RE: [dita] Use of standardized prefixes when incorporating foreignvocabularies

["Yas Etessam" <yas.etessam@xmetal.com>]

Yes, validating XML editors should be pulling the values from the
schema/dtd.  Albeit, sometimes some piece of connector code between an
XML editor and a MathML renderer (where the Math tools might not be
reading the DTD/schema) might expect a particular prefix.

If a particular MathML connector uses something like "m:" as the default
prefix, which doesn't match the namespace prefix, it is the connector
that should change when implementing the system, not the DTD or schema. 

From this vendor's perspective, it is not a requirement that DITA
defines standardized prefixes in the specification.  

Best regards,

Yas Etessam

-----Original Message-----
From: W. Eliot Kimber [mailto:ekimber@innodata-isogen.com] 
Sent: Friday, October 06, 2006 11:43 AM
To: Eric Sirois
Cc: dita@lists.oasis-open.org
Subject: Re: [dita] Use of standardized prefixes when incorporating
foreignvocabularies

Eric Sirois wrote:
> as the prefix.  We both use our favorite XML editor to write up the
topics.
> You try to process that topic within your publishing stream  which use

> your specialized version of MathML and the parser is throwing some 
> errors because it does not recognize the mml prefix.

I'm confused by the "parser...does not recognize the MML prefix" because
it's not a parser issue--that is, either the topic is valid with respect
to its governing DTD or schema or it's not. When authoring a document it
would not be possible to create one with the wrong prefix assuming the
editor is syntax driven or I'm validating as a I go.

Once I get to the point of processing, anything that is MathML-aware
better by gosh be looking for the MathML namespace URI and not a
particular prefix. If my processor is in fact expecting a particular
prefix then my processor is broken, which is a "my processor is broken" 
issue, not a namespace prefix or DTD vs. schema problem.

If somebody has the expectation that they can take two topics created
using different local schemas (DTD or XSD doesn't matter) and blindly
smash them together into the same document instance then they are simply
not thinking clearly because it's never possible if you expect to use
one or the other of the original document's DTDs to validate the result,
as a general case (of course it might work in a particular case but that
would just be coincidence).

Cheers,

E.

--
W. Eliot Kimber
Professional Services
Innodata Isogen
9390 Research Blvd, #410
Austin, TX 78759
(214) 954-5198

ekimber@innodata-isogen.com
www.innodata-isogen.com