W3C home > Mailing lists > Public > www-math@w3.org > March 2008

Re: Exploring new vocabularies for HTML

From: Bruce Miller <bruce.miller@nist.gov>
Date: Mon, 31 Mar 2008 15:14:57 -0400
To: Henri Sivonen <hsivonen@iki.fi>
Cc: Anne van Kesteren <annevk@opera.com>, David Carlisle <davidc@nag.co.uk>, ian@hixie.ch, public-html@w3.org, www-math@w3.org
Message-id: <47F13831.50201@nist.gov>

Henri Sivonen wrote:
> On Mar 31, 2008, at 20:17, Bruce Miller wrote:
>> So, Classic MathML, provided it didn't use namespace prefixes, I 
>> assume, would be valid to embed in HTML5?
> I consider that a requirement (at least when there are no OpenMath 
> annotation-xml subtrees).
>> Are you referring to the exporting of MathML as XML
>> a not-necessarily-mandated "UI feature" ?
>> That seems really bad to me.
> In general, UI is where HTML5 allows browser-vendors to innovate freely. 
> Firefox already has View MathML Source in the context menu, so I 
> wouldn't be too worried.
>> Besides, if, as you say, MathML as XML would be allowed in HTML5, 
>> there'd be no need for a browser to export the HTML5 serialization.
> This already came up in the SVG discussion. Even if you allow copying 
> and pasting (unprefixed) MathML or SVG XML source into text/html, you 
> still need an HTML5 parser and an XML serializer to extract 
> MathML-in-text/html or SVG-in-text/html from the Web into XML-only apps 
> *in the general case*, because someone out there *will* produce markup 
> that parses as HTML5 but not as XML.
> If you refuse to use an HTML5 parser either in the browser, as a 
> standalone tool or integrated into an importing editor, HTML authors 
> have a very simple copy protection mechanism to use against you. :-)
> The mistake with insisting on the syntax looking like XML in order to 
> enable reuse is the expectation that HTML authors will cooperate with 
> you. They won't--either accidentally or deliberately.

No, I'm not insisting that, at all.
Perhaps I haven't been clear, or I'm misunderstanding you.

I'm conceding that an HTML5 browser would accept a range
of MathML/SVG syntaxes from strict XML (so long as no namespace prefixes)
at one end, to the lax html-like syntax at the other end.
And, that authors would do... well, what authors do.

My concern was that the browser, after parsing whatever form
into a DOM, would be required (or _very_ strongly encouraged)
to allow exporting that DOM as XML.   Then, _any_ MathML (ditto SVG)
application could use the result, including old, strict MathML
applications on the one hand, to HTML5 browsers on the other.

Received on Monday, 31 March 2008 19:15:57 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:27:40 UTC