Re: MathML at the next CSS WG F2F meeting

On 25/02/2019 14:13, Daniel Marques wrote:
> My claim is that it is good for the health and future of MathML that
> it is properly integrated with HTML5. And that will imply that all
> MathML building blocks (fractions, stacks, scripting, stretchy, etc.)
> are also available as regular HTML5 (and CSS). It is something like
> the <table> tag that is also available as CSS (although both
> approaches are not exactly the same).

I think we agree on that too. This is one of the rationale for the
MathML Core spec. After the first experiments this year to implement
MathML in Chromium, we already had concrete details to discuss with
Google and now they seem open to work on CSS box models for math and
willing to report feedback to GitHub. In my opinion, such incremental
approach with concrete native implementation and discussion with browser
vendors is the way to proceed. If we start throwing new ideas without
any concrete plan and without interaction with browser vendors, that's
not going to work.

> This vision is fundamentally different from other people's of the
> MathOnWebPages. In the sense that I believe that a good implementation
> of MathML in the browser will inevitably result on better tools for
> handling mathematics with solely HTML5. While the position of some the
> people of the MathOnWebPages is the opposite: implement MathML using
> the available tools in HTML5 (thus not native implementation at all).

I have no idea at all about the agenda of MathOnWebPages's people but
you said they did not get any web engine developers involved and a
fortiori no concrete plan to implement features natively, so I don't
really see how they do better than the former Math WG. Moreover in
recent discussion, some MathOnWebPages people were even against the
effort made by the MathML refresh CG or by Igalia's "MathML in Chromium"
project, when these are currently the best bet we have right now to
collaborate with browser vendors on native math layout... Anyway, my
comment was really generic: The approach I suggested is used by all spec
authors and I just wish we do the same for the MathML Refresh CG.

-- 
Frédéric Wang

Received on Monday, 25 February 2019 14:06:49 UTC