May 25, 1965

Dr. K. D. Ofer
Eli Lilly and Company
Indianapolis 6, Indiana

Dear Dr. Ofer:

Iwas delighted to receive your letter of May 21 and to learn that you
would be working in this country. I also appreciate your interest in the
DENDRAL system. Your letter prompted me to compose a brief memorandum on
the current status of the reports which should heip to answer some of your
questions. I am also including some relevant material, most of which will
go into the report as it is printed. ‘The system is ectually much further
along than you might guess from seeing what has been sent to the printer,
and I would like to encourage you if possible to spend a few days, or
perhaps even a few weeks visiting here, both to go over the system in some
detail and perhaps to collaborate in some of the current projects which
are outlined in Memo No. 30 enclosed. I have no special constraints as
far as a suitable time for such a visit is concerned, and would like to
leave this to your own convenience.

\I940

I would like to stress that the system can be greatly simplified as far as
the human interface is concerned, since there is no requirement for input
form the human user to be in canonical form. Some of the rules are rather
formideble to deal with outside the computer, and there ig really no reason
why they should be. On the other hand, I do not believe that the inter-
pretation of a canonical DENDRAL formula affords any special difficulties.
Quite possibly a good deal of time could be saved if a checklist of the
more frequent cyclic forms were set up so that the user would not have to
meke so many pencil and paper constructions. However, an Your or two's
conversation about these matters could, I think, be more informative, or at
least more persuasive than a great deal of correspondence, and I would look
forward to any possibility that you can respond to our invitation.

As I hardly need to emphasize, the connection table should form a common
interface between any sets of systems whatsoever if they are in any sense
complete, and I can visualize a wide number of input formats to fit the
inclinations of different users. I do not propose this as an inherently
desirable situaohon, but I wonder if we may not have to be reconeiled to it.
This is exactly the situation that now exists in any graphic structural
representabions which, despite the absence of any general conventions, still
seems to be the most acceptable medium of expression and communication between
chemists.
Dr. K. D. Ofer May 25, 1965
Page 2

Whether or not you can visit us here, please let me know if there is anything
else I can do to answer any of your questions or to work out any ambiguities
that you may detect in these reports.

Cordially,

Joshue Lederberg
Professor of Genetics

Enclosures:

Current status of DENDRAL
Memo No. 30
Excerpts from DENDRAL II (in prep)