ASSISTANCE IN PREPARING BRIEFING ON CIRIS FOR ASD(I)

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP86M00612R000200020067-8
• Release Decision: RIPPUB
• Original Classification: C
• Document Page Count: 6
• Document Creation Date: December 15, 2016
• Document Release Date: May 19, 2004
• Sequence Number: 67
• Publication Date: February 22, 1972
• Content Type: MF

Approved For Release 2004/06/1p1h--,PtM00612R000200020067-8
22 February 1972
MEMORANDUM FOR: General Taylor
SUBJECT . Assistance in Preparing Briefing on CIRIS for
ASD(I)
1. As I believe you know, Mr. Tweedy has instructed Admiral Showers
and me to prepare a briefing on CIRIS for Dr. Hall. No date has been
set yet, but we believe it will be given in the latter part of this week.
I am focussing on CIRIS background and development to date; Mac Showers
is presenting specific examples of CIRIS use and looking to the future
and to ways of making CIRIS more useful to high level consumers.
c/n~Ir
2. The general theme is that a community information system, with a
common language, makes sense as an alternative to ad hoc.,-data calls. and
to commonality of presenting data output on! co s. It follows that
ssich a common system must have a potential utility to ASD(I) as well as
to ourselves in order to justify the effort to put it together and operate
it.
3. The particular area where your assistance would be invaluable
relates to the proper responses that should be made in case Dr. Hall in-
quires about haw the CIRIS-type data base can be made more promptly res-
ponsive to him and to his staff. In this area of the discussion, I think
there are three alternatives for consideration. I am stating them below,
and adding my non-technical observations and comments. However, I think
it would be inappropriate to enter such a dialogue with Dr. Hall without
a further technical back-up and possibly a clearer understanding of the
DCI/NIPE and CIA's aljlit. and willingness to commit resources to improving
the ASD(I) data retrieval and manipulation capability.
4? The three main alternatives, I believe, are:
(a) DCI/NIPE and OCS could transfer CIRIS tapes to ASD(I) for the
latter to run on DOD computers in order to manipulate and recover data.
This would, theoretically, produce hard copy output.
(b) A remote terminal in ASD(I) offices could be run from the CIA
headquarters building, to provide ASD I with a remote query capability,
and a printout of answers to queries.
(c) DCI/NIPS/CIRIS Staff could provide all the support required by ASD(I),
without either a transfer of tapes or a remote terminal.
CONFIDENTIAL
Approved For Release 2004/06/15 : CIA-RDP86M00612R000200020067-8
 Approved For Release 2004/06/1 5C%'~ 00612R000200020067-8
5. Transfer of Tapes, etc. This is an unsatisfactory and not
feasible solution. To be successful from the ASD(I) user point of view,
it is necessary not only to transfer tapes but also to transfer the
know-how to operate the tapes. This solution is at least a year away.
(a) The CIRIS system can be run successfully within the CIA Hdqtrs
building for NIPE use, but this is because we are all so close together
and have been working so long as a team. group know 25X1
the history of CIRIS; they have written the programs; they know the
minor quirks and peculiarities and the patching and piecing they have done
to create the software and operate it, in addition to the new programs.
(b) CIRIS is still, however, in a developmental mode. It is operational,
but it is still changing. For example, in the new CIRIS for calendar year
1972, we have introduced quite a number of new subjects and related them
to DCID 1/2 Objectives. We have expanded geographic targets. We have
added two sensors. There is nothing wrong with this process of change';
it must occur. But it means that we do not have a stable system.
(c) Since the system has not been stabilized, it would be an unprofitable
exercise to write the necessary technical documentation that would be re-
quired to accompany the transfer of tapes to ASD(I) and DOD computer operators.
I lestimates there is about one man year of programmer effort
a Will be required to "document" the CIRIS software once the system has
been "stabilized." Considerable secretarial support is also required to
transcribe this documentation.
(d) Once the CIRIS system has been stabilized and then documented, it
will still require an investment by ASD(I) of some DOD resources to operate
a computer storage and retrieval capability. This may or may not become a
problem for DOD. CIRIS has been designed in the light of computer equipment
available within OCS to operate it; this requires a 360/65 -- quite a large
machine -- because OCS programmers in writing CIRIS have not optimized
its design for core utilization. Thus, CIA has no problem running the
CIRIS on a machine with 1 million bites of core, of which the CIRIS itself
may use about 250,000 bites. A smaller size machine, or a larger machine
that is compartmented and hence has limited core, may not be able to ac-
commodate CIRIS as it has been built . This is a technical consideration
which ASD(I) should evaluate in any consideration of operating its own
CIRIS storage and retrieval capability based on tapes provided by DCI/OCS.
6. ASD(I) Remote Terminal. This does not appear to be a feasible
solution, at least in the short run, for the following reasons.
(a) It is possible to subset the CIRIS data file so as to create solely
a DOD (CDIP) data base. However, at the outset a security policy question
must be resolved, namely, whether on not the Office of Security is willing
to recommend to the DCI that a remote terminal in a non-CIA-controlled
area should be installed when the main computer driving the terminal is
also running CIA internal use only programs. This is the "inadvertent
spillage" problem,, and as far as I know, the Office of Security has never
-2-
Approved For Release 2004/06/1EUM00612R000200020067-8
 Approved For Release 2004/06MAWEVA ~86MOO612R000200020067-8
stated a flat position on this precise question. It is a sticky wicket.
(b) Unless Security is willing to act favorably on the policy issue,
the question of a ASD(I) terminal is academic from the outset. Let us
assume that Security will grudingly grant approval; there still remain
possible security aspects relating to the physical security of the non-
CIA spaces and as to protection against emissions (shielding). I don't
know the technical status of these aspects.
(c) Assuming that Security issues are all resolved, the major problem
of this course of action is that both ASD(I) and DCI/NIPE must commit to
a not insignificant investment of manpower in training ASD(I) personnel
in terminal use. The technicalities of the software must be taught and
learned; the techniques of terminal operation must be taught and learned;
and ASD(I) consumers as a whole have to know -- be willing to learn --
more than they have to date about the capabilities and limitations of
CIRIS as a data system. In addition, as we all know, terminals sometimes
are "down" for technical reasons; terminals require some maintenance after
initial installation. These technical aspects will require some continuing
level of OCS involvement to keep the ASD(I) terminal working properly; this
may not be a great deal of work but it will be continuing, and because the
user is external to CIA there are liable to be little bickerings and mis-
understandings on both sides.among technicians.
(d) It is possible, also, that a terminal in ASD(I) might noticeably
degrade responses from the master computer, thus affecting other users.
This is probably not a major consideration, but it should not be ignored.
(e) There are also some costs to ASD(I) for terminal equipment, rental
of a line from the CIA Communication Center to the Pentagon, and security
devices to be attached to the line and/or the terminal. These probably
amount to several thousand dollars, but I do not have details.
(f) There is a continuing, though probably small, cost to CIA/OCS, to
dedicate a spindle holding a disc pack to operate the ASD(I) terminal,
as well as the need referred to above to keep the circuit up and operating
satisfactorily.
(g) Another major area requiring further investigation lies in the
availability of the GIM system to ASD(I). GIM is the software package
used on the remote terminals for CIRIS. It is a package procured from
TRW, who developed it at government expense. CIA/OCS has spent several
hundred thousand dollars to expand its capability. It might or might not
be possible for DOD to obtain GIM from CIA -- at an unknown cost. A pretty
certain cost would be that of calling on TRW to contract for its installation
in DOD, assuming they don't have it now.
(h) In summary, on this second alternative, all of the technical aspects
are probably solvable, given time, and that means probably a year at least.
The security policy issue should probably be faced at the outset, for there
is no sense in working on technical aspects .if??.., security won't let the
system be installed.
-3-
CONFIDENTIAL
Approved For Release 2004/06/15: CIA-RDP86M00612R000200020067-8
 Approved For Release 2004/06/15 : CIA-RDP86M00612R000200020067-8
CONFIDENTIAL
7. Manual Support to ASD(I) by NIPE/CIRIS Staff. This alternative
is certainly the most feasible in the short run -- over the next year.
(a) The considerations here principally involve manpower costs to
both DCI/NIPE/CIRIS Staff and to ASD(I). We have to train, teach, ex-
plain; ASD(I) people have to learn and practice. Much of the time this
work has to be done in the CIRIS Staff offices; this will pose a problem
of the availability of physical working space. The CIRIS Staff is already
crowded; will get more so as a number of new people are added in con-
nection with the expansion of the NIPE (or DCI/IC) staff. In addition,
there may be times when it would be desirable to compartmentalize the
ASD(I) people from the free flow of materials and discussions within the
NIPE staff.
(b) It is possible that additional terminal facilities will be needed
in the CIRIS staff office to accommodate ASD(I) users. This would depend
on the volume of their use. For the next 6-9 months, the competition for
use of the one CIRIS terminal tied to the GIM system would make it very
undesirable to have to share that terminal with ASD(I). The present
configuration of GIM is limited in the speed of reply to terminals; it
services terminals serially,not several together. The OCS contract with
TRW for GIM-II modifications plans to make GIM capable of supporting a
number of terminals at the same time; however, that modification cannot
be expected to be in operation before the fourth quarter of calendar
year 1972. Thereafter, this problem should be resolved.
(c) In spite of inconveniences to the CIRIS Staff due to space limi-
tations, this alternative offers the best -- if not the only feasible --
prospects for making a meaningful contribution to ASD(I) as the quid pro
quo for their support in obtaining CIRIS data.
(1) This can only be accomplished, however, if the CIRIS Staff
can be expanded as provided for in the proposed T/0 that I believe
is tentatively approved. This will add two substantive analysts
to work primarily on customer servicing. It adds one "outside man"
-- probably Lt. Col. Hurwitz who becomes available on 1 April -- to
help improve the quality and speed of CIRIS inputs from DOD as well
as to simplify the data submissions for imputters. It adds the
OCS coordinator with which you are familiar, and also a secretary or
intelligence clerk for overall assistance. However, in most of
these cases, I still have not been authorized to issue a job opening
notice within CIA and thereby get started on recruitment. In all
cases, except the productive output of these new people
cannot be anticipated for several months after they are on board.
8. In summary. Can you assist in preparing for this briefing of Dr.
Hall by verifying -- or supplementing, altering, et al -- my views of the
three alternatives. Then, can you make your conclusions known to Mr.
Tweedy so that he will have a reliable appreciation of what he has to offe
to Dr. Hall, the limitations, and the costs and timing.
25X1
cc: Admiral Showers
Approved For Release 2004/06/15 : ,"?0612R000 00020067-8
 Appr
Appro%
ved F wNf lea so-1206 7
UNCLASSIFIED CONFIDENTIAL SECRE
OFFICIAL ROUTING SLIP
NAME AND ADDRESS DATE INITIALS
Admiral Showers
2
3
4
5
6
ACTION
DIRECT REPLY
PREPARE REPLY
APPROVAL
DISPATCH
RECOMMENDATION
COMMENT
FILE
RETURN
CONCURRENCE
INFORMATION
SIGNATURE
Remarks :
This is my request to General Taylor for
assistance to us in developing a DCI/NIPE
proposal, or response, to ASD(I) relative to
,at we can do for them by way of providing
IS output and servicing their requests --
Tssuming they will make some requests.
There are some technical ADP matters which
General Taylor can check out better and in a
more authoritative way than I can -- and Mr.
Tweedy will get the answers from him direct rathe
than by double-check. As the memo indicates, the
are some real short-run limits as to the help we
can give ASD I .
ENDER
DATE
22/2/72
AS -
T
e
FORM NO. 237 Use previous editions
1-67
 Approved For Release 2004/06/15 : CIA-RDP86M00612R000200020067-8
SENDER WILL CHECK CLASSIFICATION TOP AND BOTTOM
UNCLASSIFIED I CONFIDENTIAL SECRET
OFFICIAL ROUTING SLIP
TO
NAME AND ADDRESS
DATE
INITIALS
I
General Taylor
2
3
4
5
6
ACTION
DIRECT REPLY
PREPARE REPLY
APPROVAL
DISPATCH
RECOMMENDATION
COMMENT
FILE
RETURN
CONCURRENCE
INFORMATION
SIGNATURE
Remarks :
This memo deals with alternatives for
computer-related support to ASD(I) in the
use of future CIRIS data. The subject is
one that may be raised by Dr. Hall later
this week when he is briefed on CIRIS by
Admiral Showers and myself.
I don't know of any specific time yet
for this briefing. However, this memo asks
for your help in guiding us -- as briefers
and Mr. Tweedy-- as negotiator with Dr. Hall.
My personal preference would be that you
be present and handle these matters in the
Hall briefing -- provided Mr. Tweedy agrees.
TURN TO SENDER
14D PHONE NO. DATE
22 Feb 7
UNCLASSIFIED CONFIDENTIAL SECRET
FoI1M Mo. 237 use previous editions
1-67
Approved For Release 2004/06/15 : CIA-RDP86M00612R000200020067-8