X-Message-Number: 30298
Date: Fri, 11 Jan 2008 21:16:28 -0700
Subject: Only one claim of "Abuse" (was Re: [CN] How to make Cryonet b...
References:  <>
From:  (Tim Freeman)

From: david pizer <>
>I have been told by reliable sources that an Alcor
>director told some Alcor staff, advisers and directors
>to give my Cryonet messages the lowest possible rating
>so as to get me censored.

I just now looked at a bunch of Pizer's recent messages, and a few
days ago I looked at a bunch more.  Most of them had one "Abuse" vote
and none of them had more than one "Abuse" vote.  Presumably the
director in question has taken his own advice, so the interesting
observation is that nobody else took the suggestion.

For what it's worth, the rule of thumb advice from "Sociopath Next
Door" is that if someone lies to you on three separate occasions, you
should assume they are a sociopath and avoid them.  So, if we assume
Pizer's inside information is accurate, there's one Alcor director who
deserves one strike out of three because they're lying to everybody by
giving false inputs into the cryonet rating system.  Given the
anonymity going on here we can't tell who.

Maybe cryonet ratings should have the rater's email associated?

Until now the ratings have been anonymous.  However, I read through
the description of the ratings algorithm and didn't notice any
documentation that promises that.  Kevin has debug access to his
system and can find out who, if he wants to.  I wouldn't blame him for
refusing.

I agree with Stodolsky that we need a system that has a firmer
theoretical basis.  In general, the group can separate into cliques,
where clique A trusts only people in clique A and clique B trusts only
people in clique B.  Are there known good algorithms for doing this?

-- 
Tim Freeman               http://www.fungible.com           

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=30298