- From: Roger Menday <roger.menday@uk.fujitsu.com>
- Date: Mon, 19 Nov 2012 10:01:35 +0000
- To: Andy Seaborne <andy.seaborne@epimorphics.com>
- CC: "public-ldp-wg@w3.org" <public-ldp-wg@w3.org>
- Message-ID: <233F877D-DA4F-4EC8-81C0-D8CDF3346519@uk.fujitsu.com>
>> >>> ldp-ISSUE-36 (container creation): Can and how do applications create new containers? [Linked Data Platform core] >>> >>> http://www.w3.org/2012/ldp/track/issues/36 >>> >>> Raised by: Andy Seaborne >>> On product: Linked Data Platform core >>> >>> Can containers be created by the application and if so, how? >>> Does the application control the URI pattern for items created in the new container? (see ISSUE-26). >>> >>> See thread from >>> http://lists.w3.org/Archives/Public/public-ldp-wg/2012Nov/0082.html >>> >> >> I think it should be possible that containers could be created in the application, but, I think this would be part of regular application flow, and no special mechanism needs to be reserved for this case. > > That would be good - could you show how you think it would happen > (triples and verbs used)? > I just mean that a resource in the system is created, by the server, as part of the application, and will come to life already as a container. I think it is less likely, but possible, that a resource that is originally a non-container, gets turned into a container later on as a result of something happening in the application. However, I think it is often more natural to consider it as a container which is *currently* not accepting/creating. All the same a client should be able to manage with all cases using resource 'reflection'. It is feasible (but, not particularly helpful) that a client might use the update aspect of LDP to make a new 'type' assertion about the resource, and this could trigger the switch from non-container to container. Roger > >> >> There was a conflict in recent discussion on the mailing list related to this issue about instructing a regular LDP resource to become a container, because in order to "receive" that directive, I believe that it would already have to be a container first. >> >> regards, >> Roger >> >
Attachments
- application/pkcs7-signature attachment: smime.p7s
Received on Monday, 19 November 2012 10:02:31 UTC