COMMENT ⊗   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	\\M0BASL30\M1BASI30\M2NGR40\.
C00005 ENDMK
C⊗;
\\M0BASL30;\M1BASI30;\M2NGR40;\.
\F2\CSTANFORD ARTIFICIAL INTELLIGENCE LABORATORY
\CDEPARTMENT OF COMPUTER SCIENCE
\CSTANFORD UNIVERSITY
\CSTANFORD, CALIFORNIA 94305
\F0





							18 October 1974


Jerry Elkind
XEROX Corporation
Palo Alto Research Center
3l80 Porter Drive
Palo Alto, California 94304

Dear Jerry:

\J	Terry Winograd just gave me a copy of \F1Steps Towards Language
Understanding\F0 with the remark that the process of approving its release
outside Xerox had taken since August.  This raises an issue that has
arisen before in connection with our scientific relations with IBM.

	Namely, it is not reasonable for a university laboratory and
an industrial laboratory to share personnel in a field that is active in
both places, when the industrial laboratory feels free to keep results
of the research secret.  We don't obJect to PARC having its secrets
in areas in which the same people are not active in both places, but a
one-sided relation in a field like language understanding cannot last.

	I hope you will be able to correct the situation, and I will be
glad to explain our point of view more fully if necessary.\.

							Sincerely yours,



							John McCarthy
							Director, Artificial Intelligence Laboratory
							Professor of Computer Science

cc: Terry Winograd, Bob Floyd, Bob Taylor

JMC;paw
elkind.le1[let,jmc]:SU-AI