Directory talk:Jon Awbrey/Essays/Prospects For Inquiry Driven Systems

MyWikiBiz, Author Your Legacy — Wednesday April 17, 2024
Jump to navigationJump to search

Notes & Queries

Jon Awbrey 10:54, 14 January 2008 (PST)

Archival Fragments


PRO.  Discussion Note 1


MA = Murray Altheim

Re: PRO 46.
In: PRO.

MA: Before you travel further down this particular path, I wonder if you'd be
    so kind as to relate your concept/discussion of "state" with the concept
    of "context".  Is a "knowledge subspace" a contextualized space, in the
    sense of contextualization per a specific individual, community, or
    domain, or is it something else?

Sorry, Murray, you're already 'in medias res' -- but I imagine you knew that.
This is the continuation of a thread that I started last Ides of March or so,
and it's actually the "Interest Statement" that I attached to my application
to Systems Engineering in the early 9O's -- they asked for a couple of pages,
so naturally I gave them 50.

Here I'm talking about a "system" the way they do in mathemtaical systems theory.
You have to imagine a single point moving through a single space -- that's all
there is to it.  There is no self/other distinction yet, indeed, it may not
even be possible to dis-entangle the state of an agent from the state of an
environment in any of the more customary ways.  At any rate, distinctions
like that have to be discovered on a contingent basis if they do exist.

All the information about the state of the system is represented
by a point x in a space X, which may have a very large number of
dimensions and a wildly exotic topology.

That's all you have in general, but under various conditions that
you have to make explicit you may be able to get some of the more
usual kinds of apartments and furniture constructed in the space.

For simplicity, let's say that we have a finite number of dimensions,
and that we can order the components of the state vector x like this:

   x  =  <x_1, x_2, ..., x_p, x_(p+1), ..., x_(p+q)>.

In other words, the components of state fall into two camps,
with the first camp having a rank of p and the second camp
having a rank of q.  Here, "rank" is just a 4-letter word
that economizes for the 6-syllable word "dimensionality".

In this speculative scenario, let's say we have good reason to call
the first camp of components the "physical component" of state and
the second camp of components the "intellectual component" of state.
What this actually means in practice will depend on the operational
definitions of these terms, but the general idea is that the "agent"
has more control over its own "intellectual properties" that it does
over all the stuff in the "public domain" and the "state of nature",
so to speak.

The "knowledge subspace" would be generated by some subset
of the variables in this so-called intellectual component.

As far as the usual sort of "context" goes, it's a highly derivative notion.
What's basic is the state of the agent, interpreter, observer, or system, as
all of those names are just alternative figures of speech -- personifications
or reifications of the underlying process.