[LINK] astroturfing

jim birch planetjim at gmail.com
Tue Jan 10 11:09:44 AEDT 2012


The basic problem is the internet's anonymity or absence of reliable
identity, and hence, no reliable reputation history.  I can't see this
continuing indefinitely.  Interestingly, Google + requires a real name with
id's belonging to events, etc, that facebook allow becoming places.  I'm
not sure how far they go down the validation path but I believe they will
ditch your account if they find it a false name.

Personally (without having considered this to much) I think I'd be happy to
only ever receive communications from validated identities with a good
reputation, or have communications flagged with the sender reputation,
good, bad or unknown.  I think it would be possible to design a federated
system that supports this kind of approach.  This wouldn't fully prevent
scams but would make life more difficult for serial offenders.

There's a fair bit of theory (ie, real mathematical results) that suggests
that a reputation mechanism would be the way forward for a huge system like
the Internet.  See, for example, Martin Nowak's book Super Cooperators.

Jim



More information about the Link mailing list