FOAF-trolls

[fallen_x_ashes]

This entire concept of FOAF seems very intresting and promising, but I do have to ask, it also is based on the conept that a web users FOAF file would be correct, yes? There's nothing preventing someone from creating a FOAF file that isn't a real reflection of themselves

Comments

[denshi]

There's that tricky 'correct' epistimology problem as well...

[acheron_hades]

How about just adding in a PGP-style web of trust? I haven't thought the details through but I imagine it's do-able..

[revgeorge]

There is a web of trust FOAF project, which basically relies on a detached GnuPG signature. Then you can verify the authorship of FOAF files using GnuPG.

[hohf]

Yes, this will *definitely* be a problem, even with GPG, and especially
when considering that the broad masses aren't really using GPG.

FOAF is nice, but i fear it'll have the same problems soon which E-Mail
has now. (Maybe that's a RDF problem, though.) If you want to know my
opition, all those semantic web things shouldn't be technically usable at
all without a proper GPG key.

-hohf

faof trust

[dvae]

Why is it a problem. As long as people only list true friends as friends in thier foaf files then it's ok. If you don't trust someone, don't list them as a friend. Simple