Re: PROV-ISSUE-55 (are-provenance-uris-needed): Are provenance URIs really needed [Accessing and Querying Provenance]

Hi Graham,

I think it's unfortunate.

You seem to dismiss the case where there is no natural URI for provenance.
In situations where provenance is dynamic, a query has typically to be 
issued, to retrieve the provenance
related to a given context-uri. The implementer will have to extract the 
context-uri from the provenance-uri.

I appreciate your reference to the architecture document [1]. But to say:
- To benefit from and increase the value of the World Wide Web, agents 
should provide URIs as identifiers for resources.
and follow it by:
- The term "resource" is used in a general sense for whatever might be 
identified by a URI.
strikes me as somewhat circular.  So, what kind of good practice are we 
trying to follow?

For the case identified in the issue I raised, I believe that provenance 
is better referred to by a query (containinig
a context-uri). I don't doubt that this query can be encoded as a URI, 
but that doesn't make it a natural URI.

For this reason, I believe that we should not encourage one approach or
the other, and we should have a neutral presentation.

Cheers,
Luc




On 25/08/11 13:47, Graham Klyne wrote:
> I think it's entirely appropriate that we should *encourage* 
> developers to allocate and use URIs for accessing provenance. 

[1] http://www.w3.org/TR/2004/REC-webarch-20041215/#pr-use-uris