[dsdl-discuss] Fwd: Response to the W3C CDF WG feedback on NVDL FCD

I sent the attached message to the W3C CDF WG. This is based
on the discussion of WG1 in Atlanta.

Cheers,
Makoto

Forwarded by "MURATA Makoto (FAMILY Given)" <EB2M-MRT@asahi-net.or.jp>
----------------------- Original Message -----------------------
 From: "MURATA Makoto (FAMILY Given)" <EB2M-MRT@asahi-net.or.jp>
 To: member-cdf@w3.org
 Date: Sun, 20 Nov 2005 18:56:05 +0900
 Subject: Response to the W3C CDF WG feedback on NVDL FCD

----
Response to the W3C CDF WG feedback on NVDL FCD
2005 November 20
ISO/IEC JTC1 SC34 WG1
We appreciate the feedback from the W3C CDF WG.  We studied 
the comments carefully.
1) Namespace-based dispatching cannot handle RDF very well
It is true that NVDL cannot handle valid RDF fragments such as:
<Person xmlns="http://example.org/people">
  <Name>Mark</Name>
  <Height>180</Height>
</Person>
To handle such RDF fragments, we need a very powerful mechanism for
extracting subtrees.  SC34 WG1 considers that such a mechanism 
is outside the scope of NVDL.
2) Trigger elements are not clear.
We agree that the text on trigger elements can be improved.  
The project editor is instructed to improve the text probably 
using an example.
SC34 WG1 makes sure that NVDL can handle XForms embedded within XHTML
even when XForms uses the XHTML namespace.  This requires a small
change to NVDL, and the project editor is instructed to incorporate
it to the FDIS.
Alternatively, one could use namespace renaming, which DSRL is
expected to allow, and then apply NVDL.
3) Can NVDL impose constraints on locations of foreign elements?
Yes, the path attribute of context elements can provide such constraints.
4) Can qnames in attribute values be validated by validators invoked
   by NVDL dispatchers?
Yes, since the data model of NVDL has contexts, which are mappings 
from prefixes to URIs.  Validators invoked by NVDL dispatchers 
will receive such contexts.
5) More examples are needed.
At this stage, it is not easy to introduce more examples as part of
the NVDL FDIS without significant delay.  However, volunteers in SC34
WG1 intend to publish examples at DSDL.org.
6) Can attribute inheritance be handled?
Attribute inheritance is outside the scope of NVDL.  One way to handle
it to use an upcoming part 8 of DSDL, namely the Document Schema
Renaming Language, as a preprocessor to NVDL.
7) Modes are not clear.
Again, volunteers in SC34 WG1 intend to publish examples at DSDL.org.
8) When are options needed?
Neither RELAX NG nor W3C XML Schema need such options.  However, 
the final Schematron specification may introduce some such options.
-- 
MURATA Makoto (FAMILY Given) <EB2M-MRT@asahi-net.or.jp>
--------------------- Original Message Ends --------------------
-- 
MURATA Makoto <murata@hokkaido.email.ne.jp>
--
DSDL members discussion list
To unsubscribe, please send a message with the
command  "unsubscribe" to dsdl-discuss-request@dsdl.org
(mailto:dsdl-discuss-request@dsdl.org?Subject=unsubscribe)