[math-on-web] CSS TF meeting minutes, 2018/02/26

Hi everyone,

The minutes are below. The next meeting will be on March 12.

Best,
Peter.

# MathOnWeb CSS TF 2018-02-26

* Present: Dani, Peter, Arno

* Dani: agenda
  * Peter: my demos, website, what do we want to do
* Dani: re webfonts
  * I think they need to be part of the discussion
  * =>  agreed
  * Arno: I think it's typical for CSS to provide options
    * eg math tables sound good but we don't have access in CSS
    * if you want to have very good quality, then you need access to them
  * Dani: I would want to have that without JS
     * Arno: agreed.
  * Arno: e.g., placing integral limits needs the slant of the int glyph
for precise location
    * need CSS to say "offset this with the value of X of that font"
  * Dani: but you could pre-generate that data and use a webfont
  * Arno: sure but I think we should go for something that works the way
the web works
  * Dani: I think it's useful to not aim for that but to base it on current
tech (latest browsers)
    * e.g., ebooks
    * asking CSS to change standards comes after
  * Arno: so how is that differnet from current solutions (e.g., mathlive)
  * Dani: goal to try to use pure CSS as far as possible, e.g., no absolute
positioning
  * Arno: that sounds like what I'm doing. No absolute positioning but
still some data (cf. integral slant above)
* Peter: I think a key problem is quality vs font-independent rendering
  * people have become accustomed to TeX-quality. If you go for something
lower, people get confused
  * mediawiki's template is the only serious case for lower quality.
* Dani: I think it will help to keep these separate. Be clear when you're
doing what
  * we should communicate eg to standards: we can do math but we need your
help for quality
* Peter: I'm also interested in identfying features with a greater focus
for the web
  * samples on stretchy constructions
  * [discussions]
* Peter: I think it would be helpful to the community to collect examples
like this to give tools
  * Arno: the diversity of approaches is very good
* Peter: linking examples from TF page
  * codepens ok? what about the older examples?
    * Dani: just codepen
* Peter: about baseline alignment codepen examples
  * adding that link to a mathjax-style munderover
  * @Dani is your example misaligned?
    * Dani: no. The glyph renders in the usual tradition
* Dani: re quality. You might see things on a retina-like device but you
might not see it
* Arno: so what about quality?
  * Peter: I'm currently more interested in getting the "low" quality up;
or perhaps: *more* interested i that than in pushing down / making "top"
quality easier
  * Arno: e.g., you chose a thinspace in that codepen. It's good but not
quite right? But we don't have mathspaces
  * Peter: I honestly don't remember
* Arno: other examples todo? ACTION items
 * Peter: maybe italic slant and scripts/munderover?
 * Peter: I had Custom Paint for fences on my list but seems beyond my
skills
 * Peter: what about menclose?
   * Arno: I could do that
 * Dani: square roots and general roots are tricky
   * Arno: don't fences solve this?
    * Dani: not for general roots. vertical placement of index is tircky
for strethcy
 * Peter: we should still think about long term directions
   * arno: definitely. It's good collecting these examples but also good to
start thinking about the REC track which will be long term anyway.