home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!paladin.american.edu!auvm!FAC.ANU.EDU.AU!ANDALING
- Message-ID: <9208260248.AA03030@fac.anu.edu.au>
- Newsgroups: bit.listserv.csg-l
- Date: Wed, 26 Aug 1992 12:48:48 EST
- Sender: "Control Systems Group Network (CSGnet)" <CSG-L@UIUCVMD.BITNET>
- From: Avery Andrews <andaling@FAC.ANU.EDU.AU>
- Subject: Re: Modeling vs. WHAT?
- Lines: 34
-
- I'm not convinced that PCT-ers really understand that interactionist
- arguments are aimed at a fairly specific group of people, the `Planning
- weenies', a group of people who really seem(ed?) to believe that an
- Agent or a critter can make its way through the world by concocting
- elaborate plans with its eyes closed, and then following them (still
- with its eyes closed). According to Chapman, for example, AI vision
- people actually tend to think that there needs to be an exhaustive
- categorization of everything in the visual field, plus precise
- metric information. Part of Sonja's mission is to demonstrate that this
- assumption is unnecessary (and, by implication wrong). The idea is not
- no machinery, but minimal machinery, surely scientific Right Thinking.
-
- I think that PCT diverges from standard Interactionism by denying that
- much in the way of useful results will emerge from accidental outcomes
- of the dynamics of activity. One of Agre's examples is that if you have
- a stack of bowls in your cupboard, the ones you use most will tend to `drift'
- to the top, not because you mean for them to be there, but just as an
- accidental side-effect of the fact that the top of the stack is the easiest
- place to put a bowl you've used when you've finished with and washed it.
- Agre sees this as evidence of the cooperativeness of the world, while I
- imagine that Bill or Rick would say that this kind of effect is of marginal
- importance (as would I), & most good stuff has to be controlled for. (On
- the other hand, setting places up so that their natural dynamics tends to
- support the results you want is definitely a good idea, & I suspect that
- Agre could make a fortune as a business consultant if he wanted to).
-
- At any rate, we don't have to have an argument about this - analyse enough
- specific cases in detail, and a trend will probably emerge.
-
- Avery.Andrews@anu.edu.au
-
- So PCT seems cognitivist to Penni
- (I think) because of the emphasis on the internal representation (as
- reference levels) of
-