Re: //head/object

Henri Sivonen, Wed, 12 May 2010 03:56:18 -0700 (PDT):
> "Neil Soiffer" <NeilS@dessci.com> wrote:

>> Breaking this compatibility with HTML4 will  [...]

Below you tell that IE doesn't really rely <object> being in the 
<head>. Thus it seems more important that MathPlayer depends on:

1. @classid inside <object> - currently obsoleted in HTML5
2. xmlns namespace prefix declarations - a hot HTMl5 potato
3. PIs: <?import namespace="m" implementation="#mathplayer" ?>
   - parsing specced in HTML5, but syntax considered invalid.
4. In text/html: namespace prefixed MathML elements.  While:
   - IE is alone in seeing them as namespace prefixes in text/html
   - Hence prefixed mathml in text/html doesn't work cross browser

If we consider 'validity' only - ignoring cross browser text/html 
compatibility, then:

* HTML 4.0 does not serve the bill. 
* Monoglot HTML 5.0 even less so, 

While Polyglot XHTML5/HTML5 could work right out of the box, cross 
browser. 

>> including the users of our MathPlayer MathML renderer for IE.
>> I did an experiment and put the object tag inside of <body>
>> and MathPlayer was still triggered, but existing pages would
>> not work in HTML5. [...]

Becoming invalid doesn't mean it stops working. I also notice that your 
text/html based test suite performs tricks to hide the <object>: a 
comment as child 
"element" of <object>, to ensure that the <object> collapses in (at 
least) Webkit. 

http://www.dessci.com/en/reference/MathMLTestSuite/testsuite/General/Math/math1.html

By placing <object> in the body, you can be more certain that authors 
actually do the necessary tricks to hide the <object> from *all* 
browser.

I'm not against <object> in <head>. Just pointing out some issues that 
I think deserves attention first.

  [...]
> Do you have examples of typical pages that use MathPlayer?

Perhaps: http://courses.ece.illinois.edu/ece463/lectures/MLS.htm
-- 
leif halvard silli

Received on Thursday, 13 May 2010 03:35:02 UTC