Glen -
I think the sophistry around the defn of "model" is important, but a
distraction from this conversation. (I've got a few publications that target it almost
directly if anyone cares.)
Yes, it was an aside, but I think an important one to help Nick
follow/focus with us.
As Russ and Nick point out, this conversation is about the boundary and its
ontological status. Russ is leaping a bit further ahead and focusing on an
_effect_ of the boundary while Nick (and I) are focusing on the prerequisite
for symbol machines.
I appreciate your stating it this way. I did hear Nick ask if a system
could (somehow?) choose it's own boundaries and dismissed it as (yet
another) distraction but would now like to hear more. It felt like an
anthropomorphism to suggest a system could "choose" it's own boundaries,
but I'm open to having that explored if anyone can/will.
Similarly, your and Stephen's sparring about boundaries (compartments in
a refrigerator?) and the distinction of systems/subsystems, etc. was
not something I felt able to parse out completely, so I'm hoping your
post here leads to more elaboration of that question.
My claim is that Stephen's 3 examples are _not_ systems, much less complex
systems at all because they are idealized out of their context. In order to be
systems, they have to have some sort of objectively determined boundary (like a
petri dish). Any bounded gob of goo can be thought of as a system. An agent,
however, must be _closed_ under some operation. Hence, all agents are systems.
But not all systems are agents. Whether the agent's boundary is loopy,
self-defining, or not is the subject of Rosen's work (from which Kauffman's is
derivative).
Whether a symbol machine can be merely a system (with an objectively determined
boundary) or must be an agent (with some form of closure) is an important
question.
I'm waiting with 'bated breath, and trying to guess if this is directly
relevant to Russ's comment/question about "living in two worlds" of
symbol AND energy processing (if I understood his point correctly).
I wish I had more to contribute myself, perhaps I should study Rosen
again, some more.
- Steve
On 05/28/2017 08:40 PM, Nick Thompson wrote:
[NST==>Ok, but the question before us is, Does the system itself “get to
participate” in determining its own boundaries. <==nst]
On 05/28/2017 08:35 AM, Russ Abbott wrote:>
Symbolic processing, including computers, is a step beyond switches. Half a
century ago Newell and Simon defined computers as physical symbol machines.
We and many biological organisms are physical symbol machines also. I
think that's an important way to look at it.
The thing about physical symbol machines is that the rules of causation
they follow are more complex than those of physics.
============================================================
FRIAM Applied Complexity Group listserv
Meets Fridays 9a-11:30 at cafe at St. John's College
to unsubscribe http://redfish.com/mailman/listinfo/friam_redfish.com
FRIAM-COMIC http://friam-comic.blogspot.com/ by Dr. Strangelove
