[ExI] Survival

[Jef Allbright]

On 1/3/08, Stathis Papaioannou <stathisp at gmail.com> wrote:
> On 03/01/2008, Anne Corwin <sparkle_robot at yahoo.com> wrote:
> > For some reason I'm finding this ice-cream speculation terribly amusing.
> >
> > It sounds silly on the surface, but there's a lot of potential for
> > interesting discussion there.  I feel like I *know* I project a "chocolate"
> > signal, but I have no idea *why*.
>
> I would have guessed chocolate, but I'm also not sure exactly why;
> perhaps something to do with quirkiness, vanilla being in general the
> least quirky choice:
>
> http://www.sendicecream.com/15mospopicec.html
>
> (It might be possible to break this down further by geographical
> location, gender, age etc. if you had access to all the statistics.)
>
> I suspect that Eliezer and Jef guessed vanilla for me because in the
> absence of other evidence, it has the greatest prior probability (at
> least, I prefer that explanation rather than that of my evident
> non-quirckiness).

On the contrary, my probabilistic assessment (with quite high
certainty as I expressed to you) was based on a broad variety of
evidence based on my observations over many months of what discussion
lists you participate in, how you respond to novelty and conflict,
your favorite movie, how you tend to resolve subjective/objective
distinctions, your choice of words, concepts and style of thinking and
much more.  While I've seen no explicit expression of your preferences
in ice cream, all this evidence (in the context of my prior
understanding of the world) formed a fairly substantial (while fuzzy)
model of a coherent whole, with a place in it for your most likely
preference (between chocolate and vanilla) being quite clearly
vanilla.

For me the same process applies to assessing the "realism" or "truth"
of a statement.  Almost immediately I get a geometric image in my
mind, and I pay attention to the slope, linearity, monotonicity, etc.,
providing an immediate intuitive indication of the mapping of the
assertion onto my own model of reality.  I can then focus my attention
on the higher-information content portions of the geometry, asking
myself what (in my model or the other's) might account for the
deviations.

Weird?  I suppose so, but it works for me, and one of the reasons I
hang out in places like this is that **once in while** I connect with
someone who understands some of it in some of the same ways.

- Jef