[Om3] Complex Variables

[Robert Miner]

In a teleconference on MIME types just now, we discussed the encoding
attribute a bit.  I am going to put it first on the agenda for
tomorrow's WG teleconference.  

The issues we saw that tie in here are:

- should it really use a media type for the value?  That seems to partly
overlap with the clipboardflavor attribute
- we have a backward compatibility problem with MathML 2 giving the
value as MathML-Presentation.  Maybe that is
  just a question of exposition, but software is going to need to deal
with the old value

On the issue of having the cd/name attributes, my understanding was this
pair comprised an "annotation key" that specified the role of the
annotation.  I think Michael's point was he wanted something more
specific for the role other than "alternative representation" which is
the default when the annotation key attributes aren't there.  He was
suggesting introducing a new sub role called "complex variable".  

I want to carefully re-read ch5 before committing myself.  We should all
try to look at Ch 5 before the call tomorrow.

--Robert

> -----Original Message-----
> From: member-math-request at w3.org [mailto:member-math-request at w3.org]
On
> Behalf Of David Carlisle
> Sent: Wednesday, July 29, 2009 4:24 AM
> To: om3 at openmath.org; member-math at w3.org
> Subject: Re: Complex Variables
> 
> 
> Michael,
> 
> > Problem1: I remember that David and I talked about just making the
> first
> > form part of strict content MathML, and I had thought that we
decided
> to
> > do that for convenience,
> 
> Srict content mathml doesn't allow element content of ci. I suspect
> you were thinking about allowing the type attribute in strict
> 
> 
> > Problem 2: 5.3.2 says that " The value of the |encoding| attribute
> may
> > contain a media type that identifies the data format
> > for the encoding data."
> > So it should really be
> >             <annotation-xml encoding="|application/mathml-
> presentation+xml|">
> 
> 
> No, the sentence after the one you quote explictly says to use
> MathML Presentation
> 
> 
>   For example, Section 6.2.3 Names of MathML Encodings identifies the
>   strings MathML, MathML Presentation, and MathML Content as
predefined
>   values for the encoding attribute that may be used to identify
MathML
>   markup
> 
> actually the sentence in (5.2.3.2 Attributes) does need to be adjusted
> a
> bit now that there is a proposal to have a mime type for presentation
> mathml, to make it clear that the existing usage of using the name
> MathML Presentation takes precedence.
> 
> > Problem 3: the <annotation-xml> element in the example does not have
> @cd
> > and @name attributes. But I think we really should have them.
> Otherwise
> > they default to alternative-representation from the mathmlkeys CD.
> But
> > this seems to be too weak to key the intended behavior of a
> presentation
> > engine, namely to use the pMahtML directly.
> 
> I don't understand this comment. The default presentation of an
> objected annotated with "presentation mathml" is that presentation.
> I don't see how a cd attribute would effect the default presentaion at
> all (although of course a particular renderer may be affected by any
> markup changes), Perhaps it should have a cd otherwise I assume
> mathmlkes:alternate-representation will get used, but this doesn't
> seem too bad actually, and in any case, doesn't affect the
> presentation.
> 
> 
> 4.2.8.3
> 
>   When a Presentation MathML annotation is provided, a MathML renderer
> may
>   optionally use this information to render the MathML construct. This
>   would typically be the case when the first child is a MathML content
>   construct and the annotation is provided to give a preferred
> rendering
>   differing from the default for the content elements.
> 
> 
> 
> 
> > But do we really want to have the enclosing math tag around the
> > <msup>?
> 
> All examples in MathML2 didn't have enclosing math (or enclosing
OMOBJ)
> when mathml or openmath is used as annotations. I think that's the
> correct convention.
> 
> David
> 
> 
>
_______________________________________________________________________
> _
> The Numerical Algorithms Group Ltd is a company registered in England
> and Wales with company number 1249803. The registered office is:
> Wilkinson House, Jordan Hill Road, Oxford OX2 8DR, United Kingdom.
> 
> This e-mail has been scanned for all viruses by Star. The service is
> powered by MessageLabs.
>
_______________________________________________________________________
> _