Re: Next stage, and Q1

At 10:48 PM 24/06/02, Gyuri wrote:
>At 16:36 20/06/02 +0200, you wrote:
>>Now, if self-organization is taken seriously, rather than merely as some
>>vaguish metaphor, there must be something that is dissipated to create
>>and maintain information. Furthermore, not just any dissipation will do;
>>it will have to be dissipation of something of the right sort to permit
>>the relevant sort of information to form.. So, rather than give an
>>answer, I ask the following questions:
>>
>>What is dissipated in ....
>
>
>Dear John, Pedro and FISers,
>
>I agree with your thoughts as you reached to put this question.
>I am not sure, I can give a definite answer. I try only to share my thoughts.
>A closed system differs from an open one, that there is no exchange of
>anything on its boundaries. This means, a closed system does not change
>information with its environment either. As I see now (but maybe I am
>wrong), as soon as we allow any dissipation (of whatever property) we can
>speak about information transmission too. Maybe the essence of providing
>information would be in this fact (dissipation of something), and not in
>the specification what is dissipated? (Of course, this does not mean, that
>any property plays this role, I think only that many properties may do.)
>What do you think?

Dear Gyuri,

That is an interesting observation about open systems. Personally, I am
inclined to agree with your suggestion, but others who have a more
restricted notion of information will not be so ready. In any case, I think
that Pedro's observation that there are different kinds of information,
like different kinds of forces (the analogy is weak, I think, and it would
be nice to find a stronger one), suggests that there may still be a problem
with understanding how different kinds of information might be convertible.
Even in an open system in which we can talk of information flows in and
out, any internally emergent information may be of another kind than the
flows. This seems to be required for the emergence of new kinds of
information (e.g., codes, intentional information). So I think Pedro's
problem still stands. On the other hand, your suggestion does focus on the
issue of what conditions must hold for information of one type (or some
sort of proto-information) to dynamically emerge as another kind of
information.

I must confess that I have not taken this problem very seriously so far,
since I have been working with what I understand to be a unitary notion of
information that is close to, but not the same as the "it from bit" view.
(I restrict "its" to macrostates, basically, adopting the negentropy
principle of information, so I would not accept what David Layzer calls
"microinformation" in his 1975 Scientific American article, Layzer, David.
1975. The arrow of time. Scientific American 233: 56-69 as information.) In
this case, any macroscopic order that emerges would be information. So I
have concentrated on 1) how previous information systems can "recruit"
potential information, and 2) how new kinds of information can be formed by
special conditions under which self-organization occurs. The second problem
is really just another version of the conversion problem, so it can't be
avoided, I think, but perhaps the best view is one that mitigates the
problems in understanding conversion most effectively. Incidentally, this
problem is one that our small group at Fuschl found to be very pressing,
and one that discussions in Vienna with Wolfgang Hoffkirchner and his group
keep coming back to. I don't think that the common issue has anything to do
with my more or less accidental participation in both groups -- it seems to
arise out of the definition of the problem, which was set long before my
participation. So I expect that other people and groups have come up
against the same issue in one form or another, which might be called "the
conversion problem".

The conversion problem may have no single solution for all kinds of
information and proto-information. On the other hand, I think that some
sort of solution is required for a unified theory of information.

John

----------
Dr John Collier john.collier@kla.univie.ac.at
Konrad Lorenz Institute for Evolution and Cognition Research
Adolf Lorenz Gasse 2 +432-242-32390-19
A-3422 Altenberg Austria Fax: 242-32390-4
http://www.kli.ac.at/research.html?personal/collier
Received on Tue Jun 25 10:52:43 2002