[rdfweb-dev] Smushing the Semantic Web / Invalid data in IFP's

Ian Davis iand at internetalchemy.org
Tue Mar 23 16:21:33 UTC 2004


Hi Julian,

On Tuesday, March 23, 2004, 3:26:28 PM, Julian Bond wrote:
> There's one or two people (round here) that have FOAF files that are
> mostly about themselves. But they also contain foaf:Persons that don't
> have a knows relation to them and have significant information about
> that person. This is not bad RDF, but I think it is bad FOAF. In the
> sense that you shouldn't be asserting statements about other people 
> beyond a name, mbox_sha1sum and maybe a seeAlso. And you shouldn't be
> doing even that except to declare a knows relation.
I still don't agree with this viewpoint. I can say anything about
anything. Whether or not you choose to believe me is another matter.
It certainly makes things technically simpler if a given document
talks about only one thing (see Patrick Stickler's definition of
Concise Bounded Resource Descriptions in [1]). But we need to strike a
balance between simple and expressive and on this one I think I'd
rather allow more expression. Think how limiting (and dull) it would
be if we could only talk about ourselves.

Besides there are people who are only described by other people's
FOAF such as my son. Not to mention all the dead people who are
referred to via ancestral relationships.

Ian
[1] http://sw.nokia.com/uriqa/URIQA.html

-- 
http://internetalchemy.org | http://purl.org/NET/iand 




More information about the foaf-dev mailing list