Re: psychohistory

From: Scott Chase (
Date: Tue 27 Jul 2004 - 15:47:21 GMT

  • Next message: Gene Doty: "Re: testing 1,2,3.."

    --- Alan Patrick <> wrote:
    > --- Scott Chase <> wrote: >
    > > I'll have to look into it, yet somewhere in the
    > > large
    > > part of my post you snipped I made allowances for
    > > modest, short-term predictions. Now how well does
    > > this
    > > stuff hold up in the case of an unforeseen Mule?
    > If
    > > we
    > > were to use chaos maths, evolutionary algorithms
    > and
    > > big computers with CRM systems loaded in, could we
    > > go
    > > back in time, say to when there were dinosaurs and
    > > *predicted* the actual impact and subsequent
    > effects
    > > of an asteroid?
    > I snip for brevity, I hate long posts....anyway, I
    > thought it was implicit in your discussion that the
    > "Mule" - aka a major discontinuity like an asteroid
    > etc - knocks any form of predictive maths off its
    > perch. The ongoing series is how the Foundation
    > tries
    > to rescue the situation....but then I'd give the
    > plot
    > away if I went farther.
    Yeah, I guess I'll need to see how the plot develops in a post-Mule galaxy. Yet, the emergence of a random element such as the Mule does throw a monkey wrench into long term forecasting, even though the Foundation could massage events to get things back on track in an after the fact manner. I'm at a part in Second Foundation where after the Mule they realize that things have diverged a tad.

    I've looked into these CRM's (Customer Relationship Management systems) a little. Looking at this site:

    It looks like these systems hold some promise for targeted marketing and understanding customer behavior, but this article points to the events of 9-11 throwing a monkey wrench into the scheme. Sometimes the future differs significantly from the past. So forecasting might be good in the short-range but must be adjusted accordingly when new contexts arise. If fitness landscapes could serve as an adequate analogy, tracking peaks is a means of satisficing, but overcommiting to a peak on the assumption that the present strategy will be globally optimal (OK I'm using a little too much teleological language here to make my point) will be disastrous if the landscape shifts drastically. A peak could become a valley overnight.

    I'm a little cynical about targeted marketing strategies, because I'm aware of them. I think Douglas Rushkoff might have talked a little about some similar stuff in one of his books. When I walk into a supermarket (notice the term super*market*) and they offer me discount cards, is this so they can make up their costs that discounts incur by tracking my shopping behavior and tailoring decisions based on what they learn from me and others? Maybe they are going to sell this information to marketing companies so that demographic info on my area can be monitored and big corporations will know at a glance what people in my area are buying this year and what we bought in previous years. Paranoid delusional? Maybe. Maybe not.

    The typical customer thinks "Oh, that's nice, they gave me a discount card". I'm not sure how many people stop to ponder why they might be offered the card. In a worst case scenario customers are being treated like sheep, bleating away down the aisles (or Skinner rats pushing at levers?). Can't say that that assessment of customer behavior misses the mark too much, but I'm not the genius who decided to put a sign saying "hot beer" on the aisle that marks where the beer that isn't refrigerated is located. A bleating sheep like me tends to cringe when the words hot and beer are so casually associated. I wonder if I'd have a case for emotional distress and be able to win a class action settlement with all the others this sign has disturbed.

    I do vaguely recall another supermarket many years ago having pesticides and baby food on the same aisle, so these supermarket tacticians aren't always mental heavyweights themselves.

    CRM seems to be more webbased where places like Amazon and others can keep track via cookies what potential customers are into.

    I can appreciate that this stuff might work, but aren't we talking about modest short-term predictions based on stable contexts (sans Mules)?

    __________________________________ Do you Yahoo!? Yahoo! Mail - 50x more storage than other providers!

    =============================================================== This was distributed via the memetics list associated with the Journal of Memetics - Evolutionary Models of Information Transmission For information about the journal and the list (e.g. unsubscribing) see:

    This archive was generated by hypermail 2.1.5 : Tue 27 Jul 2004 - 15:51:28 GMT