perm filename EINDOR.1[LET,JMC] blob sn#798272 filedate 1985-07-08 generic text, type C, neo UTF8
COMMENT āŠ—   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	Professor Phillip Ein-Dor
C00004 ENDMK
CāŠ—;
Professor Phillip Ein-Dor
Faculty of Management
Tel-Aviv University
Tel-Aviv 69978
ISRAEL

Dear Professor Ein-Dor:

	Thanks for your report ``Representing Common Sense Knowledge: A
Survey''.  Enclosed are four of my papers relevant to this topic.  Notice
the overlap between what is said in the beginning of your paper and
the beginning of my 1960 paper, written in 1958.  Not so much has changed
in 27 years.

	In my opinion the core of common sense is a general framework
for representing knowledge about the effects of events, especially the
effects of actions.  This contrasts with the view that common sense
knowledge is like that of expert systems but concerns common sense
domains such as those you list.  MYCIN, as I mention in the second
paper knows a lot about bacterial infections, but it knows nothing
of the effects of events, so it can't make a plan or a prognosis.
It can only fill in the parameters of a prescription.

	The other two papers concern the situation calculus and improvements
and applications of circumcription.

Sincerely,