RESEARCH AND TECHNOLOGY SUBCOMMITTEE; SECOND MEETING REPORT
Document Type:
Collection:
Document Number (FOIA) /ESDN (CREST):
CIA-RDP02B05208R000100050028-9
Release Decision:
RIPPUB
Original Classification:
K
Document Page Count:
3
Document Creation Date:
December 27, 2016
Document Release Date:
July 17, 2013
Sequence Number:
28
Case Number:
Publication Date:
April 26, 1989
Content Type:
MEMO
File:
Attachment | Size |
---|---|
CIA-RDP02B05208R000100050028-9.pdf | 137.96 KB |
Body:
Declassified in Part - Sanitized Copy Approved for Release 2013/07/17: CIA-RDP02B05208R000100050028-9
,
STAT
STAT
The Director of Central Intelligence
Washington, DC 20505
Foreign Language Committee
Research and Technology Subcommittee
26 April 1989
MEMORANDUM FOR: Chairman, Foreign Language Committee
SUBJECT: Research and Technology Subcommittee; Second Meeting Report
1. The second quarterly meeting of the RTS was held on 20 April 89 at the NSA
Conference Center. The following organizations were represented:
roc ouse)
USAF (Vick)
NIC (Uyehara, Maloney)
DLI (Talmy)
Of the organizations which were represented at the first meeting in January, the
FBI, State Department, FSI, and the Army, were not represented at this meeting.
All attendees at the meeting held at least Secret clearances which were all
forwarded to the Agency to allow for discussion up to the National Secret level.
2. NSA, CIA, NIC and DLI responded to the actions which came out of the first meeting:
a survey of Language Technology initiatives, and a brief of language functions at each
organization.
NSA presented a database of 40 technological initiatives in support of working
linguists. The database was carefully sanitized by the Agency's Information Security Policy
branch lowering its original classification to CONFIDENTIAL. The utility of the sanitized
version is severely limited.
CIA representative contributed a description of projects including
machine processing of foreign language text prior to human scanning, word-for-word
machine gisting, off-the-shelf Machine Assisted Translation, development of a Japanese
Optical Character Reader, generic morphological analysis by machine, and research into
Natural Language Understanding algorithms.
NIC members described cooperation with the Air
Force's FTD to use SYS I KAN software running on a PS/2 to translate Russian, French, and
German.
Declassified in Part - Sanitized Copy Approved for Release 2013/07/17: CIA-RDP02B05208R000100050028-9
Declassified in Part - Sanitized Copy Approved for Release 2013/07/17: CIA-RDP02B05208R000100050028-9
DLI delegate Vladimir Talmy brought a "Talking Paper" on educational technology
which described various initiatives in Monterey including the German VELVET interactive
video program, Interim EIDS and EIDS, and a resident satellite dish system for local
collection of international language broadcasts augmented by a SCOLA link for languages
not covered by the local antennas.
Although the reports were not in the database format requested at the first
meeting, their contents will be entered into a table for quick reference.
3. OBSERVATIONS: This meeting was very instructional on critical issues which
probably reflect the intelligence community at large. The issues are at the heart of why
there is an Intelligence Community Staff with all of its attendant committees and
subcommittees. We ran up against some of the thorny problems of trying to network, and
cooperate in our community. In brief:
? Clearance of committee members - We must all have TS/SI clearances to allow for
discussions of the real issues and details of language technology. If there is no
Need-To-Know for a particular member or organization, there is probably no
need to attend. It appeared, from the meeting, that the more a representative
knew about the complex language problems and solutions in the community,
the higher the clearance and vice-versa. This is logical and obvious. There is a
direct correlation between clearance, involvement with intelligence community
problems, and knowledge useful to the committee.
[NOTE: this observation contradicts my original presumption that any institution
involved with language would have similar needs and equivalent potential for
contribution - in fact some members seem to have little in common.]
? Classification of shared data - A related problem is the pooling of classified data.
We will be looking into the problems of distributing our database to all
appropriately cleared members. There is reluctance to share all technical data
even with TS/SI cleared members because of gray areas in the procedures for
interagency exchanges.
? membership and institutional involvement - It was clear from this meeting that
attendance and representation are idiosyncratic; tied to the individual, not the
institution. A person's schedule decides whether an institution is represented,
not the institution itself. This reflects the lack of coordination with the senior
management of the institutions, or perhaps reflects management intentions.
We should really determine which organizations want to play in this arena, and
have them support their player. Otherwise there is no incentive for an individual
to be involved. If interagency cooperation is not a priority for member
institutions, perhaps the need for such committees should be reassessed.
Declassified in Part - Sanitized Copy Approved for Release 2013/07/17: CIA-RDP02B05208R000100050028-9
Declassified in Part - Sanitized Copy Approved for Release 2013/07/17: CIA-RDP02B05208R000100050028-9
4. PLANS: The issues above will be resolved and I am optimistic about pressing on. We
will determine how best to share information among the different institutions and will
build a database reflecting the technological "State of the Community". I anticipate that
members will want to see systems at other institutions and I foresee demonstrations of
projects where the interest is high and the sponsors are willing.
The ultimate goals of the group, which should be reiterated from time to time, are
working to determine technical solutions, trading resources and information, reducing
duplication, and maximizing the use of what's available. We will "endeavor to
persevere".
Our next meeting will again be at NSA, on 20 July.
Chairman
STAT
Declassified in Part - Sanitized Copy Approved for Release 2013/07/17: CIA-RDP02B05208R000100050028-9