OPENING SESSION

FISers--

I announce herein the start of the final session. I propose that this
one be somewhat different in nature.

It is our final session, and it would make the year's effort more
worthwhile if we could produce some conclusions. I propose that
instead of seeking concurrence on what we all agree on, let us
instead crisply identify the open controversies by focusing on
apparent fundamental differences. It is my experience that this is at
least as valuable as the consensus results and has the benefit of
better structuring research decisions.

This is also our "holiday" session, for most of us, and I expect we
may have less opportunistic exchange as a result. So let's aim for
statements that start with the papers at hand and leverage them to
reach a sense of where lie the differences in approach, open issues
and controversies.

Fortunately we have papers that are not only interesting, but
particularly apt for this purpose.

Terry Marks interprets the FIS problem as one centered on reflection,
and suggests an informed examination of myth. I'm personally
sympathetic to this "new science" introduction into the conference.
The paper focuses on self-reference and is itself folded thrice (the
myth, the story of Jung, the nature of the problem). Because it is a
psychoanalytic approach, there is an implicit notion of "history" and
(reflected) memory behind the less fundamental idea of information.

Wolfgang Hofkirchner's paper is in the process of being uploaded to
the site, I believe. He starts at roughly the same place: information
as a manifold multidimensional entity. He instead sketches an
approach that leverages structured evolution (meaning evolution with
causality rather than highly non-deterministic mechanisms) to
structure the (three) types of information. Because the problem is
reflexive, the same symmetries overlap with the types of evolution
guiding a similar, reflected organization. Rather clever I think.

Karl Javorszky starts from apposite opposition to this idea with an
equally seductive notion: instead of starting from the effect of
information, starting from the notion of information in the system
and examining its behavior. He recognizes set-theoretic properties of
information entities. (To my mind, these could possibly map to the
symmetry types of the others in some way.) That gives Karl what he
needs to develop/discover mechanics which seem to cover many of the
problems of information transfer in biological systems. Karl has
advanced these ideas in his mails, but the paper is a more complete
presentation.

Morris Villarroel's ideas seem to share the same basic philosophy of
working to comprehend the mechanics of information per se. His
conceptual hooks are partitions, a group-theoretic notion rather than
set-theoretic. And rather than the combinatorials of Javorsky, Morris
employs a symmetric application of logical composition. This has huge
advantages in inheriting geometric mechanics that (one suspects) can
be found in biological interactions. Surely this is close to existing
mechanics in the Kaufmann-influenced community. But it raises the
mathematics from numerical to logical so that "primitives" are less
mathematically pure. I wonder what the group's feel is on these
tradeoffs.

JosÈ Bea and Pedro take a tack distinct from these two. Instead of
looking at the comprehensive world of information, they look at a
single, well-defined but poorly explained problem within information.
This philosophy of investigation has a strong pedigree for providing
paradigm-busting insights. Their domain is that of social discourse
and the problem is humor. I admit that this paper, actually the
approach underlying this paper, has made me rethink a few things. The
idea is laughter (the actual sound) as a sonically characterizable
utterance with information content, compared to "conventional"
utterances. Not surprisingly, Shannon's entropy metric fails.

But wait! We have another paradigm-busting candidate. Antonio
GarcÌa-Olivares and Pedro Vidal present a quite similar examination.
Again it is of "unconventional" information transfer, couples dancing
within a well-defined vocabulary like the tango. A variety of
communicative types and modes are noted. The notion of "dance" in
biological encounters is somewhat enticing. What do you think?

(Dear authors, forgive me for any unintended slights, misreadings or
clumsy tramping through your work.)

Are we better off with set theoretic, group or category theoretic
foundations? And how pure should our working abstractions be anyway?

Are we better off thinking about information itself, or about its
effects within the system? What legacies should we jettison? And
along those lines: do we accept the reflexive nature of information
as a concept worth leveraging, or engineer the reflexiveness out of
the (mechanics of the) science?

Are we better off looking at the entire behavior or a few vexing
problem areas? If so, what are they, and in general are we better off
working from the social or the biological domain?

Do we need new abstractions? Are these discovered or engineered and
which among the FIS presentations so far seem promising to you?
(Don't vote for yourself!)

Do we need to redefine the problem for the next FIS adventure?

As always, unless the list traffic is quite slow please limit to two
postings per week (excepting announcements).

Best, Ted
Received on Mon Dec 9 07:20:44 2002