Reputations

[Tim May]

At 11:35 PM -0700 4/15/01, Alan Olsen wrote:
>On Sun, 15 Apr 2001, Bill Stewart wrote:
>
>>  Some of the cypherpunks implications were that we all saw
>>  reputation systems as a goal that Really Made Sense,
>>  but also that turn out to be much harder to implement,
>>  even on non-fictional paper, than to describe in fiction.
>>  What kinds of algorithms do you use?  How do people outfox them?
>>  How do you deal with not only the real Detweilers,
>>  but with people using the kinds of pseudonym hacks that
>>  Detweiler was constantly ranting against, such as creating a
>>  bunch of pseudonyms that all give each other positive ratings
>>  and positive reviews of each others' articles,
>>  to create a bunch of reputation capital that's undeserved
>>  and can later be burned if needed.
>
>One of the other problems with reputation capital is that reputation
>depends on perspective.

Interestingly, nearly everything that is "interesting" depends on 
perspective (including what is interesting). Entropy and randomness 
both depend on perspective. The value of some item depends on 
perspective: what is of high value to Alice may be of little value to 
Bob.

This is part of a much larger issue, and is not just verbal game-playing.

Perhaps I was so accustomed to this "relativist" point of view that 
the fact that reputations share the same characteristics was not at 
all surprising.



>
>The people who I respect and listen to are not always the ones that you
>will repect and listen to. reputation is a more individual thing. I think
>if you mapped who people found worthy of reputation that it would break up
>into a number of different groupings.

I did a major article on this some years back, about probabalistic 
belief networks...which is what reputations seem to best map into. 
The Dempster-Shafer method of propagating beliefs seems most useful.

--Tim May



-- 
Timothy C. May         tcmay at got.net        Corralitos, California
Political: Co-founder Cypherpunks/crypto anarchy/Cyphernomicon
Technical: physics/soft errors/Smalltalk/Squeak/agents/games/Go
Personal: b.1951/UCSB/Intel '74-'86/retired/investor/motorcycles/guns