[rdfweb-dev] A definition of 'knows'

Bill Kearney wkearney99 at hotmail.com
Sun Aug 3 15:24:25 UTC 2003


> (The definition of "know" here is very broad. It is if you sat down next
> to that person on a train, you would know their name if they introduced
> themselves to you, and they would know your name.)
>
> Wondering whether this is a good gloss for foaf:knows
>
> Also brings up the question (discussed in IRC recently) about whether we
> say that foaf:knows is symmetrical, ie.
> whether ?x foaf:knows ?y can be possible without ?y foaf:knows ?x.
> Intuitively I find this quite appealing, but it has sensitive corner
> cases (eg. alzheimers)...

As we've discussed, I'd very much like to see a 'qualifying' technique applied
to the relationships as well as groups.

As in:

<foaf:Group rdf:nodeID="members">
    <foaf:member rdf:resource="someone"/>
</foaf:Group>

<rdf:Description rdf:about="#members">
    <ns0:type
rdf:resource="http://ns0.example.com/ns0/schema/Groups#mailing_list/>
</rdf:Description>

The idea being that we're adding a description stating the nature of the group.
The Group class (or whatever) would then have a core set of group styles.
Something on the order of mailing lists, community groups, physical, virtual,
one-time, on-going, seasonal, team league, etc.  I'm not sure exactly what sorts
of groups or what modelling to use so I'm asking for suggestions.  Expecting, of
course, that this will encourage people to use rdfs to extend their own group
types from this basic core set.

Is there an extant vocabulary for group relationships?  This leads to the same
question for 'relationship' types as well.  The idea there to add layering
markup that describes just how one 'knows' the target resource.

A nod goes out to the Mindswap efforts on Trust as well.

-Bill Kearney




More information about the foaf-dev mailing list