[rdfweb-dev] Proposal: Names in FoaF
Morten Frederiksen
mof-rdf at mfd-consult.dk
Tue Apr 29 18:51:21 UTC 2003
"Bill Kearney" <ml_yahoo at ideaspace.net> wrote:
>Ah, then using http://example.com/foaf.rdf#persons-rdfID might help
>clarify it.
>As in:
>http://www.ideaspace.net/users/wkearney/foaf.xrdf#wkearney99
Well, not really - at least by means of rdfs:seeAlso, which is just a
"pointer" to additional information, nothing else could or should be
infered.
However, if you used <foaf:Person rdf:resource="..."/> it would work,
but that assigns a URI to me as a person, something that FoaF is somewhat
trying to avoid by using inverse functional properties like mbox,
mbox_sha1sum and homepage, thus <foaf:Person foaf:mbox_sha1sum="..."/>
would be the prefered method.
>I think, however, the convention is that the 'first' foaf:Person fragment is
>'the entity' to which the document can be considered 'describing'. I realize,
>however, this is an assumption fraught with any number of possible errors. I
>suspect others haven't considered it.
I have this "problem" with my FoaF Explorer, and making up a rule is
very hard - I've seen (and try to handle) quite a few variations.
>My foaf document does have an rdf:Description for itself (at the end) and that
>uses a dc:creator pointing to the #wkearney99 identifier. Now, does a reference
>to the foaf document itself infer it's /my/ foaf?
It's /your/ foaf in the sense that you created it.
Other than that, it could be said that you are a trustworthy source for
information about you, and then that the information provided about you
from that URL should be trusted "more".
>> That is indeed an interesting point, one that foafnaut solves by
>> checking for signatures on the documents.
>Yeah, this is why asked this question months ago. I still haven't seen a
>satisfactory answer.
I'm sure Jim would be happy to tell us how he does it, I'm quite
interested as well.
Regards,
Morten Frederiksen
---
A diva who specializes in risqu'e arias is an off-coloratura soprano.
--
<URL: http://xml.mfd-consult.dk/foaf/explorer/?foaf=morten.rdf >
More information about the foaf-dev
mailing list