[rdfweb-dev] fwd: Broadband Mechanics requests and proposals

Morten Frederiksen mof-rdf at mfd-consult.dk
Tue Jul 22 16:57:06 UTC 2003


Marc wrote:
>2. We're planning on going beyond just making friends.  We wanna
>help make all sorts of relationships, but ONLY friends need to be
>'verifiable'.
I think you have realized by now that the verification is not something
you'll be able to handle with a centralized system - RDF and FoaF is
distributed, and an e-mail confirmation isn't going to keep anyone from
making statements at will.

However, I can imagine that some messaging would be nice in the context
of a "system of friends" - just make sure that nobody gets the
impression that it's the only way to make friends (and statements)...

>So what we'd like to do is formerly extend the vocabulary used for the
>'second part of the triple' (or whatever it's called)  - when it comes
>to friends links.  
That second part of the triple is usually called predicate or
property, with the first and third part called subject and object,
respectively.

I agree that a wider set of predicatess relating people would be nice -
how about it if we do a little brainstorming here, based on pixel's
relationship predicates, and then perhaps have pixel update it if he
agrees on what we find?
If he doesn't, there's always the option of creating af "competitor"...

>4. There are apparently two different ways to represent friends of
>friends.  One seems to be the 'RDF' way - which declares a
>rdf:resource="marccanter", and then use sthat syntax to refer to friends
>while the other more FOAFy approach - seems to be to just declare
><person><knows><name>Marc</name></knows></Person>. Can we assume that
>this duality approach will continue?
Heh, there's only one way: Triples...

The above (and the other 2-3 "ways") are just syntax variations in
RDF/XML, and not variations that you should care about - just use an
RDF/XML parser, and you'll end up with the triples you need.


Regards,

Morten Frederiksen
--- 
!07/11 PDP a ni deppart m'I  !pleH
-- 
<URL: http://xml.mfd-consult.dk/foaf/explorer/?foaf=morten.rdf >



More information about the foaf-dev mailing list