Ray,
I agree: the W3C XML schema should become normative.
We can also make the DTD normative, provided that we make sure that everything works when both DTD and schema are used with one instance-document.
Would there then be anything left in the text that is normative? We would need to be clear about which bits were normative and which became informational.
Cheers,
Guy
On Thu, 20 May 2004, Ray Plante wrote:
> Hi VOTablers,
>
> I'll confess that I have not actually read the WD "VOTable Format
> Definition v1.09" start to finish; however, my previous post about COOSYS
> brings up a relevent point for this document. As my comment reveals, the
> prose document does not define the format to the same level of detail as
> the XSD file. I therefore recommend that we establish a close coupling
> between the standard document and the schema.
>
> That is, we should (as you did in the 1.0 version) include the Schema
> and/or the DTD as an appendix to the document. Then in the document
> (perhaps in section 1.3), we should say that when using the XML Schema
> approach (marked by the use of xmlns attributes), this is the schema that
> defines a compatible VOTable; when the DTD approach is used, this is the
> DTD that defines a compatible VOTable. Because of the greater ability to
> constrain syntax with Schema, the XSD file should be considered the more
> definitive standard for syntax. Even when an application is using the DTD
> to validate, the application should assume the stricter constraints
> defined in the XSD.
>
> Finally, I may have mentioned this one about the namespace. I think it
> would be appropriate to base the namespace URI on www.ivoa.net. In
> particular, I recommend we follow the convention used with the resource
> metadata schemas, and use "http://www.ivoa.net/xml/VOTable/v1.1".
>
> cheers,
> Ray
>
>
Guy Rixon gtr-at-ast.cam.ac.uk Institute of Astronomy Tel: +44-1223-337542 Madingley Road, Cambridge, UK, CB3 0HA Fax: +44-1223-337523Received on 2004-05-21Z11:55:47