W3C home > Mailing lists > Public > www-style@w3.org > October 2016

Re: [mediaqueries] MathML

From: Peter Krautzberger <peter.krautzberger@mathjax.org>
Date: Tue, 4 Oct 2016 13:05:18 +0200
Message-ID: <CABqxo800m1XRVCkc46ZsMt0iL-r7R+7qBz8=0E0-bp9VMiCE4w@mail.gmail.com>
To: Manuel Rego Casasnovas <rego@igalia.com>
Cc: Florian Rivoal <florian@rivoal.net>, www-style list <www-style@w3.org>, W3C Digital Publishing IG <public-digipub-ig@w3.org>
PS Apologies. That first line should read "I think most people interested
in MathML are aware (I am anyway)."

On Tue, Oct 4, 2016 at 1:04 PM, Peter Krautzberger <
peter.krautzberger@mathjax.org> wrote:

> Hi Manuel,
> I think most people interested in MathML are as well (I am anyway).
> They do not change my personal assessment of either those implementations;
> evidence from crawlers indicates that nobody uses native implementations,
> though it's obviously too early to say that much about Safari 10
> The fundamental problems with MathML as a web standard remain. And as you
> point out, the implementations, in critical parts, does not follow a spec
> (in the W3C or WHATWG sense) but that document written by Fred Wang (who is
> the last volunteer standing, even if it's good to see him find a formal
> home at Igalia).
> Presentation MathML does not specify layout sufficiently and it is non
> semantically meaningful, thus not accessible. And even if MathML was
> suddenly implemented widely, we'd still need CSS modules to match,
> otherwise MathML implementations will lock perfectly reasonable layout
> features into specific tag names / namespaces.
> Regards,
> Peter.
> On Tue, Oct 4, 2016 at 12:45 PM, Manuel Rego Casasnovas <rego@igalia.com>
> wrote:
>> Hi,
>> On 03/10/16 04:47, Florian Rivoal wrote:
>> > - MathML has been stagnating for a long time, and implementations do
>> not seem to be making much progress. There is no reason to believe that
>> implementations (either visual or screen readers) will improve sufficiently
>> any time soon (or ever) to make the previous two problems go away.
>> Actually, there's been some work recently around MathML that some of you
>> might not be aware of.
>> As you probably know MathML has been supported by Mozilla for a long
>> time in Firefox, you can check it using the MathML Torture Test [1].
>> WebKit also had an implementation that is now more or less similar to
>> the one in Firefox thanks to the work done by Igalia in the last year:
>> https://webkit.org/blog/6803/improvements-in-mathml-rendering/
>> On top of that, we've an experimental branch to add MathML support in
>> Chromium too [2], and we're in conversations with Google to see if we
>> could rely on this work to bring MathML back to Chromium.
>> Just to add more context, there's a MathML implementation guide [3] and
>> test suite [4] developed by the MathML Association, that have been used
>> as reference on the implementations of these 3 engines (Gecko, WebKit
>> and Chromium).
>> Apart from that, past year we were also doing some work regarding MathML
>> and accessibility in Firefox, to add speech support for MathML content
>> to the Orca screen reader:
>> https://blog.grain-of-salt.com/index.php/2015/09/23/new-in-
>> orca-3-18-firefox-support-rewrite-and-mathml/
>> Last, Microsoft has stated that they'll take a look to MathML support:
>> https://twitter.com/SampsonMSFT/status/727199790736384001
>> Of course, like for many other specs there is a lack of resources to
>> implement and properly support MathML on different browsers and platform
>> screen readers. However it seems clear that there is some stuff ongoing
>> around MathML lately. As you can see, at least from Igalia, we've been
>> doing some contributions to improve MathML and we could do even more if
>> we've a stronger support.
>> Cheers,
>>   Rego
>> [1]
>> https://developer.mozilla.org/en-US/docs/Mozilla/MathML_Proj
>> ect/MathML_Torture_Test
>> [2] https://github.com/fred-wang/chromium.src/tree/mathml
>> [3] http://www.mathml-association.org/MathMLinHTML5/
>> [4] http://tests.mathml-association.org/
Received on Tuesday, 4 October 2016 11:05:52 UTC

This archive was generated by hypermail 2.4.0 : Monday, 23 January 2023 02:15:01 UTC