[expertfinder-dev] Fwd: evolving foaf

Axel Polleres axel.polleres at urjc.es
Thu Feb 1 12:10:46 UTC 2007

I suggest you also browse the archivesof the foaf-dev list...
there was recently some movement starting again, it seems.

I think you raise some interesting points.

As for addresses there is ongoing discussion (also here as well as on
foaf-dev) how to integrate with e.g. vCard, etc.
The SWAP archives seem to be worthwhile crawling also, etc.


paola.dimaio at gmail.com wrote:
> (thanks Alex)
> Greetings all
> I am thinking of developing foaf to fit the emergency requirements as
> shortly discussed in the expertfinder workshop, and talking to people
> on and off lists about this idea.
> I plan to study the suitability of foaf for our requirements, what
> would be the best way to recommend some changes to foaf, or the best
> way to implement
> a modified version of it?
> I mean, its the only people schema that we know of to date, and I dont
> think we should create another people schema , but occasionally 
> reservations
> come up with the current schema. See the mail below that someone sent
> me . I would prefer to see foaf grow than having to create an new spec
> let me know what you think, and let me know how you think I  should go 
> about it
> cheers
> paola di maio
>  A glance at the current FOAF spec reveals some problems:
> 1) Name properties are western-centric.  Not clear they would work for
> traditional hispanic names and certainly not for arabic names.  I'm
> sure there are other culture-specific naming systems that would
> experience similar problems.  Secondly, the FOAF spec says you can
> encode the language of the name in an attribute.  What if you need
> access to that information?  Are not languages objects like any other
> that require first class treatment?  In short, naming is a hairy
> phenomenon.
> 2) FOAF has a bunch of properties relating Agent/Person to addresses
> of various kinds directly without respect to time.  Two problems here
> -- these "addresses" are not addresses of anything but services that
> anyone with the proper account name and password can access.  There is
> nothing at all essential about their link to persons.  Secondly, I
> know for a fact that some user names I used to use for some services
> have been recycled after my use of them has lapsed.  FOAF ignores all
> these details.
> _______________________________________________
> expertfinder-dev mailing list
> expertfinder-dev at lists.foaf-project.org
> http://lists.foaf-project.org/mailman/listinfo/expertfinder-dev

Dr. Axel Polleres
email: axel at polleres.net  url: http://www.polleres.net/

More information about the expertfinder-dev mailing list