IMO, the statement:
http://viaf.org/viaf/64055977/ skos:note “He smelled very bad”
contains no provenance information about the _statement_. It is
dangerous to infer any from the subject URI (although one might infer
something about the provenance of the subject URI itself). If people
interpret that wrongly as Kevin mentions, that seems like an education
issue.
Now, perhaps if I downloaded that statement as LOD from
http://viaf.org/viaf/64055977/ then I might reasonably infer something
about provenance (perhaps that OCLC asserted, approved or collected the
statement; to be specific one would need out-of-band knowledge of OCLC
and that dataset).
If one really cares about the expressing the provenance then it should
be made explicit and systems will need to keep track of it. This could
involve recording where things came from in "identified datasets" as
Jeff Young suggests (separate branch of the thread) or more heavyweight
methods such as named graphs, nquads etc...
Being able to served LOD from a subject URI does seem like a valid
motivation to mint new URIs that are owl:sameAs existing ones. However,
I don't think that should be conflated with provenance interpretation.
Cheers,
Simeon
On 7/10/14 2:30 PM, Ford, Kevin wrote:
> Dear Karen,
>
> The complexity arises if someone other than OCLC (?) asserts something
> about a VIAF URI.
>
> So, for example (and, please note, I’ve used a different URI, because,
> well, the one your provided is problematic for my purpose here), what if
> /I/ asserted:
>
> http://viaf.org/viaf/64055977/ skos:note “He smelled very badly, all the
> time.”
>
> On the surface it looks like VIAF/OCLC said Raphael smelled badly, all
> the time because, provenance-wise, the domain tracks back to OCLC. In
> reality, I said it, but you cannot tell.
>
> If we use global URIs, and additional statements are made against those,
> then something like this could happen.
>
> Yours,
>
> Kevin
>
> *From:*Bibliographic Framework Transition Initiative Forum
> [mailto:[log in to unmask]] *On Behalf Of *Smith-Yoshimura,Karen
> *Sent:* Thursday, July 10, 2014 1:40 PM
> *To:* [log in to unmask]
> *Subject:* Re: [BIBFRAME] Bibframe and Linked Data (Authorities)
>
> Kevin –
>
> Re:
>
> >There's been lots of talk about provenance and the like in a global
> graph of data, but I feel most of those discussions rely on fairly
> technical mechanisms, the complexity of >which outweigh the simplicity
> of minting one's own URI. (Also, the provenance statements will need
> their own URIs!)
>
> Doesn’t http://viaf.org/viaf/54202464 show the provenance is VIAF?
> What’s complex about this?
>
> Karen S-Y
>
|