Le mardi 14 décembre 2010 à 10:00 -0500, Steve K Speicher a écrit : > > > > Is there something in the specs describing the standard representation > > of changes on a resource. > > No, not at the moment. >
ACK. > > In bugtrackers, we often have history views of the properties changes, > > with author and date, that help track who/what. > > This would be useful, it just takes someone from WG to own/drive it. > > Some things to consider when defining this: > - audit trails/history definitions that include what action had been > invoked > - how much detail needs to be exposed on what properties were modified? > - is this queryable? > > > Are such properties in the scope of OSLC-CM V2 (or more generally OSLC > > Core) ? > > Since OSLC-CM V2 is final, we could not add it to V2 proper. Though it > could be developed as an extension or part of a post-2.0 definition. > > We do have on our backlog [1] various properties like this and they came > out of the evaluation [2] we looking at common properties across many > different CM providers. I would recommend we start with some concrete > scenarios in CM, possibly creating a minimal definition for this and then > evaluating along the way how/when it should be a common resource > definition / property. > > Sound like a reasonable approach? > Sure. One may also imagine something more in the top-down approach that would consider the general problem of history of changes on any OSLC Resource, and the "theorization"/standardization of RDF resource modification tracking, which would then belong to Core, I guess. Your approach seems more reasonable in the short term ;) I'll think about the needs we have for this, and eventually come back with more concrete proposal. Thanks for the feedback. Best regards, -- Olivier BERGER <[email protected]> http://www-public.it-sudparis.eu/~berger_o/ - OpenPGP-Id: 2048R/5819D7E8 Ingénieur Recherche - Dept INF Institut TELECOM, SudParis (http://www.it-sudparis.eu/), Evry (France)
