- From: Roger Menday <roger.menday@uk.fujitsu.com>
- Date: Mon, 19 Nov 2012 17:19:53 +0000
- To: "Wilde, Erik" <Erik.Wilde@emc.com>
- CC: Linked Data Platform Working Group <public-ldp-wg@w3.org>
- Message-ID: <32A83E4A-2AE9-4F60-8511-D5D77FB09038@uk.fujitsu.com>
>> ldp-ISSUE-38 (filtering, inlining): filtered representations and inlining >> [Linked Data Platform core] >> http://www.w3.org/2012/ldp/track/issues/38 >> Raised by: Roger Menday >> On product: Linked Data Platform core >> Section 5.1.2 describes a how to retrieve non-member properties of a >> resource. It might be good to generalise retrieving a 'filtered' >> representations. An further extension might be a way of specifying how to >> "inline" representations of linked resources. > > repeating my IRC comment for the issue tracker log: AtomPub has a model > for that as well, which is that the client decides how to add entries > (inlined or referenced), and this is opaque for the server. the server > just operates on the structures known to the service, so it just blindly > regurgitates whatever has been stored by clients. that's the only job the > server has: accept things, and then retrieve them on demand. if we want to > implant any specific knowledge of the things we manage into the server, we > have made the problem much much bigger. This specific issue wasn't about the 'storing' (the 'write') part. I was specifically concerned with the reading part. e.g. from GET /bugs, I'll get a list of links to individual bugs, but, then I need to cycle through the list to find out about each one. But, if I do GET /bugs?inline=bugs:has_bug, then I can make this more efficient. One question is related to issue.32: how can I discover this (rather then have query string construction algorithms). Roger > > cheers, > > dret. > >
Attachments
- application/pkcs7-signature attachment: smime.p7s
Received on Monday, 19 November 2012 17:20:46 UTC