[rdfweb-dev] RE: FOAF-interest usage

Leigh Dodds ldodds at ingenta.com
Mon Mar 15 09:44:40 UTC 2004


> > I've never been particular comfortable with foaf:interest,
> > mainly because of it's range of foaf:Document. I'm not interested
> > in a document, but a subject.
>
> Well, the way it's currently described, it actually does mean that:
> "The foaf:interest property represents an interest of a foaf:Agent, through 
> indicating a foaf:Document whose foaf:topic(s) broadly characterises that 
> interest." -- http://xmlns.com/foaf/0.1/#term_interest
>
> The advantage as compared to a more "free-form" description of ones interests 
> is that it is much easier for machines to realize when two people share an 
> interest.

Sure, but thats all predicated on the fact that people will use the same 
document as exemplars of their interest. This is relatively straightforward for 
say, XML and RDF as one can point to either the spec or the activity, but 
not that useful for something more general like "Information Architecture", or 
"Gardening".

And if I make up my own URIs then there's even less chance of useful 
aggregation.
 
Perhaps there's a best practice issue here, as I've got similar reservations 
with how foaf:tipjar (newly added to the spec I see) is specified. 

This term also has a range of foaf:Document, but notes that "We expect, but do not 
currently specify, that this will evolve into a hook for finding more 
machine-readable information to support payments..".

This means, to me at least, that we may ultimately want to use something 
other than foaf:Document in that element, e.g. ns:PaymentMechanism. 

If so, isn't it always better to use a more general range of Resource when 
describing a term whose usage is expected to develop "organically", rather 
than overspecifying things initially?

Cheers,

L.



More information about the foaf-dev mailing list