[math-on-web] CG meeting minutes, 2018/03/01

Hi everyone,

The minutes are below.

The next meeting will be on March 15. Agenda items: an update from the CSS
TF, continue discussion on Ivan's input, move other TFs forward. Additions
welcome!

Best,
Peter.


# Math On Web 2018-03-01

 * Present: Arno, Neil, Peter

 * Arno: on Mathlive improvements
   * accessibility is in
   * HTML improvements coming in
* Neil: fmath was interesting to look at, quite feature rich
* Neil: malignmakr etc. what's CSS state of art there?
  * Peter: the general problem came up in the Publishing WG,  there's
latinreq by Dave Cramer https://www.w3.org/TR/dpub-latinreq/
  * but no progress.
  * Neil: what's the hold-up?
    * Arno: CSS doesn't really target text content
  * Neil: seems like it there should be lots of examples
  * Peter: agreed. E.g., LaTeX intertext within alignat environments came
up in my world the other day
  * Neil: but should be more general
  * Peter: yes. Seems like any table tool has alginment at decimal points.
Just not HTML/CSS
  * Arno: both vertical and horizontal have use cases
* Peter: about Ivan's point on "the plan"
  * Neil: personally would like to clean up MathML, remove the problems for
the web, I still believe in markup, believe it can be fixed
  * Arno: maybe we should arrive at "(cleaned up) MathML as an interchange
format" but not be part of HTML5
* Neil: @Peter don't you think there should be a markup?
  * Peter: depends what you mean. PresMML is markup for visual layout. It
provides next to no information beyond that. I don't think visual layout
should be markup.
  * For a11y, I'm aware that many AT users just want equality here, i.e.,
if a11y means just access to visual layout then that's equality with visual
users and enough for many
  * But that's hasn't been the web's approach for a11y and it does not
match my experience with equation layout (nobody says "x with dot above",
they say "derivative of x". Cf. users asking for "voice it like my teacher"
  * Neil: we had that a lot in clearspeak development with ETS. We added a
semi-flexible annotation system that would allow some modifications
* Arno: re "a plan".
  * Future of MathML re browsers: CG consensus seems to be: "we wouldn't
want ot encourage that"
    * instead, "make it easier to create visual display" is what we want to
encourage
    * using the different techniques
  * comprehensive plan seems a good idea
  * for Ivan/W3C they just hear "kill MathML" and the reaction is "we don't
want that"
* Neil: @Arno, do you see no need for set of tags for authoring math?
  * Arno: easy things can be done in HTML/CSS
  * for advanced, a markup language might be appropriate. But it doesn't
need to be "native"
  * Neil: but markup without a standard lacks power
    * I think there's value there
* Peter: I wonder if web-componenty, stdlib-ish, human-auhtoring focused
attempt at defininig semantically rich, visually specific is perhaps the
right direction
* Neil: people do write semantically but they don't specify it
  * Peter: true but print is holding this back a lot
  * Neil: I'm not sure if people will do it on the web or take it elsewhere