COMMENT ⊗   VALID 00002 PAGES
C REC  PAGE   DESCRIPTION
C00001 00001
C00002 00002	\\M0BDR25\M1BDI25\M2NGR30\M3XMAS25\M4NGB25\.
C00007 ENDMK
C⊗;
\\M0BDR25;\M1BDI25;\M2NGR30;\M3XMAS25;\M4NGB25;\.
\F2\CARTIFICIAL INTELLIGENCE LABORATORY
\CCOMPUTER SCIENCE DEPARTMENT
\CSTANFORD UNIVERSITY
\CSTANFORD, CALIFORNIA 94305
\F0

Dear Dr. Morgan:

\J	I have been dithering about whether to ask you to send
the proposal out to review or to try to come to an agreement with
you about cutting it down.  It seems to me that there are some
basic questions that should be settled first.

	The proposal as written is intended as a plan for developing
a set of useful home terminal services and testing whether
a group of people indeed find them worth having.  This plan is based
on several considerations:  First, home terminals are desirable -
my ideas on why this is so are contained in the paper \F1Home Information
Terminals\F0.  Second, the main problem is getting a useful set of
servicces rather than a hardware problem, because terminals
are being improved for other reasons.  Third, while home terminals will
achieve their greatest utility in connection with adaptation of the
institutions of publishing and commerce taking into account their availability,
it is a good be to try to develop useful services that do not require
such changes.

	You have mentioned that some aspects of the proposal are vague,
and I would like to give a reason for this.  Namely, what can be done
depends on what graduate students turn out to be interested in working
on and on the usual compromise between what is best technically and
how the work can be divided into separarable projects.  Secondly, we
shall have to get reactions to the first services we provide in order
to determine what services to do next.  
Thirdly, we hope that some of the work will be financed from other
sources; the Associated Press is interested in supporting further work
on our news program, and ARPA finds some of the research of interest in
connection with making effective use of the network and with a project
on office automation.

	The main question I have for you is this.  Do you agree with the
goal and the propositions about how to achieve it in the second paragraph?
If you do, then we can try to work out the best way of achieving it
within your financial limitations.  If you have serious doubts about
the goal or general approach, then I would definitely like to have the
proposal put out for review in the hopes that the reviewers will be
more enthusiastic about it than you are.\.



					Sincerely yours,



					John McCarthy
					Professor of Computer Science
					Director, Artificial Intelligence Laboratory