X-Message-Number: 9530
From:  (John P. Pietrzak)
Newsgroups: sci.cryonics
Subject: Re: Pietrzak 5
Date: Tue, 21 Apr 1998 01:25:46 GMT
Message-ID: <>
References: <>

Finally, we seem to be getting down to the heart of the matter.  Let
me deal first with the points Mr. Ettinger makes which seem to apply
somewhat to the two questions I have posed to him, and consider
separately the other side issues he brings up.

(To summarize, my questions are (1) what (if any) objective data has
he used, and (2) can he refute that his probabilistic study results in
a circular calculation.)

On 20 Apr 1998 16:20:04 GMT,  (Ettinger) wrote:

>2. Bayes does NOT take a "step away" from the "objective ideal." This is a
>fundamental misunderstanding. Bayes is MORE objective (than the Fisher
>approach, say), in that it does not ignore available data.

In part of my studies in AI, I worked with a program called Minerva (a
derivative of the classic MYCIN program), where the a priori factor
included in the system of Bayesian probability it used was the obvious
place to implement its heuristics, allowing human-quality answers to
questions within its domain of knowledge.  In other words, it was
exactly because Bayes allowed SUBJECTIVE data to be added in
calculating probabilities that it was useful for this application.
(Bayes places no constraints on how you use that a priori factor!)

>My technical example (exponential life parameter) applies here. Read
>it again.

It does.  I did.  It allows for the use of subjective data.  What's
your point?

If possible, let's switch this discussion to the objectivity of your
data.  (Just because a function is set up to accept subjective data
does not mean that purely subjective data has been used.)

>3. It is NOT "circular" to say, first, that many difficult projects of
>serious technology have succeeded, while few or none have been proven
>impossible; and then to conclude that the cryo-repair project has a good
>chance of success.

Ah, now this _does_ more closely approach my second question.  Of
course, you're just appealing to modus ponens here, not your 16 page
treatise on probability:

1) Goals of technology have good chances of success
2) Cryo-repair is a goal of technology
---------------------------------------------------
3) Cryo-repair has a good chance of success  (1+2)

This is a much different argument, and I certainly have a lot more
faith in its methodology. :)  All we need to do here is substantiate
axioms #1 and #2.  (But I suspect you'll find it hard to get good
objective data on axiom #1.)

(This still doesn't really answer my second question, though, that
your probability theory is, as applied in the paper, a simple identity
function over the input it is given -- when used as you use it in the
paper.)

>4. "it is impossible for you to end up with a probability value that was
>not already implicit in your intuitions. CIRCULAR."  There are two
>possibilities. First, I use only my "intuition" (or rough guess based on
>experience); in that case my guess was either near the mark or it was not.
>Certainly there is a possibility of error. Second, I start with intuition
>or guesswork and then decide to make it explicit by compiling statistics.
>In that case, my result will be more reliable. But in neither case was
>there any circularity.

Circularity has NOTHING to do with accuracy.  When I say an argument
is circular, what I mean is that the inputs to the argument are also
it's outputs (regardless of what those inputs are).  Thus, "The sky is
blue because the sky is blue," or "My name's John because John is my
name."  Or, in your case, odds-on for the chances of success in
cryonics because you already believe that the chances are odds-on.

>I started by looking at cases that were similar but different, cases
>already on the record. Read it again.

Uhm, ok.  The only thing that I can think applies here would be the
section on "Intuition and Probability" (pp. 17-18).  So, that would
mean your similar cases would be how adults intuitively avoid cars
when crossing the street, the intuition of Goddard and Tsiolkovsky on
the feasability of moon rockets (which, depending on how you read it,
applies to either scientific or political feasability), and da Vinci's
intuition about flying machines.

In other words, your similar cases are there to support the likelihood
that your own intuitions (on cryonics, or for that matter on
practically anything else in the wide world!) are reliable.

My question here is, why does this support your claim?  I would
suppose that I avoid cars in much the same way as you, and I doubt
that either of us are closely related to either Goddard, Tsiolkovsky,
or da Vinci (I know I'm not), so presumably my own intuitions on
cryonics would be just as reliable as yours.  To be perfectly honest,
none of my intuitions give me odds-on results for any of the three
points you bring up on page 20.  What, then, supports your intuitions
over mine?

-----------

It would appear, after going through your three points above, that
both my questions are still mainly unanswered.  Could you venture to
address them directly?  Thank you.

-----------

On to the more minor issues:
>1. I have not endorsed Koopman--only pointed out that, in some ways, his
>approach is in practice not much different than von Mises'. Read it again.

p. 5: "B.O. Koopman [...] has produced a detailed axiomatic theory,
based on intuition, which seems to be in many respects as practical as
v. Mises' frequency theory and yet broader."

p. 6: "In short, it seems consistent with modern ideas to assert, in
flat contradiction to Koopman, that intuition derives from experience,
and can in principle be explicitly analyzed in terms of experience.
In fact, I claim below to show how probability intuitions arise."  (I
read this as being in support of Koopman's system, at least in
general.)

>5. I don't think Mr. Pietrzak has yet acknowledged that his Michelson-
>Morley counterexample was wrong.

I haven't, because it isn't.

>The M-M experiment was not a failed project of technology.

You're certainly right there.  The "interferometer" was a wonderful
_technological_ success, and is still being used all over the place
today.  The base for it that Morley dreamed up was also a
technological marvel (chemically, if not physically).  However, you
were talking about goals of _science_, not technology, when I brought
it up.

>M-M did exactly what it was designed to do--test the hypothesis of the
>"luminiferous ether."

Michelson didn't want to test whether the ether existed -- he was
already sure (intuitively!) that it did.  He was just trying to find
it.  His failure to find it perturbed him (to say the least).

>The fact that the result may have been displeasing to the experimenters
>is irrelevant.

Ah, and this is certainly a reassuring note coming from you, in that
your own test involves people's lives.  I for one would hope that
_your_ measure of success in this calculation involves a positive
result from your experiment, not just the existence of _any_ result.

>6. I don't think he has acknowledged either that his Fountain of Youth
>counterexample was inapplicable.

What is this??  I'm pretty sure I've already said that I was not the
one who brought this up, and in any case it was never mentioned as a
goal of either science or technology.  Rafi Haftka mentioned the
"Fountain of Youth" in a letter pertaining to the foundational
reasoning behind your system, and I replied to him.  You can find our
discussion in cryonet messages #9492, #9501, and #9510.  If I may make
a suggestion, "read it again."


John

Rate This Message: http://www.cryonet.org/cgi-bin/rate.cgi?msg=9530