- From: Ashok Malhotra <ashok.malhotra@oracle.com>
- Date: Thu, 16 May 2013 13:19:02 -0400
- To: public-ldp-wg@w3.org
Hi Raul: Issue-57 http://www.w3.org/2012/ldp/track/issues/57 and, perhaps, also, Issue-32 http://www.w3.org/2012/ldp/track/issues/32 which are open, address at least a part of the problem you are referring to. All the best, Ashok On 5/16/2013 10:25 AM, Raúl García Castro wrote: > Dear all, > > The current specification does not impose any absolute (MUST) restriction on LDPR representations. Therefore, "almost" any server returning text/turtle and satisfying some other protocol restrictions would be an LDP-conformant server. > > Besides, it is difficult for a client to discover if a server or its resources are LDP or not; therefore, it is difficult to know their behaviour (e.g., that linked resources can de dereferenced). > > Proposal: > To require, similarly as for LDPCs, that LDPR representations are typed (i.e., "The representation of a LDPR MUST have rdf:type of ldp:Resource, but it MAY have additional rdf:types."). > > Kind regards, >
Received on Thursday, 16 May 2013 17:20:00 UTC