Re: PROV-ISSUE-627 (URI-specified-or-REST): Should PROV-AQ specify PROV service URI or always use template?

A key element of the current design is that it permits (but does not require) conneg as a mechanism to select alternative service description semantics.  Using RDF raises some challenges because it has multiple content types corresponding to syntactic variations, so using conneg for semantic alternatives  in RDF is less easy.  (e.g. A different "+xyz" suffix for each RDF syntax seems cumbersome.)

My view is that the nature of RDF allows an additional selection mechanism to be applied within the RDF data.  Conneg is still usable, but in RDF vocabulary-driven mechanisms can be used.  Like RDF types or the property-based approach Stian proposes.

But using conneg or not seems to me to be somewhat orthogonal to whether we should be trying to fix the URI patterns that a provenance service responds to.  REST principles clearly say not.  I think we've taken a clear step down the REST path, which seems appropriate to me for a Web specification, so I would urge sticking to those principles for maximum flexibility and evolvability.

#g.


Paul Groth <p.t.groth@vu.nl> wrote:

>Interestingly this impacts my view of the rdf syntax requirement of the
>service description.
>
>A key question for me is does the mandate for rdf for the service
>description hurt potential uptake from people who provide services
>using other very common techniques (e.g. Json web services) . My prior
>view was the definition of a best practice URL was a way of helping
>adoption.
>
>Hmmm - I will look again at service description formats. I think
>content negotiation may be the way out...
>
>Paul
>
>On Feb 7, 2013, at 20:27, Graham Klyne <graham.klyne@zoo.ox.ac.uk>
>wrote:
>
>> Tim,
>> 
>> Things have changed from last time, because the same link relation is
>being used 
>> to access multiple query mechanisms (something you argued for :^) ). 
>This is a 
>> design choice that is a large extent guided by REST principles.  So
>now I shall 
>> argue more strongly for sticking with a full-REST pattern.
>> 
>> The client should not have any prior knowledge of how the server URI
>space is 
>> used.  That's a basic constraint of REST.  Hence the client MUST
>learn about the 
>> URI from the service (in this case from the servioce description). 
>That's what 
>> the template defines.
>> 
>> #g
>> --
>> 
>> On 07/02/2013 16:18, Timothy Lebo wrote:
>>> On Feb 7, 2013, at 9:52 AM, "Provenance Working Group Issue Tracker"
><sysbot+tracker@w3.org> wrote:
>>> 
>>>> PROV-ISSUE-627 (URI-specified-or-REST): Should PROV-AQ specify PROV
>service URI
>>> +1
>>> 
>>>> or always use template?
>>> -1
>>> 
>>> (just as in the last time we had this thread)
>>> 
>>> -Tim
>>> 
>>> 
>>>> http://www.w3.org/2011/prov/track/issues/627
>>>> 
>>>> Raised by: Graham Klyne
>>>> On product:
>> 
>> 

-- 
Sent from my Android phone with K-9 Mail. Please excuse my brevity.