RE: [MathML3-last-call] media-types mentioned in encoding attributes

Hi.  This message records the resolution of the last call comment quoted below.  The changes were:
 
- fixing incorrect casing of the encoding attribute values MathML, MathML-Presentation 
  and MathML-Content in chapters 3, 5 and 6.
- changing the encoding to application/vnd.wolfram.mathematica for the Mathematica example.
- changing the encoding to application/x-tex for the examples of TeX annotations
 
--Robert
 
CVS logs:
presentation-markup.xml
revision 1.307
date: 2009/11/22 19:05:12;  author: rminer;  state: Exp;  lines: +2 -2
corrected casing of encoding values

mixing.xml
revision 1.127
date: 2009/11/03 22:22:17;  author: rminer;  state: Exp;  lines: +2 -2
 
world-interactions.xml
revision 1.67
date: 2009/11/22 19:05:13;  author: rminer;  state: Exp;  lines: +3 -3
corrected casing of encoding values
 
 
Dr. Robert Miner
Vice President, Research and Development
 
Design Science, Inc.
140 Pine Avenue, 4th Floor
Long Beach, California 90802
USA
Main:   (562) 432-2920
Direct: (651) 223-2883
Fax:    (651) 292-0014
robertm@dessci.com
www.dessci.com

________________________________

From: www-math-request@w3.org on behalf of Paul Libbrecht
Sent: Wed 10/7/2009 8:03 AM
To: Public MathML mailing list
Subject: [MathML3-last-call] media-types mentioned in encoding attributes




Hello all,

MathML3 now define three media-types in the appendix B, this is the 
objective of another mail I will send to ietf-types. My objective is 
to discuss the encoding attribute values within examples.

Since MathML3, the annotation and annotation-xml elements carry the 
encoding attribute with the following recommended content:
1.- MathML-Presentation
2.- MathML-Content
3.- MathML
4.- the media-type of the annotation
5.- some text value you expect someone will recognize

If you look at the specification we have the following example values; 
I note that there are spurious occurrences of versions of 1, 2, and 3 
with a different casing which we should clean. Aside of 1, 2, and 3 
there are:

- well known media-types: image/png, text/plain, application/xhtml
+xml, image/tiff, image/svg+xml. No issues here.

- not yet standardized media-types:

--  application/openmath+xml: should probably be done by the OpenMath 
society one day, this follows directly RFC-3023 so there's no big 
debate here

-- application/x-maple: is justified by some documentation pages on 
MapleSoft's documentation and a private email exchange with Andrew 
Smith of MapleSoft. This is done for the sake of an "unregistered 
media type".

-- application/mathematica has been requested to change by Jason 
Harris of Wolfram Research to
    application/vnd.wolfram.mathematica
in the mail I reproduce below. I hereby propose to perform this change.

- non-standardized string: TeX, For this type, I have to say, there is 
no standard and it seems impossible to safely define one. We could 
replace it by application/x-tex which is what the Apache server 
delivers for a .tex file. On this point, I request suggestions.

thank you in advance for your comments.

paul


> De : Jason Harris <...>
> Date : 1 octobre 2009 19:17:04 GMT+02:00
> À : paul@activemath.org
> Objet : Rép : MathML3 followup  [TS 37366]
>
> [...]
> ----------------
>
> For the MathML 3 sepc, can someone update the draft in the 
> appropriate places to reflect the following change: Mathematica's 
> MIME type will change in the next version.
>
> In detail:
> We're registering new MIME types for future versions of 
> Mathematica.  The one which would replace application/mathematica 
> is...
>
> application/vnd.wolfram.mathematica
>
> Assuming this is accepted by IANA (We have already received initial 
> comments with only minor requests for updates, so I have no reason 
> to expect a rejection), this is what we'll use in future versions of 
> Mathematica. application/mathematica will become a legacy type.
>
> One of the differences in application/vnd.wolfram.mathematica is 
> that the version specification is now *optional*, not required.
>
> Sincerely,
>
> Jason Harris
> Wolfram Research
> [....]