xml-dev - Re: draft-whitehead-mime-xml-04

[MURATA Makoto <murata@apsdc.ksp.fujixerox.co.jp>]

John Cowan wrote:
> Since one of the primary purposes of Content-Type: is to instruct
> a receiving program (email client, Web browser, or what have you) on
> the proper application to deploy in order to interpret the
> MIME entity, it seems to me that it would be useful to standardize
> a "type" parameter with application/xml and text/xml.  Legal
> values would be "document", "entity" (including both external
> parsed entities and external parameter entities), and "dtd".

Thank you for your comment.

Although the Internet Draft proposes text/xml and application/xml, the 
XML WG plans another mechanism for XML packaging (packaging of a 
document entity, external entities, styles, a catalog file for 
resolving FPI's, and so on.)  The Internet Draft is intended to 
provide minimum functionalities for successful interchange of XML.  
The rest is left to XML packaging.  The XML WG might want to introduce 
many more parameters of text/xml (e.g, references to DTD's, references 
to stylesheets, references to programs, etc.) or might want to 
introduce another media type.  I do not know yet.  (I also have to 
admit that XML packaging has not been seriously studied, though.  
It would be a service if somebody thoroughly studies this issue and 
submits a technical note to W3C.)

Regarding your proposal, an XML document entity can also be an 
external parsed entity.  XML declartations and text declarations are 
designed to allow this duality.  Thus, I do not think we should 
distinguish "document" and "entity".  If we distinguish the two types, 
users cannot make an XML document that is also an external parsed 
entity.  (However, one could argue that the type attribute should 
be multi-valued.)

An external parameter entity can also be an external DTD subset.  
Thus, "dtd" and "entity" should not be distinguished, as I see it.  The only 
reasonable thing is to distinguish "dtd" (including parameter 
entities) and "entity" (including document entities).

Then, is it necessary to distinguish "dtd" and "entity"?  Certainly, a 
text/xml-aware MIME recipient can invoke different programs for 
"dtd" and "entity".  However, it is also possible for an XML-syntax-aware 
program to autodetect "dtd" and "entity".  

Are there any strong reasons to introduce the "type" parameter in a 
hurry?  I can be pursuaded, but I am inclined to postpone this issue 
to XML packaging.  After all, we cannot make a perfect solution at this stage 
of the game.  In other words, text/xml and application/xml are not 
expected to provide enough information for launching XML applications.  
They merely provide the charset parameter.

> Typically, "documents" would be passed to an XML browser, whereas
> "entity"s and "dtd"s would be simply stored; alternatively, "dtd"s
> could be passed to a DTD viewer.  Passing "entity" or "dtd" MIME
> entities to an XML browser would or could produce an unwanted
> parsing error from the browser.

Makoto
 
Fuji Xerox Information Systems
 
Tel: +81-44-812-7230   Fax: +81-44-812-7231
E-mail: murata@apsdc.ksp.fujixerox.co.jp

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To (un)subscribe, mailto:majordomo@ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo@ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa@ic.ac.uk)