A global possibility ... LO6906

John O'Neill (jao@cook.dsto.gov.au)
Tue, 23 Apr 96 16:47:41 +1000

Replying to LO6875 --

I'm really replying to two threads here.

Marion Brady has spent a considerable amount of time developing the thesis =
that the key to learning is articulating and understanding the models we use =
as individuals, as groups, and as a society. I think this is a key issue for =
BOTH education and business.

At the same time, we have a thread that states that object-oriented ideas =
are useful for thinking about the world in multiple ways simultaneously. In =
practice, object-oriented software development has/will meet many of the =
same pitfalls of any other software development - it is not adaptive, it is =
difficult to change an object-oriented (or any other) software system as =
your business requirements change.

Why? What is missing from our software designs?

I think Marion has provided the answer - the models are missing. If we had =
the models, the underlying assumptions, and the undrelying problem space it =
would make adaptation of software systems much simpler (not easy and quick =
necessarily, but much, much simpler than today's maintenance nightmare).

What is being done about providing models?

John Warfield's research into complexity and system design has produced a =
method called Interactive Management that enables models to be elicited and =
constructed by a group of people for complex, novel problems.

I'd like to think that my PhD work will also contribute one day. I'm =
starting from the assumption that we design software systems with the models =
explicitly articulated, then look at how we can adapt these models for =
situations exhibiting some novelty.

I'm finding that Newell and Simon were right, we are goal driven (see "Human =
Problem Solving" 1972). However, AI and computer science has failed to =
consider the fact that we may have multiple levels of goals, and different =
types of goals for ongoing work, and situation specific work. If I =
substitute the word intent for goal (and intents are much more than goals), =
then John Warfield has developed structures that map the relationships =
between goals called Intent Structures. (BTW, John Warfield's work has so =
much depth and breadth I don't think I'll ever get to the bottom of it).

My question is "are intents (or goals) the basis for learning, how we frame =
problems, how we solve problems, how we conduct business"?

John O'Neill
DSTO C3 Research Centre, Australia
email: John.ONeill@dsto.defence.gov.au=


"John O'Neill" <jao@cook.dsto.gov.au>

Learning-org -- An Internet Dialog on Learning Organizations For info: <rkarash@karash.com> -or- <http://world.std.com/~lo/>