- From: Florian Rivoal <florian@rivoal.net>
- Date: Tue, 4 Oct 2016 23:27:19 +0900
- To: Paul Topping <pault@dessci.com>
- Cc: "Tab Atkins Jr." <jackalmage@gmail.com>, Avneesh Singh <avneesh.sg@gmail.com>, www-style list <www-style@w3.org>, W3C Digital Publishing IG <public-digipub-ig@w3.org>, Peter Krautzberger <peter.krautzberger@mathjax.org>
> On Oct 4, 2016, at 07:07, Paul Topping <pault@dessci.com> wrote: > > I will admit to not being an expert on media queries but is the state of MathML implementation in browsers really a deciding factor on whether there should be a math MQ? It doesn't help. If we were in a world where all browsers had a flawless implementation of MathML, and we were just trying to tell them apart from more primitive UAs with no support at all, then the mathml MQ would have an easier time. Arguments saying that MathML might not want to be the format you want to use for distribution anyway may still stand, but if you did want to use it you could at least rely on the MQ. If we were to devise an army of a few hundred MQs, allowing you to test every single feature of MathML you plan to use in a particular formula before deciding whether to show it or not, you could theoretically make a more informed decision, but it is unlikely that any author would bother writing such a beast, and even then, it still wouldn't tell you if the feature is implemented well. > I appreciate there's a difficulty in deciding how good an implementation must be but that is surely a problem shared by other complex media types. Isn't MathML's value in accessibility enough to justify the creation of a math MQ? I am not a good judge of this myself, but the opposite statement was being made: MathML is not semantic markup and therefore not very good at accessibility, and there are more capable alternatives (e.g. involving HTML or SVG together with ARIA). - Florian
Received on Tuesday, 4 October 2016 14:27:51 UTC