[MathOnWeb] CSS task force minutes 2018-11-19

Hi everyone,

Here are the minutes from the meeting of the CSS task force this week.

Best,
Peter.

# Math On Web CSS TF, 2018-11-19

* Present: Dani, Peter
* Dani: baseline follow up
  * Peter: do we just point them to the spec we found?
  * Dani: not enough, we need to align at child still
  * same for mathaxis
* Dani: next steps?
  * Peter: hoping Arno might have more suggestions from the rest of TAPC
  * https://github.com/w3c/csswg-drafts/issues/1339#issuecomment-300293130
    * tabatkins and fantasai agreed there
      * parent has "I have special baseline"
      * child has "I'm the baseline"
  * Dani: I ran into the following: what if multiple children say "I'm the
baseline"
    * Peter: oh, good problem.
      * could we find CSS specs that might have a similar problem?
      * eg. subgrids -but two subgrids don't create conflicts.
  * Dani: then there's looking through baseline container, finding a child
that is baseline container
    * seems like it should stop looking?
    * Peter: eg nested fractions
  * Peter: yes, unless this child is itself a baseline container
    * bottom up: the child informs the parent but that parent might inform
its parent etc
  * [some discussion around this]
  * Peter: in terms of what to do with multiple children declaring
baselines, we just choose first or last and let CSS WG say what they prefer
  * Dani: maybe we should continue the thread with this
    * Peter: sounds good.
  * Dani: then what's next?
    * Peter: after discussion with Arno, maybe a Houdini-based polyfill
      * before going to a C++ patch example
    * Dani: Houdini is a bit different though
      * basically, "this box has a completely different layout"
    * Peter: I need to read up :)
    * Dani: but seems doable
* Dani: about stretchy constructions
  * maybe we can file an issue
  * we probably need to come to an agreement whether border glyphs are good
enough
  * Peter: I don't mind either way, I think border glyphs have independent
use cases and any implementation opens up the path to the other
    * if they're interested, I'd put our foucs there
    * Dani: sure but perhaps other options are just as interesting
      * we should at least try
  * [some discussion]
  * Dani: ACTION let's gather some ideas (eg. in a google doc)
    * be creative, figure out what we can do
    * we need to figure out
* [some discussion about stretchy vs size switching]