Fwd: Re: ISSUE-36 (Kill Radion?): Should RADion be killed off? [DCAT]

-------- Original message --------
Subject: Re: ISSUE-36 (Kill Radion?): Should RADion be killed off? [DCAT]
From: Chris Beer <chris@codex.net.au>
To: Dave Reynolds <dave.e.reynolds@gmail.com>
CC: 

Is there a simple solution? (Thinking out loud here) - go for the crosswalk - a skos:sameAs ref in each for the relevant parts *only when they are actually the same thing*

I see both vocabs being useful when used in context : a repository object might be a dataset (under development for instance) and a dataset might be datum on repositories or repo objects, but the two are not mutually exclusive.

When there is a "collision", such as when the repo object *is* a dataset, then DC principles would dictate that this is ok - the radion:distribution is describing an attribute of a discrete repository object which happens to be a dataset, and dcat:distribution is describing a dataset attribute, noting that the dataset may or may not be held in a repository depending on the user/publisher.

All very practical LD from the sounds of it?

(Only pitfall I can see is a potential circular ref when a dataset containing repository data is held itself in a repository ;) )

Chris


Sent from Samsung Mobile