LETTER TO US ARMY FROM H.E. PUTHOFF

• Document Type: CREST
• Collection: STARGATE
• Document Number (FOIA) /ESDN (CREST): CIA-RDP96-00788R001100040005-2
• Release Decision: RIPPUB
• Original Classification: U
• Document Page Count: 2
• Document Creation Date: November 4, 2016
• Document Release Date: June 18, 1998
• Sequence Number: 5
• Publication Date: October 2, 1980
• Content Type: LETTER

? Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100040005-2
SG1J
U.S. Army
HQ INSCOM
Ft. Meade, Maryland 20755
Attn: Lt. Col. Murray B. Watt
Thinking further on a discussion we had at your place with regard to unevaluated
data, I would like to put forward the following idea for your consideration.
Last summer, at - behest, we looked into the feasibility of putting
together a computerized data management system for our own results on special
tasks. In response we came up with the following design. Each data packet
we sent out was to have with it an analyst evaluation sheet (something like
what I understand you are putting together). The sheet only requires checking
off items on a 3 or 4 point no-correspondence to perfect-correspondence scale
with regard to general geographic locale description; major stationary elements
(e.g., buildings), activity; timing elements if any; utility with regard to
task at hand; etc. (You may have noticed we used one version of this in
evaluating Hunter-Liggett data.) The idea behind the design of this list is
that it would take less than 3 minutes of an analyst's time, and he only had
to check boxes--no real decision making or essays to write. This provides a
first cut at evaluation to be returned to us. To this is added our own coded
instructions which indicate method of targeting, length of session, identity
of interviewer, whether first or later scan, etc., for data manipulation. One
would then, at the touch of a finger, call out eval numbers as a function of
who was the interviewer, how long was the session, what was the method of
targeting, etc. Over scores of trials in each category, meaningful graphs,
computer-graphics displayed, would be made instantly to check out any hypothesis
that came to mind (afternoons are better for Smith, multiple short sessions
versus few long sessions are better for Jones, etc.). This approach could
still be applied to a large portion of your past data, since eval sheets could
still be sent out to your scources.
SRI International
333 Ravenswood Ave. ? Menlo Park, CA 94025 ? [phone] ? Cable: SRI INTL MNP ? TWX: 910-373-1246
Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100040005-2
 Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100040005-2
Lt. Col. Murray B. Watt
Page 2
2 October 1980
SG1J
We are not able to follow up on this on _ portion of the Joint Services
program, since his marching orders are directed toward other items. Given that
you need results directly applicable to your next-year decision, it seems to
me that this could provide an invaluable aid. Therefore, before the ink is dry
on the S.O.W., you might consider whether, e.g., some of the audio analysis P-p
funding might be diverted to have us provide to you an optimum eval sheet form
(we could work on it together) and the data base management computer program.
You could then have a local computer programmer put it on your local computer,
and away you go.
If you're interested, please let me know ASAP. I am very excited about it
myself, as I would like to see this level of organization overlaid on the kind
of data we are all generating, and it would seem to me to be especially relevant
for your upcoming program review.
H. E. Puthoff, Ph.D.
Approved For Release 2000/08/07 : CIA-RDP96-00788RO01100040005-2