April 16, 1965

TO: Orr Reynolds, Director
Bloscience Programs
Office of Space Sciences and Applications
NASA

FROM: Joshua Lederberg
Professor of Genetics

SUBJ: Preliminary Notes on PLANETLAB Program

Since our conversation the end of last month we have been
thinking very hard about the program that will be needed to realize any
effective mission in 1975. Every aspect of this is very complicated and
we certainly need to give a good deal more attention to our part of the
program. Even to do this properly we have had to consider adding to our
present staff as soon as possible and have some good laads (but need
whatever further advice you can offer).

The main conclusions that emerge even at this stage are these
urgencles: .

1. Planning for substantial facilities (phase 1) to be ready
no later than September, 1970. This entails final design to be under
way by September, 1968. It is certainly not too early to start thinking
about at least the budgetary planning for this right away. We would want
more study before even hazarding an informed guess, but the cost might be
anywhere from $10-50 million, depending in part on the intended overlap
with the phase 2 facility.

If we get phase 0 underway immediately (and the ARPA award plus
current NASA funding gives us at least a little headway), we should have
the experience to be able to contribute to the technical planning in
September, 1968, and,of course, continue backup thereafter.

2. The importance of a sturdy Mars reconnaissance mission at
the 1969 opportunity. This is the latest chance for basic redirection of
PLANETLAS strategy. Later sorties can influence engineering detail at one

extreme or go/no-go at the other.
0. Reynolds 2 April 16, 1965

Before we say very much more about the expansion of phase 0, we
want to make a very detailed study on our own. The general picture that
we have is to tmplement up to about 20 channels of time-shared computer
control, about 10 of which would be directly related eventually to Mars
oriented experiments. The major purpose of phase 0 is not, however, to
verify the configuration of the PLANETLAB but to outline what the main
system planning problems are going to be when we have to service large
number of experiments simultaneously from the same computer. In that
sense the work that we already have going and the additional work which
is contemplated under the request that Dr. McCarthy and I have already
submitted would constitute some 4 or 5 of the indicated channels. However,
until now we have intended to shy away from expensive hardware development
at the other end of the line as far as we could except for a few expert-
ments (for example the Pasteur Probe that we have been vitally interested
in ourselves).

In addition to the planning for the phase 0 system we are also
contemplating going through some additional exercises on our own towards
other paper configurations of a PLANETLAS that would help to give more
tangibility to the concept. How far we can go on this depends mainly on
the people that we can recruit during the next few months.

I will, of course, arrange for the earliest discussions in some
detail with Chuck Cline and our other colleagues at various places, and
you will hear more about this. I want to assure you of our own recognition
of the urgency of the timetable, and if we are still somewhat cagey about
the actual responsibilities that we might wish to undertake, it is only
that we want to be very sure of what we are offering to get into so that
we can deliver on any promises that might be implicit. Needless to say,
any tangible developments that would reinforce the possibility of the
location of the phase 1 laboratory (in particular in this general vicinity),
would lend even more weight behind our own planning that might contribute
to it.

Actually the main new thing that we might plan to get into on
the premise of a significant expansion of phase 0 effort would be the
automation of various "wet chemistry processes” of which amino acid
analysis by column chromatography and the snythesis of fluorometric reagents
and of other special materials might be rather good examples. The essence
of the matter is the timetable, and I include a very simplified sketch which
does, after all, contain the meat of the problem.

JLias

Enclosures