[topicmaps-comment] RE: [humanmarkup-comment] Base Schema-channel

[paul <beadmaster@ontologystream.com>]

Len,

I am forwarding the conversation into a small forum for reference, and
perhaps some comments will occur that is relevant to the rather interesting
task that you and Rex brooks are engaged in (along with a few others).
Please forward to those in your circle who might be interested.

NIMA (National Image and Mapping Agency) has a BAA out for applied research
on a "Glass Box" for intelligence vetting.  I can send the BAA (.pdf file)
if anyone wishes.  This is non-classified work.

My working paper on this is at:

http://www.ontologystream.com/cA/papers/cA-SPS.htm

The proposal is almost finished with the accountants at SAIC.

It is on detecting events in computational spaces as part of an
action-perception cycle involving humans.  Of course the kicker is in what
to do in interactions with humans if an "synthetic intelligence made of
computer programs and algorithms" detects interesting events.

What I have here is something that might be complex, in the sense that the
humans in the loop would see the computer doing things that involved an
complex interior, much the same as if interacting with a human (almost).  We
have coined the term "Knowledge Operating System".

My interest in a state - gesture game (simulation) in which the
representations of the behaviors of humans are encoded into schema is shown
in a number of my papers.  One can imagine that entertainment-type computer
games will be one consumer of a human factors mark up schema standard.
Stratified theory (and the tri-level architecture) would have each of these
schema form out of a process of assembly (of something) in the context (of
something).  This is the process model for a "formative topic map" that I
had hoped would be developed by the topic maps group.  (Sigh...)  "My" work
is strongly influenced by Russian semioticians Pospelov and Finn.

My question is about if there is anyone who might like to work on this with
my group.  The Prime is SAIC, but I have scientific control over the
project... (at least as much as the system might allow).

The science advisors are:  Drs. Peter Kugler (psychology and computer
science), Karl Pribram (it is his 1991 book that motivates much of the deep
theory that I have advanced), Daniel Levine (leader in the neural networks
community and one of the early Grossberg PhDs), Richard Ballard (Founder of
Knowledge Foundations Inc. and developer of many knowledge base systems),
and Robert Shaw (leader in the ecological psychology community).

Additional science advisors (human factors... human mark-up) might be added
if we felt that there is a common sense of what the problems are.

A NIST ATP Gate 1 proposal is due on June 10, and I will be making a
submission to them also.  This proposal 6 page executive summary is at:

http://www.ontologystream.com/admin/KnowledgeNet.htm

and a PowerPoint (saved as HTML) is at:

http://www.ontologystream.com/admin/KTEcosystem_files/frame.htm

I invite collaboration.

The NIST proposal on a "Knowledge Net Software Framework" and other
infrastructure for knowledge science, could be a real winner all the way
around.  There is room in the tent.




-----Original Message-----
From: Bullard, Claude L (Len) [mailto:clbullar@ingr.com]
Sent: Tuesday, June 04, 2002 3:58 PM
To: 'paul'; Rex Brooks; 'Norm Badler'
Cc: categoricalAbstraction; allbeck@graphics.cis.upenn.edu;
humanmarkup-comment@lists.oasis-open.org; cognite@zianet.com
Subject: RE: [humanmarkup-comment] Base Schema-channel


(There are lots of folks in the address boxes from Paul's message.
Let me know who doesn't want to receive this.  I don't subscribe
to any but the comment list, so I don't see parts of the conversation
and assume many have the same problem of email overload.)

Thanks, Paul.  This is helpful.

The schema as a division into parts has to serve multiple
purposes, the first being to name entities to be manipulated,
and to provide a means to create structures from these entities.
This is all artificial out of the box.  That it can't absolutely
represent a living organism or model it in its complexity is the
well-known model problem aka, "Words aren't what they represent".

However, a model as a tool has a utility so as long as we agree
the model we are building has a utility and we understand the
limits of that, I am not too disturbed, but note that when
we started this project, I occasionally exhorted some of the
more ambitious members to be leery of overstating its capabilities,
particularly since we had nothing to show.  One might say at
this time, we are naming data of interest without saying too
much about why it is interesting or to what.

That the data is interpreted by the application is a fundamental,
and I hope, doesn't have to be repeated too often understanding.
That is really something I've often had to state about XML and
markup systems in general:  these are data objects.  Their
meaning is derived from their application.  I believe that
semiotics, with the notion of interpretant, covers this, yes?

In some of the emails, I refer to part of what we schematize
as observables.   Observables imply a viewpoint and the concept
that viewpoints have some virtual aspects to them is not
unfamiliar.   That belies a behaviorist perspective and
we do have to get past that and also enable a cognitive
perspective, else, we are not in a good position to model
internal states except mechanically.

>From a perspective of application, I tend to think of these
similar to the way a relational database designer thinks
of a cursor or recordset:  a value (perhaps and object),
passed to a procedure or function where the local system
(the function), does the work with the data that function
is designed to do including perhaps, returning a value
to the calling function.  Yes, that is a CS point of view,
but it is an implementation point of view.   To the degree
possible, we should design the primary schema without too
much reliance on implementation issues.  I say that with
some tongue in teeth, knowing just how hard that is to do.

But that's the gig... :-)

len

-----Original Message-----
From: paul [mailto:beadmaster@ontologystream.com]


When one moves into this definition of Kinesthetic sense, one realizes that
this is largely a brain stem function (or can be seen in this way).  Pribram
clearly sees it in this way in his 1991 book, "Brain and Perception".  But
one also sees that all of the senses are mixed into an experience of world
that is unified.

The typical argument against a schema, where the parts are treated as if
they can be removed from the whole, is that this is a reduction of the
function of the part (hearing, for example) as if hearing can be disembodied
from the living system and the other senses.  To a certain degree it can be,
but this disembodied understanding of hearing becomes abstract and
theoretical - since hearing can only be done by a living system.


The problem is not insignificant in terms of the hoped for uses from a human
mark up standard involving schema and crisp ontology.

Len, you are aware of these class of problems (yes?).  How might you address
the criticism (constructive I hope) that you are seeing from my words?

I do have a proposal for how to address these issues, but this proposal is
not so easy to state quickly and when there is opposition to a non-crisp
non-reductionist viewpoint.

(Oh well, I will state anyway..  One might use a descriptive enumeration of
the qualities of human communication and behavior, while stating that the
"meaning" of the schema are left to an interpretation.  This means that
scope and viewpoint are to be left underconstrainted.  Example:  A sixth
sense might be used to talk about the co-occurrence of an idea that is
patented at the same time by two individuals who do not know each other and
have no common direct friendships.  The notion behind the patent is then
"sensed" by a sixth sense that is tuned to the needs of the market.)

What is not a crisp ontology, but still an ontology?  Well perhaps a
ontology that is formative in the specification having a late binding but
also in that the ontology is understood to have an interpretation by a human
mind after all is said and done (this is the core notion of applied
semiotics - if I understand this correctly.)





-----Original Message-----
From: Bullard, Claude L (Len) [mailto:clbullar@ingr.com]
Sent: Tuesday, June 04, 2002 2:54 PM
To: 'Norm Badler'; Rex Brooks
Cc: cognite@zianet.com; humanmarkup-comment@lists.oasis-open.org;
allbeck@graphics.cis.upenn.edu
Subject: RE: [humanmarkup-comment] Base Schema-channel


Let's use that as the working definition for kinesthetic in the schema for
now.

len


From: Norm Badler [mailto:badler@central.cis.upenn.edu]

I'll look at it tonight.  Meanwhile note that (the movie notwithstanding)
the
6th sense is usually considered to be kinesthetic: the understanding of the
internal state of the body -- where one's body parts are relative to each
other and gravity (or other forces), e.g., joint angles, proximities,
orientation.  Touch includes external perceptions such as contact, pressure,
and temperature; kinesthetics can also include internal attributes such as
aches, pain, discomfort, pressure, soreness, etc.

----------------------------------------------------------------
To subscribe or unsubscribe from this elist use the subscription
manager: <http://lists.oasis-open.org/ob/adm.pl>



----------------------------------------------------------------
To subscribe or unsubscribe from this elist use the subscription
manager: <http://lists.oasis-open.org/ob/adm.pl>