[OpenID] reputation

Fen Labalme fen at 2idi.com
Sun Feb 25 19:38:48 UTC 2007


(catching up on old emails) I agree with *both* Bob and Joaquin, and further
agree that "reputation" is a tricky word to define.  The problem has to do
with the fact that reputation is not only specific to a person/entity (the
Reference) and a context, but also according to who is making the claim and
when (as people's opinions can change).  While working on OpenPrivacy in 2000,
I assembled this definition:

Reputation: A value that represents the collective opinion of some reference.
A reputation is really just another name for an Opinion, as it is the
calculated opinion of a Reference by the issuing Reputation Calculation
Engine. Reputations are ephemeral, and the weight applied to an Opinion
representing the reputation of some Reference is subjectively applied by the
end user (person or program) that requests it.

(Other terms defined also on <http://www.openprivacy.org/opd.shtml>)

I think in today's lingo, "Reference" would be "Subject" and there are
probably quite a few other changes I can make to update those definitions (and
I am welcome to suggestions!).

On a related note, I just blogged about the importance of reputation here:
http://blog.fen.net/

=Fen




Joaquin Miller wrote:
> Bob is right on target about the word, 'reputation'.
> 
> Whether data about a party or claimed to be about a party is true or not
> is irrelevant to reputation. 
> Reputation is about the opinion of some people about some other party. 
> The only fact that has anything to do with reputation is how a party is
> regarded.
> 
> I offer authorities, a demonstration, and an example; see below.
> 
> Please, let's not destroy another perfectly good word by giving it a
> technical meaning that is not at all its ordinary meaning.
> 
> Cordially, Joaquin
> 
> 
> Authority:
> 
> OED: "...estimate of a person with respect to character or other
> qualities; the relative estimation or esteem in which a person or thing
> is held..."
>  (Or, for those of us who are net-type-people:)
> Wikipedia: "...a social evaluation..."
> Wiktionary: "what somebody is known for"
> 
> 
> Demonstration:
> 
> If Joe claims to be a doctor and is a doctor and is regarded by people
> as doctor, then (in that respect) Joe has a good reputation.
> If Joe claims to be a doctor and is a doctor and is regarded by people
> as no doctor at all, then (in that respect) Joe has a bad reputation.
> If Joe claims to be a doctor and is not a doctor and is regarded by
> people as a doctor, then (in that respect) Joe has a good reputation.
> If Joe claims to be a doctor and is not a doctor and is regarded by
> people as no doctor at all, then (in that respect) Joe has a bad reputation.
> 
> 
> Example:
> 
> Does it seem like there is some puzzle here?  There is not. 
> 
> If a person keeps leaving blog spam, it is a fact that that person is a
> spammer.  So far we know nothing about that person's reputation. 
> 
> If a site reports that that person keeps leaving blog spam, then that
> person has a bad reputation (with that site).  If a reputation
> aggregator reports that many sites report that that person keeps leaving
> blog spam, then that person has a bad reputation (with those sites).  If
> you and I and the rest of us on this list think that that person keeps
> leaving blog spam, then that person has a bad reputation (with us). 
> 
> The fact that that person is or is not a spammer has nothing to do with
> the reputation: Quite possibly that first site is lying.  Quite possibly
> all those sites are mixing up that person with some other person.  Even,
> quite possibly, we just don't like that person and the messages they
> post, although those messages are in fact not actually spam, but we
> can't see that because we dislike that person so much.
> 
> 
> Bob Wyman wrote:
>> I would counsel folk to be careful about bundling too much into the
>> bucket of things that we think of as "reputation information." ... I
>> believe that "reputation" should really be restricted to subjective
>> statements about "behavior" or perceptions. 
> 
> 
> ------------------------------------------------------------------------
> 
> _______________________________________________
> general mailing list
> general at openid.net
> http://openid.net/mailman/listinfo/general

-- 
http://xri.net/=Fen.Labalme



More information about the general mailing list