Re: request for comments HTML WG comments to MathML WG

On Thu, Nov 5, 2009 at 7:26 AM, Simon Pieters <simonp@opera.com> wrote:
> On Thu, 05 Nov 2009 14:15:23 +0100, Shelley Powers <shelley.just@gmail.com>
> wrote:
>
>>>> MathML entities have now been moved to a separate document. These
>>>> entities will not be available with MathML embedded in HTML, served as
>>>> either HTML or XHTML, because there is no doctype.
>>>
>>> They will work in HTML because the HTML5 parser supports them regardless
>>> of
>>> doctype.
>>>
>>
>> These are not part of the HTML4 specification, which is what MathML
>> references.
>
> Ah. Referencing HTML4 seems useless since HTML4 doesn't support embedding
> MathML inline.
>

Color names are pulled from the HTML4 spec, and it's mentioned again
in Chapter 6[1], with an unnamed aside to HTML5:

"To be effective, MathML must work well with a wide variety of
renderers, processors, translators and editors. This chapter raises
some of the interface issues involved in generating and rendering
MathML. Since MathML exists primarily to encode mathematics in Web
documents, perhaps the most important interface issues relate to
embedding MathML in [HTML4] and [XHTML], and in any newer HTML when it
appears."

So anything I comment on related to MathML in HTML, rather than XHTML,
is somewhat moot, because the document just does not recognize MathML
in anything but an XML-based syntax, such as SVG or XHTML.

Speaking of which, there are new elements, and new constraints on
structure. We may have to do our own updating in the HTML5 document,
if we decide to reference MathML 3.0 instead of MathML 2.0. I didn't
examine this aspect of the document.

>
>> And I don't see them as part of the HTML5 specification.
>
> http://www.whatwg.org/specs/web-apps/current-work/multipage/named-character-references.html#named-character-references
>

I did not find some of the named entities listed in the MathML named
entities draft in the entities given in the HTML5 draft.


>
>>> They work in XHTML if you have a special doctype and your browser knows
>>> about that doctype (which Firefox does but Safari and Opera don't
>>> currently).
>>>
>>
>> Again, I was focused on XHTML5, not previous versions of XHTML.
>
> The special doctype can be used in XHTML5.
>

Oh, yes. Well, this is also another area of difference between the
MathML document and HTML5.

>
>> Yes,
>> there is a doctype that will work with MathML, but I don't believe it
>> references all of the MathML 3.0 entities.
>
> Ah. I guess it should be updated to include the latest set of entities.
> (Minting a new FPI for MathML 3.0 seems bad since it will break compat with
> shipped Firefox.)
>

There are differences from both XHTML 1.0 and MathML 2.0, as shown in
the new working draft [2].

Would you like me to add this as a comment?

> --
> Simon Pieters
> Opera Software
>

Shelley

[1] http://www.w3.org/TR/MathML3/chapter6.html
[2] http://www.w3.org/TR/xml-entity-names/

Received on Thursday, 5 November 2009 13:56:43 UTC