[CrackMonkey] ...from the corporate IRC

[Mr.Bad]

>>>>> "B" == Bernard  <nutella at zork.net> writes:

    B> Hmmmmm... No, I'd say you're mixing up a model with an analogy.

I was, but I'm prepared to defend that.

    B> By definition an analogy takes its cue from something else, in
    B> this case some situation which is more familiar, but a model
    B> for data could be created (and be testable) without any
    B> *obvious* prior art.

I'd disagree. I don't think we can have conceptual models that are
built without explicit analogy or some kind of implicit analogy, even
if that analogy is one of mathematical formulae.

    B> With the genome example I am happily familiar with
    B> the concepts of genome, chromosome, locus, gene, whatever
    B> and these are the most closely tied to the experimental data,

Yes, but your sophisticated model wasn't fabricated from whole cloth.
I would wager that your understanding of these concepts started with
simple analogies, such as Bb-BB-bb, and became more complicated as you
became more familiar with the field. At each step you transformed your
original model based on understandable transforms (more letters, more
kinds of letters, a spiral staircase with bends and twists in it, a
broken staircase, etc.) that were themselves analogies.

Even if the eventual model is shorthanded to great abstraction, it's
still step-by-step based on analogy.

OK, that's my spiel.

~Mr. Bad

-- 
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 /\____/\   Mr. Bad <mr.bad at pigdog.org>
 \      /   Pigdog Journal | http://pigdog.org/ | *Stay*Real*Bad*
 |  (X \x)   
 (    ((**) "If it's not bad, don't do it.
  \  <vvv>   If it's not crazy, don't say it." - Ben Franklin
 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~