Re: Annotation creation and updates

Hi,


> So, maybe the conclusion is that oa:annotatedAt is good enough:
> - in case of versioning, it expresses the creation date of the versioned annotation
> - in case of no versioning, it expresses the most recent update date of the non-versioned annotation



This sounds strange: I would say it's easier to keep just one semantics for the property (say, the most recent modification version), and invite the data consumers to check a PROV graph for determining extra information (e.g. the date of first creation). Only, ie. they wouldn't have to check the PROV graph *also* to find out which 'flavour' of oa:annotatedAt they are being faced with.

Otherwise while reading the thread I found the solution of having versions of an annotation as chains of different instances of oa:Annotation very seducing. Actually I think it is why I prefer the interpreation of "most recent modification version" because I'm convinced than when a user changes an annotation, if she changes the target (and/or the body) then it's really a different annotation, at the conceptual level.

But in that case, the 'deprecated' versions should be put in a graph different from the one where the 'most recent valid' annotations are.
If all annotations are in a same graph, then the burden would actually be on the data consumers who don't care about versions. They would have to read the entire provenance graph (which they're not supposed to care about) to find which annotations are valid. Otherwise they would get tons of useless annotations for the target objects they're interested in.

Cheers,

Antoine

Received on Thursday, 1 May 2014 15:19:13 UTC