[foaf-dev] FOAF-based whitelisting project

Dave Brondsema dave at brondsema.net
Mon Mar 12 19:27:29 UTC 2007


Tom Heath wrote:
> Hi Elijah, Hi all,
> =

>> A unidimensional trust metric is often not particularly useful.
> =

> I'd agree that your statement holds true in most cases, however I
> think there's a risk of overcomplicating the issue in the context of
> this project. What's needed in this case (imho at least) is simply
> some measure of: the extent to which I trust that the sender of a
> particular email is a real person who wants to contact me for some
> legitimate reason to which I'm likely to consent. More contextualised
> trust is great/essential in other settings (I may trust you to repay a
> loan, but not to recommend restaurants, or vice versa), but would be
> overkill here. Knowing that a mail comes from a friend of a friend (of
> a friend...) is probably a sufficient basis for this form of trust.
> =

> On that basis I think a foaf:knows-based approach is good enough (at
> least to start with). The question then becomes "who do you trust to
> send you email", where "who" is defined in terms of number of hops in
> a network of foaf:knows relations, starting at 1 hop with the people
> you say in your/a FOAF file that you know. Earlier in this thread
> someone argued that overloading foaf:knows for this purpose was not
> valid; I disagree, all this approach would be doing is taking the
> fairly loose semantics of foaf:knows and using them to derive some
> metrics, the results of which may then be republished as
> ex:trustsAsMailSenderToWhateverDegree statements (or something
> similar).
> =

> I quite like the idea of a 0-1 score (rather than 0-100), and in
> common with Steve's mail that's just arrived, I'm also skeptical about
> negative trust ratings; what would we *actually* do with them?. Surely
> all we need is a score somewhere between 0 and 1, or NULL where we
> don't have any data?

Yes, but how do you relate a 0-1 score with a foaf:knows that has no
value in it?  Do we have some setting that equate foaf:knows to 0.8
trust and foaf:interest to 0.4 and foaf:spouseOf to 0.9 (as examples)?
And then would the setting be system-wide, or per user that is issuing
the trust query, or does the truster/knower in the relationship set what
he wants his foaf:knows to mean, or can the truster/knower specify
specific values for each relationship if he wants to be detailed and
precise?

> =

> Lastly, overcoming the risk of spam foaf:knows statements (SpOAF?)
> should simply be a case of combining a user preference
> "TrustSendersUpToHops: X" (or "TrustSendersWithTrustScoreGreaterThan:
> X) with a heuristic that only attends to foaf:knows relations that run
> from the email recipient outwards, but not the reverse. So if I get
> email from someone who someone I know says they know, that's fine (if
> hops threshold > 2), but if email arrives from someone who says they
> know someone who knows me, but my FOAF doesn't corroborate this, then
> this mail should be rejected (however 'true' the statements are).
> Perhaps this directionality should also be a user option.
> =

> Thoughts? Cheers,
> =

> Tom.
> =

> On 12/03/07, elw at stderr.org <elw at stderr.org> wrote:
>>
>>
>> > My idea is that we use a single trust metric for everything, I would
>> > suggest an integer in the interval [-100,100], -100 indicating strong
>> > distrust, 100 indicating full trust. That metric could be returned from
>> > either a web service or a locally hooked up system, but I think that it
>>
>> A unidimensional trust metric is often not particularly useful.  [This
>> was
>> something I struggled with mightily in MA thesis work, some five years
>> ago, looking at how then-current moderation systems were using trust
>> metrics.]
>>
>> Better would be to define terms that allow you to specify terms and
>> degrees of trust, as relates to particular domains of discourse or
>> domains
>> of agreement, such that I can trust your opinions or attitudes in domain
>> #1 but choose to think you're full of hoo-ah in domains #2, #3, #4.  I
>> think this is a better fit with FOAF/RDF/SW technologies in general,
>> though it does demand a little more definition of vocabulary terms....
> ...
> _______________________________________________
> foaf-dev mailing list
> foaf-dev at lists.foaf-project.org
> http://lists.foaf-project.org/mailman/listinfo/foaf-dev
> =



-- =

Dave Brondsema
Software Developer
Cornerstone University

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 888 bytes
Desc: OpenPGP digital signature
Url : http://lists.usefulinc.com/pipermail/foaf-dev/attachments/20070312/d1=
2ccc04/signature.pgp


More information about the foaf-dev mailing list