Minutes: MathML Core meeting 23 Jan, 2020

MathML Core Meeting of Jan 20, 2020



Attendees:

   -

   Fred Wang
   -

   David Carisle
   -

   Neil Soiffer
   -

   Rob Buis
   -

   Patrick Ion
   -

   Murray Sargent


Regrets:

   -

   Brian Kardell


Agenda:
https://github.com/mathml-refresh/mathml/issues/8#issuecomment-576076397



The meeting was recorded:
https://benetech.zoom.us/rec/play/65R4d7j5_zI3H9PE4wSDBv8oW9S0ffmsh3cbr_NYyRmxB3NXNFGkZeATNOdguW6VwouSaj2-e_0x-a-I?continueMode=true

MathML meeting in May

FW: Some ideas for meeting:

https://mathml.igalia.com/news/2020/01/20/mathml-conference/#new

https://github.com/Igalia/webengineshackfest/wiki/MathML-Conference-2020

NS: conference good for spec work, actually implementing things

PI, NS: should mention this on appropriate mailing lists.(AP NS?)
Operator Dictionary (David & Neil)

   -

   any updates on fixing up entries
   -

   Make things more consistent and general review & fixup #87
   <https://github.com/mathml-refresh/mathml/issues/87> #6
   <https://github.com/mathml-refresh/mathml/issues/6> #151
   <https://github.com/mathml-refresh/mathml/issues/151>
   -

   #161 <https://github.com/mathml-refresh/mathml/issues/161> #176
   <https://github.com/mathml-refresh/mathml/issues/176>


NS: last week week we discussed removing ‘accent’ from the Op Dict, making
it true for all operators. Needs spec change also.

NS: Is there consensus?

DC: I’m in favor of it.

FW: It’s a big change.

NS: I doubt anyone will notice it because they don’t use the operators as
accents.

MS: In Office, we have separate ‘accent’ and ‘over’ properties.

<a bit more discussion as to what the spec says>

Resolved: take accent out of dictionary and change spec to say it is
default=true

DC: one way to compress the dictionary is list all the characters. We could
invert the logic and take the inverse approach: list all the
classifications and then list all the characters in that classification.
E.g., list all the infix characters.

NS: it’s nice to have all the info in one place

DC: let’s park this idea for now.

maction/semantics and embellished op #182
<https://github.com/mathml-refresh/mathml/issues/182>

FW: The basic idea is they only display one element. We already did this
space-like element -- for mrow, if all the children are space-like, then

FW: I want to do the same thing for embellished: switch to the same
definition for embellished operators

NS: Rob, you are probably most affected.

RB: I haven’t looked at it.

FW: I wrote the original code, so I know it will simplify things.

RB: Will this need fixing in Mozilla?

FW: Mozilla displays invalid markup if wrong number of children, so it
doesn’t really matter there.

NS: I’m ok with it -- simplification is always good.

RB: same  -- simplification is always good.

DC: I think it ok to go with the proposal. The goal of core is to make
things simpler.

Resolved: accepted.
Reset CSS rules on the math too (UA stylesheet): #34
<https://github.com/mathml-refresh/mathml/issues/34> #35
<https://github.com/mathml-refresh/mathml/issues/35> #36
<https://github.com/mathml-refresh/mathml/issues/36>

FW: the basic idea is to put some default values on the UA stylesheet.
Google asked what would happen with overflow and scrollbars:

https://github.com/mathml-refresh/mathml/issues/45#issuecomment-576126653

FW: Rob was going to add a “!” rule to the UA stylesheet to ignore them for
all but containers (e.g., mspace).

FW: I’m not sure it is a good idea for scripts.

NS: I can imagine someone wanting scrollbars for large scripts

FW: Google thinks it might be simpler in the end to support width/height.

RB: If you introduce w/h, then you end up with scrollbars.

FW: Rob maps w/h to CSS w/h. If someone specifies

FW: Currently the core spec says w/h are ignored, but the implementation
overrides the attributes.

RB: Need to add tests for this.

FW: also there is an issue if someone specifies h/d, then how does baseline
interact when someone specifies height in CSS. This tricky and not sure
what to do.

#35: reset dir to ltr on math element

FW: The MathML 3 spec set it to ltr on math, ignoring the environment.
Makes sense because many RTL languages do LTR math.

FW: CSS writing-mode also to get reset. Use case: mtext in MathML in
vertical paragraph. Should a user expect vertical text or horizontal text?
If we reset the writing mode, they would get horizontal text.

NS: let’s postpone a decision until next week when Brian will likely join
us.