Re: Lee Corbin's Goodbye

From: Mark Walker (mark@permanentend.org)
Date: Mon Aug 18 2003 - 12:21:29 MDT

  • Next message: asa@nada.kth.se: "Topology optimization: steps towards automated engineering"

    ----- Original Message -----
    From: "Charles Hixson"

    > We are all still humans. Espousing TransHumanism doesn't make one a
    > trans-human. So don't be surprised that normal human politics it the
    > actual reaction pattern even among those who promote other patterns as
    > preferable. And when the rules are vague and subjective, then those
    > making the decisions will be greatly influenced by their own emotions.
    > Will be, not may be. In that situation there *is* not alternative.
    >

    I certainly have sympathy with this line of response: there will not be any
    significant improvement until we change our natures. However, transhumanists
    sometimes suggest that even in the absence of changing our nature the
    application of science and technology to our (human) lives will emcipate us
    from at least some extant social ills. I guess to such transhumanists you
    would say: 'not likely'. In any event, what I suggest is an experiment: I
    take it you will bet on the null hypothesis here: the application of
    technology and information about our social natures will do little to change
    the normal course of human politics. You may be right, running the
    experiment would be one way to decide the issue.

    > I don't trust any centralized authority run by people, because over and
    > over again people have proven untrustworthy. But I do trust traffic
    > signals. They only play the favorites that they are designed to play
    > (e.g., emergency vehicles, or timed signals).
    >
    > OTOH, I don't trust voting machines. Consider the reasons:
    > 1) how they work is hidden.
    > 2) there have already been instances where they were appearantly
    corrupted.
    > 3) reputable authorities have stated publically that they don't have
    > good safeguards against being corrupted.
    > 4) Influencing the vote so that your client wins has a tremendously high
    > payoff.
    > But do you really think that the prior elections were without
    > corruption? Even after the last one, where several different modes of
    > corrupting the vote became public knowledge?
    >
    > Centralized authority is an inherrent weakness when the authority can
    > benefit from misusing it's power. It's an evolutionary process, and if
    > you don't consider the likely mutations, you can't predict the end
    > result, but it won't be what you wanted.
    >

    I'm not sure I understand the relevance of the voting discussion. I'm not
    suggesting that we vote people off the list, I'm suggesting that people
    might change their behavior in response to the signals they get about the
    value of their posts. Think of it as somewhat like a market. We are
    producers of messages and we want as many people to "buy" our messages as
    possible. If I send out messages that have a high perceived noise to signal
    ratio many people will put me in their killfile, so I will have fewer
    "buyers". I agree about the perennial dangers of centralized authority, and
    while what I am suggesting is not immune from distortion, I also think the
    dangers of corruption are fairly minimal. The public display of the
    killfiles might look like a table with everyone's name on this list (there
    is what about a 1000 people on this list?) on the X and Y axis. I could look
    up your name and see if I am in your killfile and you can look up my name
    and see if I am in your killfile. One conjecture then is that if people can
    signal their dislike of someone's posts by using the killfile this may
    reduce the acrimonious exchanges that sometimes go on. If we are having such
    an exchange I might put you in my killfile and you might retaliate by
    putting me in your killfile--and so would end the exchange. Next to each
    person's name would be the number of people that have entered him or her
    into their killfiles. The conjecture is that people might change their
    behavior so as not to be entered into too many killfiles. Take an extreme
    example. Suppose an individual posts tons of Nazi propaganda to the list
    every day. I conjecture that almost all will enter this person into their
    killfiles, hence the Nazi's voice will go unheard. This Nazi cannot complain
    about a centralized authority squashing her voice since the killfiles are
    set by individuals. As I said, the right to free speech does not entail an
    obligation on others to listen. The reason this system would be relatively
    impervious to corruption is that each person's killfile would be displayed.
    If there was an attempt to discredit someone by artificially raising their
    killfile quotient this could be easily detected because each killfile is
    assignable to a specific individual. If the list managers tried to
    artificially raise your killfile quotient they would have to assign it to
    some individuals, say one of them is me. When I look and see that my
    preferences have been tampered with I will scream bloody blue murder on and
    off the list.

    Cheers,

    Mark

    Mark Walker, PhD
    Research Associate, Philosophy, Trinity College
    University of Toronto
    Room 214 Gerald Larkin Building
    15 Devonshire Place
    Toronto
    M5S 1H8
    www.permanentend.org



    This archive was generated by hypermail 2.1.5 : Mon Aug 18 2003 - 12:33:43 MDT