[foaf-dev] FOAF-based whitelisting project

[Dan Brickley]

R. Steven Rainwater wrote:
> On Sun, 2007-03-11 at 00:22 +0100, Kjetil Kjernsmo wrote:
>> using FOAF-based trust networks for whitelisting email. 
>> <...>
>> Importantly, I want to build it using the most basic data,
>> really identifying people, context-dependant trust
> 
> You might want to take a look at mod_virgule, the software used to
> implement www.advogato.org and robots.net, and a few other community
> sites. It includes a trust metric system and trusted members of the
> community now have FOAF files as part of their profile.
> 
> If nothing else, I'd be happy to modify mod_virgule to submit trusted
> user's FOAF data to whatever sort of repository your anti-spam system
> will use. Or is it more of a distributed system where you'll just direct
> spamassassin to the source of the FOAF file? 

Would definitely be cool to see more of the trust-metric info within 
advogato flow out in machine-readable form.

I did some prototyping in this area a few years ago (urk, many years 
ago!), http://www.w3.org/2001/12/rubyrdf/util/foafwhite/intro.html -- 
not so much the trust metrics side, ... just basic info about who you 
deem a non-spammer.

Kjetil, did you see 
http://www.w3.org/2001/12/rubyrdf/util/foafwhite/intro.html ?

Re mod_virgule, ... I guess a lot can be done simply with existing info, 
since we don't need to know how *great*, *wise* etc someone is, ... 
merely that they are not a spammer. For 99.x% of people, ... being a 
foaf:knows of someone else in the network is enough. But I guess 
spammers do have friends. The trust metric stuff would still be great to 
have, ... to help folk prioritise information, etc.

Dan