Re: odrl-ISSUE-1 (renato): Namespace for Vocab terms [ODRL Version 2.0 XML Encoding (Contributors) ]

On 5 Dec 2011, at 05:28, Alapan wrote:
> 1. One feature of Namespaces is that it allows for validation; e.g. check that in the Namespace defined "o", there is something defined as "target". Depending on how the schema is constructed, quite a bit of structure validation can take place. By moving to URI, we loose this feature.

XML Schema's are really good at validating structure, not so good at content models (unless they are a fixed enumerated list).
(Note: V1.1 of XML Schema will address this to add more schematron like assertions…)

We use the Common Vocab as values to attributes. Since we don't mandate these, we leave the content model open for these attributes, except for the datatype of URI.

> 2. The idea of a URI is that it is resolvable to some resource on the Internet. In the proposed approach, we would have to create some sort of resolvable resource for each URI. My concern in this regard, relates to what is the envisaged resource that will be resolved to. For the given example, target is a directory (in HTTP convention), so what will target contain? 


*Eventually* I can see each URI resolving to an RDF/XML expression about that resource.
So you will get a machine readable expression describing "http://w3.org/ns/odrl/2/publish" for example.
This may include "sameAs" properties and other metadata.

Cheers...
Renato Iannella
Semantic Identity
http://semanticidentity.com
Mobile: +61 4 1313 2206