Re: Science of choices falls out of research

In article <1146761711.663482.59590@xxxxxxxxxxxxxxxxxxxxxxxxxxxx>,
"nando_ronteltap@xxxxxxxxx" <nando_ronteltap@xxxxxxxxx> wrote:

stew dean wrote:

Andre actualy made a very important statement which you've failed to
understand.

Is rubbish, I've shown to understand the statements in repeating them
in my own words. The statements are simply untrue.

Probabilities are our way of measuring things. A weather
system doesnt use probability, it uses causality.

No causality is described as equations such as F=ma. Causality does not
cover any probabilities, probabilities are described in terms of one to
many, that is; one decision several possible outcomes.

The problem is that you refuse to acknowledge that you are either using
the term probability to refer to something different from what the rest
of the world does, or you are infusing it with properties which are not
part of the proper definition of probability. You haven't shown anything
to suggest that my statements have been untrue. You have asserted this
on numerous occasions, but an assertion is very different from a
demonstration. The possibilities which you refer to above do not
necessarily have any real existence and stem from a given model's
inability to make complete predictions about future observations.

Consider a situation in which someone throws a ball at a target. For
sake of argument, let's assume that this experiment is taking place in a
vacuum where atmospheric conditions play no role. In this case, we're
dealing with a simple case of projectile motion which means that the
trajectory of the ball is fully determined by the laws of physics the
instant it leaves the persons hand.

In principle, this means that if we know the speed and direction of the
ball when it leaves the persons hand, we also know whether it will
strike the target or not. In practice, however, we may not be able to
exactly measure these quantities which means there will be some
uncertainty involved, and this is where probabilities come into play.

A scientist measuring the speed and direction of the ball when it leaves
the persons hand will have to take the error in his measurements into
account and to do so will express whether the ball hits the target as a
probability rather than as a certainty or an impossibility. If the same
scientist measures the velocity of the ball at a different point in its
trajectory, one which is closer to the target, he will arrive at a
different probability because the errors produced by his measuring
device will become less significant as the ball nears its target. If a
second scientist makes the same measurements using a more accurate
measuring device, he will also arrive at different probability values
because there is less uncertainty in his measurements. None of these
probabilities have anything to do with whether the ball hits its target
or not. They are simply expressions of uncertainty and may change over
time or from observer to observer. Nothing in this implies a decision
being made, and in no sense does taking a second measure involve
'rewinding' the event as you keep bringing up.

Now, if we take this entire experiment and put it back on earth where
there is an atmosphere to contend with, things become more complicated.
Now we no longer have to just consider measurement errors; we also have
to consider factors like air-resistance, unexpected changes in wind
speed or direction, etc. Our current models are such that even if we
knew with absolute precision the initial speed and direction of the
ball, we would not be able to fully predict where the ball would end up.
Thus, we are forced to express the answer probabilistically.

Note, however, that our current inability to fully predict the outcome
says nothing whatsoever about whether the outcome is, in fact,
determined by the initial conditions. It may well be the case that given
a more sophisticated model of how large bodies of unevenly heated gas
(the atmosphere, in this case) behave, such a determination might be
possible.

It's also quite possible that we will never be able to make such a
determination. There are various possible reasons for this: it might be
the case that the precision needed with respect to this initial
variables might be in principle unattainable; it might be the case that
the mathematics involved is too complicated to be tractable (or even
computable); or it may be the case that the system in question is truly
non-deterministic in the sense that the outcome is not a function of the
initial state (and the same options exist with respect to everything
which we cannot currently predict, ranging from the weather to the
behaviour of subatomic particles to human behaviour).

What's important to realise here, though, is that the use of
probabilities does not imply any of these three options. Probabilities
simply express uncertainty in our ability to make predictions within a
given model regardless of whether that uncertainty stems from imprecise
measurement or an incomplete model. To assume that there is some actual,
reified 'probability' associated with an object is an abuse of
terminology, and trying to claim that this 'probability' has an actual
locus is even more problematic.

You, of course, are welcome to claim that such a thing exists, but if
you do (i) you will have to actually provide some evidence for the
existence of said thing, and (ii) you should call it something other
than probability since that term already has a well-defined meaning
which is different from what you are using.

With respect to your electron example, I have not addressed this simply
because your claim is incoherent. You keep talking about the 'zero
point' in the probability distribution of the electron, but you haven't
explained what you mean by this (if you simply mean the point at which
the probability is zero then you are introducing something which isn't
present in physics) or shown that such a point actually exists.

The position of an electron is one area where we are forced to use
probabilities and where, at least on some interpretations of QM, this
stems from the actual non-deterministic nature of the phenomenon itself.
Note, however, that the claim that this is truly non-deterministic is
not something which can be empirically demonstrated. The Bell
Inequalities suggest, at least as to the extent that I understand the
EPR paradox, that true indeterminism may be involved, but there are
other options. It's possible that information can be propogated at an
unbounded velocity (something which most find unsettling, though
ultimately the universe will work however it works, regardless of our
opinion of it) or that our understanding of QM is fundamentally flawed
(which seems unlikely). Importantly, though, the use of probabilities in
QM is not dependent on any particular interpretation of QM being correct
(whatever that means).

Therefore, if you really want to continue to press your notion that a
decision is somehow tied to a change in probability, I suggest you start
by changing, and more importantly, *defining* your terminology. You will
never convince anyone of anything if you insist on using established
terms in nonstandard ways. Once you've done this, you need to actually
present some evidence supporting the fact that things actually work the
way in which you claim they do, which to date you have failed to do.

Andre

--
Andre G. Isaak
n.b. there are no monotremes in my email address

.

Relevant Pages

  • Re: Lottery facts for Sherry
    ... Totally incorrect. ... The probability that any ball will be picked first is 1/49. ... buying one ticket in a 6/49 environment gives you one ...
    (rec.gambling.lottery)
  • Re: infinity
    ... and a leak in the dam with 1 gallon per second ... Assume that the water in the lake is ... > pong balls where the ball to be removed is picked at random. ... So what is the probability that ball 1 gets removed? ...
    (sci.math)
  • Re: randomness
    ... nonzero probability that it's not even there. ... ball impacts 3 ball, and 3 ball drops into the corner pocket. ... probability that the 9ball doess not even exist. ... support for spacetime from a general construction ...
    (sci.math)
  • Re: 0, 1, 2, 3...
    ... expect Dravid to score century and not Umesh Yadav. ... whatever instead of getting out in next ball. ... The probability of Warner getting out at 48 when on 48 was ... If he'd got to 49 his probability of getting out at 49 would ...
    (rec.sport.cricket)
  • Re: Which is rarer?
    ... Comparing actual events to get the probability ... Waqar Younis takes a wicket every 30 balls ... Therefore Hitting 6 sixers in 6 balls is about 1600 times more ... Probability would only really be applicable if each ball was identical. ...
    (rec.sport.cricket)