[rdfweb-dev] Re: Ecademy FOAF
wkearney99 at h...
Mon Dec 23 19:17:00 UTC 2002
> But I believe that alternative solutions, such as a peer 2 peer search
> model, may be more effective. And most searches that people will undertake
> will centre on themselves, either as the source of the human network or the
> destination. If I was running a crawler, I might want the crawler to go up
> to 3 links away from me and then stop - for many applications (such as
> "introduce me to someone who also likes kitesurfing") this will be enough.
Or to make the integration even more interesting, "I'm going to be in the San
Diego area on the 3rd of March, tell me who in that area is interested in
kitesurfing". That's taking not only subject matter but also temporal and
spatial metadata into account.
> The idea of a global index of FOAF users may not be as important as a global
> index of HTML documents.
I agree being able to find /content/ is much more likely to be useful than
finding people. However, once content is found, being able to 'pivot' the
facets based on grouping will open up whole new vistas of discovery.
It's sort of like reading slashdot's comments. I'll read them often to find
postings from people that offer contrart viewpoints. Basically to look for
someone that debunks the obvious drivel usually found posted there. Using that
anti-metadata as a pivoting point out to other like-minded groups and
individuals would probably be a LOT more telling that folks that all chant the
More information about the foaf-dev