DC: we use a little at NAG, but not a big deal.

FW: the problem is the attribute.

BK: there are accessibility and direction issues

FW: already in all browsers

DC: is the problem the attribute support?

FW: yes

DC: not putting the quotes would be a change in existing documents, but I
could live with it because there are other changes.

FW: the plan is not to implement in Chromium for now.

BK: maybe we can use CSS for the common cases where lquote and rqoute are
the same

NS: I’m pretty sure that covers almost anything everyone does.

BK: FW has that suggestion in the issue, but claims there are accessibility
and other issues.

NS: there are issues around accessibility of before/after, but they will
get resolved independent of what we do, so I don’t think we should worry
about accessibility if we use them.


We will have a meeting next week.

<http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
Virus-free.
www.avg.com
<http://www.avg.com/email-signature?utm_medium=email&utm_source=link&utm_campaign=sig-email&utm_content=webmail>
<#DAB4FAD8-2DD7-40BB-A1B8-4E2AA1F9FDF2>