home *** CD-ROM | disk | FTP | other *** search
- Comments: Gated by NETNEWS@AUVM.AMERICAN.EDU
- Path: sparky!uunet!europa.asd.contel.com!darwin.sura.net!paladin.american.edu!auvm!EID.ANL.GOV!GABRIEL
- Message-ID: <9211110604.AA14515@athens.eid.anl.gov>
- Newsgroups: bit.listserv.csg-l
- Date: Wed, 11 Nov 1992 00:04:44 CST
- Sender: "Control Systems Group Network (CSGnet)" <CSG-L@UIUCVMD.BITNET>
- From: "(John Gabriel)" <gabriel@EID.ANL.GOV>
- Subject: Re: Hamming it up
- Lines: 24
-
- (John Gabriel 921109:16:45 CST) --
-
- Can't resist - a new insight from Bruce's mail. If Penni's or my KB
- really is a space of 2**N truth values for N Aristotelian propositions
- defining a mutually agreed universe of discourse, and this space is
- sparsely populated by instances (not hard for say N=256), then one
- can triangulate a new instance by its' Hamming distances from old
- ones. BIG question - how do we put the benchmarks in?? I think by
- a set of shared experiences - sort of like the light signals of
- relativity that define coincidences in space time - these are a
- physicist's abstraction for the shared experiences that we all have
- as a society, and which are the basis for communication. Another
- question - how much is the lack of acceptance for PCT by many
- cognitive scientists caused by lack of expriential benchmarks
- in control theory, how much by badly triangulated ones, (I remember
- a quite well known ecologist claiming quite seriously that animals
- eat entropy - terrible confusion between Boltzmann entropy and Shannon
- entropy), and how much simply by different value systems. This is
- an extension of a comment by Bill Cunningham about my saying that
- most of my fights with my boss were about different values. He said
- No! many of the bigger fights are about Aristotelian propositions,
- i.e. missing or badly triangulated benchmarks.
-
- John Gabriel 921111 00:02
-