Re: Recommendation for concise definition of LDPR

Henry,


On Sat, May 18, 2013 at 11:36 AM, Henry Story <henry.story@bblfish.net>wrote:

>
>  On 15 Apr 2013, at 09:35, Pierre-Antoine Champin <
> pierre-antoine.champin@liris.cnrs.fr> wrote:
>
>  Hi Cody,
>
>
> On Fri, Apr 12, 2013 at 9:13 PM, Cody Burleson <cody.burleson@base22.com>wrote:
>
>> Team,
>>
>> Please consider my draft recommendation for a concise definition of LDPR
>> in the introduction of the spec (so that we can be consistent in providing
>> true definitions for all terms in the Terminology section).
>>
>> Current text:
>>
>> Linked Data Platform Resource (LDPR)
>> HTTP resource that conforms to the simple lifecycle patterns and
>> conventions in the LDPRs section.
>>
>> Revision:
>>
>> Linked Data Platform Resource (LDPR)
>> An HTTP resource that can be represented by RDF, which is managed within
>> or served from a Linked Data Platform.
>>
>
>  This would preclude LDP servers to manage or serve "passive" RDF
> resources (i.e. that do *not* conform with the LDPR lifecylce patterns and
> conventions)?
>
>  If you GET and RDF representation from an LDP server, and that
> description makes no use of the ldp: vocabulary, then you should probably
> not expect the corresponding resource to behave as an LDPR.
>
>
>  Are you saying that all LDPRs must contain
>
>  <> a ldp:Resource .
>
>  either in their header or in their body?
>

Having this triple in the body would be an obvious way to do it, yes.
But I was not necessarily suggesting to impose that, and my phrasing was an
example.

For another example, and to revive another thread, I would consider that
having the header containing

  Content-type: text/turtle?profile=ldp

(note the explicit LDP profile) would be sufficient to assume that the
resource is an LDPR.

My point was: and LDP server should be able to serve plain turtle without
the clients to expect it to behave like a full LDPR. To be allowed to
assume that, clients should be provided with more explicit information,
either in the body or the header.

  pa


>
>
> So I think I like the old definition better.
>
>    pa
>
>
>> - Cody
>>
>>
>>
>
>        Social Web Architect
> http://bblfish.net/
>
>