DEAR BOB:

• Document Type: CREST
• Collection: General CIA Records
• Document Number (FOIA) /ESDN (CREST): CIA-RDP87B01034R000200050004-4
• Release Decision: RIFPUB
• Original Classification: K
• Document Page Count: 5
• Document Creation Date: December 20, 2016
• Document Release Date: August 2, 2005
• Sequence Number: 4
• Publication Date: August 10, 2000
• Content Type: LETTER

Approved For Release 2007/10/23: CIA-RDP87BOl034R000200050004-4
1133 AVENUt U- rHE AMERICAS
NEW YORK, NY 10036
[phone]
Telex: 421686
Association for Computing Machinery
PETER J. DENNING
President
Reply To: Computer Sciences Department
Purdue University
West Lafayette, IN 47907
[phone]
February 24, 1982
-a?o1/11;z
Admiral B. R. Inman
Deputy Director
Central Intelligence Agency
Washington, DC 20505
Dear Bob:
I am writing in regard to the proposed revision of Executive
Order 12065 on National Security Information. Copies of this
order have been circulating and the gist of the changes were
summarized in a recent issue of Science. I have carefully
studied the present and proposed orders. I would like to offer
some general comments in case someone is interested in taking
them into consideration before the final draft is issued.
I know you are aware that this issue is stirring up a
hornet's nest of controversy in the scientific community, that
at least one congressional committee is sympathetic to the view
that through this order the government is moving toward a policy
of secrecy in science, and that the press is unsympathetic to
proposed controls on publication. I know moreover you are aware
that the members of the scientific community who are interested
in a cooperative dialog with the government about national security
are being discouraged by government actions that are apparently
being planned without their input. I therefore will not dwell
on these points.
Tone of the Proposed Order
According to the explanatory materials, the proposed order
takes a "positive attitude" toward classification. This new
attitude shows up throughout the order in many ways. For example:
1. The principle that "when in doubt use the lowest applicable
level" is replaced with "when in doubt, use the highest
applicable level".
0-1