DociD 14011 53 IJGBPiJVVffiWGBPiJ 1 15 Bl J ijVV GBPiJ 15W B U1UiU UlI5U1Ui 15 Il OOI5UJlDl5 f OOUJUi l1WW IBm uml1rn 2nd Issue 1992 EOL4 c P L 86-36 1 I u u 1 I L GRAl OUIOlI JU DJtZ$S o o o o o o o o o o o o SZVEIII ARGOICDl'l'S FOR POBLISBIBG o o o o o o o ClIOr i 3 1 o li IIIIiIlIIl lloIiIUIj CI l IB SDULIZBD RBPORTS o l' Marian Brown o TIll ACCIDDTAI LJ XJ OOGRAPIIJ ll o o o o o o o Stuart Buck o o 1 WHO All I ARD _ All I DOIBG 1IJ llJ o o o o l l IQJOR BRUltTBROOGII IB COIIBI1IIATORIAI IIlI TBDmTICS 86-36 I I l A i I STBAftGIC COIrSIIlJ llM'IOKS J'OR WSA PROCBSSIBG LETTERS o o o o o o o o o o o oo o o o o o o o o o o o o o o NlSSIOKS rotICTIOKS OJIIDIlIu TIOKS PDSOIlJIEL OK TO TAXOIICMr or TIll OYSTU o o o VlU EOICTORY o o o o o o o o o o o o o o o r 1 25 3 o ' ' 2 2 3 A VISIT TO TDIJ o o o o o o o o o o 3 ELECTROWIC i'OBLISBIBG QUIZ o o o o o o o o o o o o o o o o o COHI'ERENCIIi UPORT RADIO IlROADCASTERS o o o I A SIDEBAR TO SIGIBT HISTORY o o o Betty Wanat o EDITORIAI o o o o o o o o o o o o o o o o o o o o TO CORTRIBOTE o o o o o o o o o o o o o o o o o o o o o o o o o I THIS DQCI TMEPJT CQPJT1 I NS CODEWORD MA'fERIAfj An 3 3 3 4 4 '''1 o 'f' A _n peclassified and Approved for Release by NSA on '10-'10- 20'1 2 pursuant to E O '135 26 vl DR Case # 54778 NOT RELEASABLE TO CONTRACTORS DOCID 4011853 CRYPTOLOG Published by P05 Operations Directorate Intelligence Staff VOL XIX No 2 2nd Issue 1992 1 -- PUBUSHER _ BOARD OF EDITORS EDITOR 1 --_ _---'1 3-7595 Collection Marian Brown 963-1197 Computer Systems 1 5877 Cryptanalysis 1 1461 Crypwlingmstics J 963 82 Information Resources J 963-3258 Information Science I 963- 34 6 Information Security 968-8013 Intelligence Community I 5S O IntelligenceReporting J l '5068 Language 963-3057 Lingmstics o 1 14 Mathematics I 963-3709 Puzzles 1 gg3- 1461 Research and Engineering I 961 62 Science and Technology o 1 9634958 Special Research o o o Vera R Filby 968 65 58 Traffic Analysis 1 1 69 Classification Officer 1 Bardolph Support I Clover Support 1 Macintosh Support I Xerox Support I lllustrawrs o o J 963-5463 oo oo o oo o J 963-4382 3369 963_7060 61 62 963-6867 963-3360 To submit articles and letters please see last page For New Subscription or Change ofAddress or Name MAIL name and old and new cnganizations and building to Distribution CRYPl'OLOG POS4l OPS-l 01' via PLATFORM c rypt1g @ curawr via CLOVER cryptlg @ bloomfield moperative at present Please DO NOT PHONE about your subscription or matters perta iDiD c to distribution Contents ofCRYPl'OLOG may not be reproduced or disseminated outside the National Security Ageney without the permission of the Publisher Inquiries regarding reproduction and dissemination should be directed to the Editor All opinions expressed in CRYP1'OLOG are those of the authors They do not represent the official views of the National Security Ageney Central Security Service P L 86-36 DOCID I 4011853 Address to the graduating class of CY-500 21 May 1992 by the Honorable Helen Delich Bentley R Representative Second District Maryland Good morning ladies and gentlemen and distinguished guests There is never a good time for both the Executive Branch and the Congress to be under siege from the press and the public but in the political It is a pleasure to be asked to address you at a dynamic where timing is critical conditions of the time like this in your lives a time of satisfaction a world demanding strong responses strong support time oflooking forward graduation means not from the leader of the free world make a weakonly an ending but a beginning ened Congress and a beleaguered President and the nation particularly vulnerable I do enjoy such an occasion because in my own life in my own experience there have been so We face a $4 trillion debt ceiling this year interest many endings to quite a few careers reporter costs in excess of $1 billion per day 12 to 14 mileditor columnist television producer and journal- lion unemployed or underemployed and a debiliist Agency chief international trade consultant- tated infrastructure bridges sewers roads all of the leave takings leading eventually to Congress And a Europe in flux from the Volga through the Balkan Peninsula into the European Community It has been a demanding path but rewarding and turning on itself breaking off on all manner of exciting trade agreements threatening GATT announcing the end of NATO with the establishment ofa And in Congress at this particular time I have to Franco-German corps focus on the rewards of serving one's country in difficult times The Congress as a body is struggling against a public perception of privilege and corruption and ineptitude To one identified with the Hill it makes no difference that I bounced no checks I didn't even have an account at the House bank that I won a primary in March handily-with a fairly heavy turnout-that most of my mail is favorable etc does not affect the overall atmosphere It is a very sad time for the establishment and a very bad time for the country The dissolution of the balance of power-some would caU it a balance of terror-between the two su r powers was anticipated was discussed and written about but there is no evidence from the behavior of this government that we were any more prepared to deal with it than were the Russians themselves No aid package was in place that addressed the particular and peculiar needs of a fragmented USSR The answer to the food shortages of the winter were massive shipment of grain tried and 2nd Issue 1992 CRYPrOLOG page 1 EOa O RGb b BSE 8NL l DOCID 4011853 true business for the last twenty years for our commodities dealers-when there was a stable upnsmg Had we supported the domestic insurgency in the fall of the year we would not have government infrastructure in place to handle had to land our troops there in December the movement and disbursement of Appropriate intelligence is of no value if it is not commodities properly used The horror of a modern war I led a drive in the Congress with 119 comparable to the scale ofWW II is beyond the congressional supporters to have at least a part of ability of the average American to imagine and the shipments be shelf-ready canned goods the costs beyond the ability of any nation to pay We must short-circuit any of these threats by We also should have been ready to barter aid for using intelligenc urbrains-to counter modern influence in the decisions made on the sale of force nuclear weapons I should have considered food or money in exchange for keeping weapons out of If it is true as some believe that the cost ofeverthe hands of extremist groups escalating weapons developments forced Russia into glasnost it pushed us into bankruptcy Hard ball You bet Business as usual will not work when there is not one strong government to Dr Martin Van Crevald's book Transformation of deal with but a group of floundering governments War presents a strong case for the disappearance driven by the need to survive of great wars and the emergence ofa series of low-intensity conflicts Call them insurgencies or This is true also on the Balkan Peninsula It is counter-insurgencies label them riots in Los not a situation that responds to the old diplomacy Angeles or revolts in Bangkok civil wars in The EC and the United Nations should be Yugoslavia-locale specific or country-contained considering plebiscites called by constituencies driven by local issue no monolithic ideologies which occupied territories in place prior to fueling the madness the United States is placed 1941 Every effort must be made to reconstitute in a leadership role which demands policies based the nation of Yugoslavia prior to Tito's terrible on regional histories anthropology and sociology policy of redistribution of minorities across ethnic and an acceptance of standards for the countries territories involved that are sometimes alien to the West Without proper planning a reactive response frequently makes us a victim-as much as any of the participants-because without preparation our responses are either inflammatory or inept But once the house is really on fire-guess who is supposed to field a halfmillion men and weaponry to put it out Iraq-Kuwait is a wonderful example of what we can do when the situation gets of of handhowever along with such success goes a huge bill But I hear little discussion about the U S policy prior to that time vis-a-vis Iraq and Kuwait Had we been wiser had we handled Iraq better would we have had to be there in the first place Certainly the Panamanian action revisited tends to convince one that we were not well briefed on the local political scene or if we were we did not have a plan in place to respond to a popular It is long past the time that we can go into the world as soldier-preachers and receive any kind of welcome If we would spread our values it will be by example only In order to do this well timely information-all kinds ofintelligence-about these countries will be of the greatest value If Dr Crevald's premise about the changing nature of warfare comes close to what is actually occurring then the nature and focus of the intelligence community will have to change also The question becomes then who will initiate the changes The National Security Act of 1992 HR 4165 seemingly has received little support inside or outside of the government In a series of hearings held over this winter the majority of the testimony from experts-former Agency officials and academia-seems to suggest that the legisla- 2nd Issue 1992 CRYPTOLOG page 2 FOR OFFfCblrL f fS 6NLl DOCID 4011853 But of greater importance to me and to the business community in this country is the report from testimony by Mr Gates on April 29th before the CIA chiefRobert Gates testified in answer to the Judiciary Committee outlining some new thrusts proposed bill on changes already under way in the for the remaining 50% of resources tasking structure of the Agency and in the efforts-across against foreign economic espionage all intelligence-gathering departments of all Remember the ultimate purpose of warfare is to agencies-to better integrate information and to seize the wealth of another country to be able to share analytical findings use its resources both human and material for It should be of special interest to you in NSA that the enhancement of the aggressors' own nation A when Director Gates was explaining some of the successful leader achieves this at the least posproposed changes in the area of strengthening the sible cost to his own people with the least damage management direction and coordination ofintelli- to the territory taken gence collection he stated in making the strucModern warfare cannot satisfy these demands any tural changes that I am about to describe I have more It is too destructive It is too costly used as a model some aspects of the National Security Agency where one individual not only is Outside of the intemecine wars the popular able to task all of the signals intelligence collectors uprisings major nations in the 21st century will available to the DoD and the Intelligence Commu- wage war with dollars instead of missiles tion goes too far and may be seen as Congress trying to micro manage too much nity but also has the responsibility for establishing standards ensuring interoperability and budgeting and strategic planning in this area He qualifies that position slightly by pointing out that the collection disciplines are sufficiently different in that they all cannot and perhaps should not exactly be modeled on NSA Indeed none can He continues by commending the basic idea of having one individual ultimately responsible for each discipline with a specific responsibility for the coordination and management of requirements for that discipline including oversight responsibility for st8ndards and strategic planning I am heartened that Director Gates addressed the economic threat at such length However I am gravely concerned that neither the Congress nor the Administration has addressed changes in the legislation which recognizes this threat One particular area of concern to me dual-use technology There is a pattern over my time in the Congress of technology b ing bartered offshore for foreign policy concerns by the State Department for economic concerns the sale of T Bills and foreign investment by the Treasury Department for the profits ofU S businesses by the Commerce Department The only Agency with any concern for the longterm strategic position of the United StatesDoD-frequently is outvoted by the big three The actions of CIFIUS a major case in point where not one sale ofcompanies owning valuable state-of-the-art technology has been stopped by the Committee even when semi-conductor Mr Gates' testimony on April 1st reported changes in focus in intelligence as early as 1980 when only technology was involved 58% of the Community's resources were dedicated We are in a global economy one into which it against the Soviet Union By 1990 the figure had seems we slipped and slid tumbled and fell withdropped to 50% I am sure you are aware of this out any preparation to protect the wealth of but most of the Congress and certainly most of this great nation represented not only by its the American people are not iWllars and raw resources but by its markets and its intellectual properties Now my mother always told me that imitation is the sincerest form of flattery More than that it tells me that the NSA model has proved itself over time and that is worth much more than flattery 2nd Issue 1992 CRYPTOLOO page 3 FOR OFFfEURtAL Sf ONLY DOClD 4011853 Small u s companies are encouraged to go jointventuring with foreign corporate giants never understanding until much too late that a contract drafted under u S law is binding only so long as the small company has resources to stay in court for months sometimes years enforcing the terms The IRS is running into the same court confrontations with these corporations that our small companies have faced And if you think that Uncle Sam has the deepest pockets and will win consider this currently there are 45 000 cases under dispute Even the government doesn't have pockets that deep and even though estimated underpayments ron as high as $100 billion They are frightening on the health of our people both physical and mental Recall when the intelligence community using Russian data-I think it was in the '70's-eame up with mass alcoholism as a major problem in Russia The Russians promptly stopped printing the information Overlay those models on the U S Someone has been waging a very successful war against this country evil man or nation or a concordance of the three We are in trouble and as the President has said you are the first line of defense Get the best information possible Disseminate it throughout the government Treat it honorably If I am frustrated that the great informational We cannot heal ourselves ifwe do not know what resources of this country are not being used to is wrong-not only in the world but inside our plan and prepare our policy makers well how own system Be courageous of the truth-there is much more frustrated analysts must be We work no other way to secure freedom with statistics in our office as broken out and As Americans we are the descendants of the basic as we can get toughest survivors of every nation on the face of Ladies and gentlemen they are dreadful on the the earth We will endure economy of the U S I wish you well God bless SUBSCRIPTION TO CRYPTOLOG o Please enroll me as a new subscriber o Please check my current mailing address o Please change my mailing address o Please change my name o Please remove my name from the mailing list because I am o retiring or resigning o going to a field site o taking long-term training o too busy to read CRYPl'OLOG though I know rm missing a lot OLD NAME AND ADDRESS CURRENT OR NEW NAME OR ADDRESS Na11U CLast first miJ Na11U Last first Org Mail to Bldg mv Org Bldg Distribution CRYPI'OLOG P05410ps-1 2nd Issue 1992 CRYPTOLOG page 4 FeR eWlCtAi aSH eNLJPY DOCID 4011853 Over the years I have often found myself extolling It is hard to explain why publishing is not more the value of publishing pointing out to younger aggressively pursued I find it difficult to believe members of the workforce the value to the that the value in doing so has escaped everyone's organization and to the individual of writing for notice Security and the attendant shyness we publication I do it for several reasons First I have regarding undue public notice explains some believe it contributes positively to the professional of it although there are numerous outlets for culture and climate of the organization second classified articles including oUr own Cryptologic there seems to exist within NSA an unhealthy Quarterly and CRYPTOLOG there are many disinclination to publish possibly due to a misvaluable lessons and insights from our work that reading or misunderstanding of our security rules can easily and adequately be dealt with in an and focus and third and most important because entirely unclassified manner and portions of the no one ever did it for me I had to learn the secret agency's work are unclassified anyway Another for myself and it took me the better part of my reason based upon the testimony of many employcareer more than twenty years to do so Invariees is that supervisors are not encouraging it ably when I finish my little exposition the emnor are they providing an example by publishing ployee tells me that no one in his or her entire themselves The sad result of this is a lost opporcareer had ever pointed this out before and thanks tunity for the organization the individual and the me for doing so It is as if I have let the person in profession I would hope that some supervisors on a secret of success that had hitherto been concan be persuaded by this article And finally cealed there is the matter of simple inertia Publishing does require effort But there is a payoff Since at the rate I am going I will never reach more than several dozen of the agency's employees Having given the subject a fair amount of thought and since I now believe that there are literally I have come up with seven reasons why someone thousands of others who might profit from my among the NSA workforce might wish to consider message I decided to turn my little speech into an writing for publication article so that more might be exposed to it 2nd Issue 1992 CRYPTOLOG page 5 f6It 6F fICtAL USE 6NLY DOCID 4011853 Reason 1 It forces one to think more deeply about the subject We often think we understand a subject until we try to explain it to someone else or write about it It is only then that we realize that we do not understand it nearly as well as we had thought And so often when we read something we have just finished writing we are often embarrassed to discover that what we just produced simply does not hold together This forces us to step back from the subject think about it more deeply and work our way through the subject again Oftentimes this even leads us to challenge some of our original assumptions and beliefs The end result is that our understanding deepens This is an extremely valuable process the benefit of which accrues even if we never actually publish the article annual essay or writing contests with cash prizes NSA's largest reward for writing the Cryptologic Literature Award carries a first prize of$2500 not an insignificant amount One ofClA's publications Studies in Intelligence automatically enters any accepted article into an annual contest for the best article The odds are quite good There are four issues a year and approximately nine articles per issue Last year the publication awarded nine cash prizes Thus the odds of receiving a prize once the article is accepted is about one in four And the prizes range from $200 to $2000 Reason 4 A published work is an entry in a Personnel Summary or Resume Resumes offer a chance to list your publications NSA's own Personnel Summary has set aside a specific space for such a listing It may be useful Reason 2 It can be an excellent source ofpsychic to know that there are some of us who always look income specifically at the Publications entry to see what if anything is included I realize there are not many No one should be embarrassed to admit enjoying people who do this and therefore most would not being the beneficiary of a little psychic income now notice if there were nothing listed However most and then Seeing in print something you have of these same people would notice if something written is one ofits best sources Justifiable pride were listed And certainly any entry is better derives from the realization that someone else than a blank thought enough of what you had to say to publish it so that others could read it too Actually the Reason 5 Publishing is a professional obligation psychi c income can occur in two increments The Most people consider themselves professionals first increment occurs when you first see your piece in print The second increment occurs when though they may have done little to contribute to the advancement of their profession It seems to you discover that the published article has been cited by someone else To have an article cited by me that if you wish to call yourself a professional you should pay some dues to your profession another author has to be one of the highest forms What you do on the job constitutes the dues you of praise I remember vividly the first time this pay to that job professional activities such as happened to me I was in a bookstore browsing publishing constitute the dues you pay to your when I noticed a book on a subject about which I profession Although there are many ways in had once written I picked it up and was which you can contribute to your profession conthumbing through it when I happened upon a tributing to the literature of the profession is one footnote reference that began with my name I of the best Writing and publishing a professional promptly bought the book paper confers upon the author considerable statReason 3 One can make money from it ure in the profession Also publishing is a professional activity with enormous leverage Many Under present law even if you wrote an article people belonging to the profession can be stimuentirely on your own time you could not accept lated by the article What you write can inspire money for it But you can enter it in some contest new thinking on the part of someone else The or other and there are such contests all the time reader then expands on your idea in another Many of NSA's professional societies sponsor article In this way the profession advances 2nd Issue 1992 o CRYPIOLOG o page 6 FeR eFFICtAi tJSK 6NLY DOCID 4011853 Failure to publish is to miss this significant proReason 7 It 11Ulkes it easier for your boss to get fessional opportunity and in my view to ignore an you promoted important professional obligation For those of you who have not been moved or Reason 6 A published work constitutes a part of influenced by any of the first six reasons this your legacy seventh and last one may have some persuasive appeal At some point in life you realize that you are running out of time and you begin to think about I like to think of our promotion system as what you will or will not leave behind You begin consisting of two gates The first gate is a binary to consider your contribution to the world your gate it determines whether a person will or will legacy Your legacy is what you pass on to future not be promoted This first gate is controlled by generations It could take the form of material one's immediate supervisor and perhaps by the possessions creative products enterprises or next person up The second gate is an analog gate simply influence on others Legacies are It determines when the promotion will occur and important Ultimately our legacy constitutes many many people control this second gate whatever claim we have on immortality particularly for promotion to the higher grades Legacies are not generally a concern of the young People tend not to think in these terms until they reach mid-life I have noticed for example that people become more concerned with their legacies as the date of their retirement nears This has certainly been true in my own case In fact it was probably an influencing factor in my decision to write this article A key group of people who exert influence on the second gate are promotion boards Among other evidence promotion boards receive written and oral testimony Because they read and hear so much testimony most of it glowing they become anesthetized to adjectives It would take a very creative testifier to come up with an adjective that a promotion board had not heard Every person that the board is considering is hard working highly motivated tenacious highly effective etc If it were were not so that person would not have been recommended for promotion in the first place Nevertheless the testifier has to offer up a suitable serving of adjectives just to keep that candidate even with the others I am sure that you have had the experience in which a valued long-time employee finally hangs it up and retires In the thirty-some years that one employee had worked she had amassed a considerable amount of knowledge and experience But because she had not written much ofit down when she retired that knowledge and experience went with her Now several months later the The testifier is then likely to relate some recent organization she left behind faces a thorny probcontribution that the candidate made that is lem its members note that ifThelma were still considered particularly significant The problem around she would know what to do But of course is that there is a strong likelihood that most of the Thelma is not around and neither is any reflecboard members will not understand nor appreciate tion of her considerable experience It is a sad the significance because the contribution is in a situation sad for the organization and even field or discipline foreign to theirs Nevertheless sadder for Thelma Sad for her because there is a the contribution is presented To the typical natural generative imperative in all of us to want member of the promotion board the contribution to leave something behind The Talmud contains certainly sounds positive Clearly having made the observation that there are three things one the contribution is better than not having made it should do in the course of one's life have a child But how does it compare with the specific plant a tree and write a book All three are contribution of another candidate in an entirely generative activities all three involve leaving different field both of which are foreign to the something behind that is likely to survive us all direct experience of the board member How is three constitute part of our legacy Publishing an the board member supposed to compare the contriarticle may be the first step to your book 2nd Issue 1992 CRYPTOLOG page 7 FOR OFPICtAi tiSE 6NLY DOCID 4011853 oooooooooooooooooooooooooooooooooo bution of a mathematician with that of a lawyer o o or a linguist or a logistician Yet that is exactly The Lead Sentence o o what members of promotion boards are required to in a do Again the testifier has to present such a Serialized contribution but because of the difficulty in como o paring them the net effect may be no more than o to keep her candidate even with the competition Report If however the testifier is then able to produce some published work of the candidate that candidate is likely to slip ahead of the others Why Well first of all because most of the competition will not have anything published Second and perhaps more important all members of the promotion board can identify with a published work They may not understand what was written they won't take the time to read it anyway but they know what it means to publish They all know what it means to write and they all know what it means to have one's work deemed worthy by some outside arbiter and actually published Having done a bit of testifying before promotion boards myself I can attest to the utility of this strategy I have used it successfully on several occasions So there you have it seven good reasons to write for publication I hope I have stimulated you to consider doing so I look forward to seeing some of your names as authors of future articles SIREN SIGINT REPORTERS NOTES P L 86-36 ______ fo5211 The lead or topic sentence of a well-written SIGINT report should concisely convey the essential facts of the report to provide the reader especially the time-pressed executive reader with a well focused precis of the contents of the report This lead sentence should always answer when possible the Five W' questions -who what where when and why It should expand upon the title focus on the main theme of the report and highlight the most significant foreign intelligence and any conclusions drawn from oo the SIGINT facts Reporting the who is o vital to the SIGINT user for assessing the importance of the information o o WE WELCOME REVIEWS of books articles software audio cassettes and video cassettes that relate to any of our disciplines orthat pertain to our mission While the lead sentence should expand upon o o the report title it should not be cluttered o with clarifying data such as unit subordinations coordinates abbreviations such information should follow later in the report In a short report this information would be in the second sentence or paragraph and in a long report in the section labeled 'DETAILS Note that short reports do not require sections headings but long reports require at least the sections heading labeled SUMMARY and DETAILS Elaborations of this guidance are contained in USSID 300 Sections 4 3 and 6 2 oooooooooooooooooooooooooooooooo 2nd Issue 1992 CRYPTOLOG page 8 FOR OFFICtAL USE 6NL l 7 EQ 1 4 c DOCID 4011853 eeNf'IBEN'f'tAf P L 86-36 P L 86-36 Specifications --_ Lockheed U-2R c - z 103 '-1 821Ml10 1nd_ -- --- y 'l JPYfI1IClht 15 101 pounctt 41 000 e n 11 OD01trIl Pr J75 P I JB f f --- 'T'1 rn o c ow iF b 1_ _' Figure 2 P L 86-36 EQ 1 4 c Lockheed U-2R Ae Remote airborne operations have long been a and were retrieved by ground crews to be reused major part of this country's overall SIGINT effort when the aircraft landed Over the years we have employed several types of airborne platforms and have enjoyed excellent y the late 1960s the aircraft had undersuccess in accomplishing our mission However gone dramatic changes The new U-2R's R for these conventional platforms have limitations that revised overall length was increased to 62 7 feet with a wingspan of 103 feet and a range of 6200 somewhat restrict our collection potential This article will show the support to military operamiles However the true utility of the U-2R was tions provided by theJ seen in its ability to operate at altitudes in excess of 60 000 feet This capability benefited SIGINT operations by making platform detection and through its involvement with the highly capable interception more difficult and by allowing a U-2R aircraft deeper look into the target area than afforded by mSTORY conventional airborne efforts U The history of the U-2 is unique Developed in the early 1950s by Lockheed the prototype resembled a jet-powered sailplane with a slender fuselage It was 49 feet 8 inches long with a wingspan of 80 feet 2 inches To balance the wings a set of dropable stanchions with small dolly wheels called pogos were attached at mid-span on each wing These fell away when the aircraft took off 2nd Issue 1992 CRYPTOLOG page 9 COi 'tffBEi 't'fIJ L wi Qbe VIA COMUff CfIAf4ffflL8 ONE i EO 1 4 c P L 86-36 EO 1 4 c P L 86-36 DOCID 4011853 REFERENCES U Larry Davis U-2 Spyplane in Action Texas Squadron Signal Publications Inc 1988 P L 86-36 EP 1 4 c 2nd Issue 1992 CRYPTOLOG page 10 C9NFIBEPfi'IAL COMHIT CIIAtlPffiLS maN IM lBLB VIA DOCID 4011853 SEGRE'I' Stuart Buck ret The Accidental Lexicographer I Fifty years ago we were in the darkest days of World War II Experienced linguists were urgently needed for translating and for code and cipher work Stu Buck was arrwng the seasoned scholars of language called to the colors U 8-0007 Operationally he worked on Japanese Chinese Romanian French Mongolian Tibetan andDzongkha as a bookbreakerandcryptolinguist as well as a lexicographer U This article is based on a talk to the CryptoLinguistic Association at an unknown time was acceptable All my cards were checked by another French linguist then we turned them over tol heeditorcitH hief- -anda superlative Japanese linguist P L B eeo We saw the trees he viewed the forest and took personal responsibility for the overall product I recall that there were about a dozen linguists in our group working from lists of J apanese words defined in some other language- r in a few instances from Japanese sources Streams of cards flowed tol who accepted or P L 86-36 rejected items in terms of his basic objectives The high value of the dictionary that was finally published is suggested by the fact that the publishers of Kenkyusha appear to have incorporated it intact into their latest edition I FOUm Sometimes I think that I have been involved in lexicography to a greater or lesser degree throughout my entire career at NSA The projects have ranged from small informal card files stashed away in my desk to a regular full-fledged dictionary For our discussion today I shall confine - - - - - - - - - - - - - - - - - - - - - - - - my remarks to three dictionary projects each of which had certain distinctive features The Japanese Technical Dictionary 8 000 At the end of World War II I was asked to assist in compiling a Japanese technical dictionary Certain aspects of this particular project were unique in my experience First of all the task itself was taken seriously No one viewed it as busy work Linguists were carefully selected for their special skills-and tested in order to determine that these skills were not illusory Once the group was set up it stayed together until the project was completed Guidelines were made crystal clear to all involved-and were enforced It was all very simple we were told to extract from various dictionaries any technical terms not included in Kenkyusha the standard dictionary 8 000 I was assigned to work on a FrenchJapanese dictionary so my basic task was to translate the French expression and to determine as best I could if the Japanese equivalent 2nd Issue 1992 o CRYPrOLOG o page 11 SEORE'I' a R bl l A 86-36 GOMINF GUA H'Jb8 ONIX EO 1 4 c P L 86-36 DOCID- P L seaT U I soon found myself writing essays on all the key words encountered including as many example of usage as I could find I had little choice 86- 3 6but to use the eclectic method so I examined for every word under study everything said in every available dictionary compared the results added what I had gained from my own experience and made out a card for my files Eo 1 4 P 4011853 c U Thewhole thing was totally foolhardy In a sense was trying to do all alone what had been accomplished byl lhighly trained team L 86-36 CO 000 Give me credit however I realized that it was madness but I never took mv eve off that Thus I tended to go easy on the flora and fauna and to bear down heavily on common vocabulary particles function words idioms likely loan words personal names titles place names organizational terms hierarchies of all types etc Until the very end I viewed the dictionary as a technical aid required to solve a specific Iproblem Perhaps I went overboard but it seemed to me that the problem demanded quite a lot Even so I was surprised to discover that the dictionary when completed contained some 833 pages L - I U Included in the listings were a gazeteer and a EO 1 4 c list of personal names There were three sorts P L 86-36 o transliterated Dzongkha in true dictionary order o a phonetic transcription according to a system devised by Indian scholars and o English meanings U I like the system of deriving vocabulary from current texts and the ability to correct update and edit periodically but I miss the essays or articles that were characteristic of key words in my Tibetan dictionary Problems in Lexicography U Certain problems were commo ' to all three dictionary projects described above Most important of these was the answer to the question what shall I include and what shall I leave out Exactly what am I trying to do tend a dustbin or forge a precision tool U Also who will answer for the final product a lot of anonymous contributors or a single individual What shall I do about things I do not understand-ignore them and hope they will go away Suppose I understand some tricky expression or usage that is devilishly hard to explain to someone else Can I assume that everyone else understands it just as well and save myself the trouble of clarifying it U Is it enough to provide one-for-one equivalents if there is such a thing or am I as lexicographer obliged to get into language structure o 2nd Issue 1992 o CRYPrOLOG o page 12 EO 1 4 c P L 86-36 FJCREi' Ioh VIA COMI lT CIYtPH'ffiLS OnLY DO tDs6_AOllB53 ---------------------------------- I 'Doing Jfere EO 1 4 c P 1 86-36 f i i EO eeO Once upon a time there we e graphic linguists and voice linguists and t there were cryppies At that time and place I was C3S tE Q a -- --1 cryppie Not a cryptanalyst mind you bu language person w-ceo A good command of the language was and is absolutely essential for this work MF f But once in Z I discovered that there was a whole other world out there While in some languages cryptolinguists need a great degree of depth-years of experience target knowledge and excellent language skills-- lther Z people worked on targets which required somewhat less depth but more breadth-introdl1ctory course in language possibly in a number oflanguages but requiring far more knowledge of cryptanalytic techniques 1 C-COO Well since I did no reporting and translated only an occasional cryptanalysis-related message I didn't feel like a graphic linguist and so I started thinking of myself as a cryptolinguist Perhaps this was in self-defense as linguists would always see me as a cryppie while 8 GGO So what is the correct definition of the cryptanalytst would refer to me as a linggie term cryptolinguist The Glossary of - - - - - - - - - - - - - - - - - - - - - Cryptanalytic Terminology 30 Sept 91 defines cryptolinguistics as the branch of study embracing the characteristics oflanguage which h_a_ve_ _- some particular application in cryptologyl 2nd Issue 1992 CRYPTOLOG page 13 SfiORJft' IIMfDLE TlfA OOMI-N f OItMifffiLS mall EO 1 4 c P L 86-36 DOCIO 4011853 L - 0 000 The question remains What is a cryptolinguist And Am I one ----JI iC-OOo Over the years many people have met for countless hours to try to put together a COSC tracks the primarily CA track and the priInarily language track working with both tl1eCA and Language career panels S-CCO I EO 1 4 c P L 86-36 I for cryptolinguists Yet the toughest part of that is defining what a cryptolinguist is Is this a broad field encompassing both those majoring in L - _ CA but using language daily as well as those who now will they be treated by language boards Is major in language as much value given to cryptolinguists as it is to voice linguists For a clue look at FLIP allocaL I How do you set up criteria tions Ifwe had not been split up would there be for a COSC that covers both Must a enough of us to consider as a separate group cryptolingist be professionalized in both fields 0 000 What do you think linguists 0 000 Or are these really two totally separate cryptolinguists and cryptanalysts out there Is fields It has been suggested that the there any chance that we can finally come to a cryptanalyst using language is a cryptanalyst consensus on what a cryptolinguist is Can we and should be considered as such by tech track come up with a cose which encompasses both boards etc But what of the languagethose strong in language and those strong in cryptolinguists Are they cryptanalytic enough cryptanalysis Or should these two groups of to be considered fairly by CA boards or must people be treated separately they earn their accolades elsewhere-through A fi'OUO I would like to hear your views You may and B Group language boards call me on 963-5071 On e-mail I can be reached 0 000 If we define a cryptolinguist as someone rhardco can be mailed to who is somewhere on a continuum that stretches me at Z443 Ops-l Your responseswoWd be most between two poles language and cryptanalysis P L 86-36 welcome I 1 Advanced Function Printing 11 Job Entry Subsystem 2 All-Points Addressable 12 Magnetic Ink Character Recognition 3 Computer-aided Acquisition and Logistics Support 13 Optical Character Recognition 4 Cathode Ray Tube 14 Overlay Generation Language 5 Dynamic Job Descriptor Entry 15 Print Description Language 6 Dots per inch 16 Print Service Facility 7 Disk Operating System 17 Revisable Format Text 8 Forms Description Language 18 Standard Generalized Markup Language 9 Generalized Markup Language 19 What You See Is What You Get 10 Job Control Language 20 Xerox Escape Sequences 2nd Issue 1992 o CRYPrOLOG o page 14 SEORB'I' lilA OOMDn' OUJdRfBLB OULV DOCID 4011853 p L -- Major Breakthrough in Combinatorial Mathematics Amajor breakthroughincombinatorialmathematics has been achieved in R51 where researchers constructed a Hadamard difference Bet in a group oforder 100-a feat thought to be impossible by most experts in design theory a highly developed science in which conventional wisdom held that the order of such a group could not have a prime factor greater than g The new difference sets provide a means of constructing highly structured block designs and related Hadamard matrices which are square arrays with entries 1 and-1 and whose rows are pairwise orthogonal Such arrays have a myriad ofapplications in areas as diverse as statistical design and coding theory which exploit their pseudorandom correlation properties These properties are similar to those ofM-sequences bentfunctions and perfect binary arrays which correspond to difference sets in more familiar abelian groups Until this recent discovery the only known Hadamard groups had order of the form 4N2 where N 2 a 3b Since the family of Hadamard groups is closed under products the new result maybe combined with the old to produce Hadamard groups of orders ofthe form 4N2 where N 2s 5s 1 or N 2 a 3b lO c There is hope that these constructions may be generalized to produce Hadamard groups of any order 4N2 N 2a gb5c and perhaps others as well The new result also provides a counterexample to a long-standing conjecture ofThomas Storer ofthe University ofMichigan that a nontrivial difference set could exist in a nonabelian group only if there exists a difference set ofthe same size in an abelian group of the same order It is ironic that the nonexistence ofabelian Hadamard groups oforder 86-36 IR51 100 was proven recently by University of Minnesota-Duluth researcher R L McFarland who was the one who promoted the subject ofdifference sets in the Agency when he was stationed here as a young Air Force lieutenant twenty-five years ago The break-through work was performed as part ofa project to determine which if any of the sixteen groups of order 100 could contain a difference set The four abelian groups were ruled out by McFarland and of the twelve nonabelian groups six had been ruled out by the combined efforts ofR J I cFarland c fNork astlleJ foCllSed on the six reIJlaining undecided groups andinparticula r on the socalled Gg P L 8 6- 3 6 1 Ken Smith ofCentral Michigan University here on sabbatical last year outlined an approach to this problem and did a preliminary computer search which reduced the size of the final search the sought-after difference set is a 45-element subset of the group of 100 elements but trying all such subsets is out of the question d a more careful costing ofalgorithms to effect the final search and did more computing to build a database which would constitutethe search spacefor a branching algorithm In early January the approach looked feasible but had not been programmed D resented a status report on the project at the Baltimore meeting expressing optimi$m at the prospects of Gg Other duties then intervened I I In anticipation ofthe AMS meeting at Lehigh 11-14 Aprill lresurrected his work and enlisted t writetbefinal program Ted quickly wrote a very clev r C program which effectively found all difference sets in Gg and thu$ immortalD ity I 2nd Issue 1992 CRYPl'OLOG page 15 POR OPPIOW USB ONLJPY P L 86-36 DOCID 4011853 P L 86-36 Strategic Considerations for NSA Processing te7 The purpose of this article is to evaluate the likely effect upon NSA processing of technological and non-technological factors over the next decade and to propose strategic considerations for evaluation in determining how best to deal with these changes In particular this document seeks to update the prior Future SIGINT Capabilities FSCS documents with respect to NSA processing The User Interface System UIS is a multilevel ADP architecture that is intended to satisfy current and projected analytic support requirements while providing an individual user with access to data and services regardless of physical location It represents a long-term continuing effort already well underway to specify and provide an architecture that adapts to changing and evolving processing and support requirements As such it becomes the starting point for the strategic considerations developed in this document The basic UIS 3-Tier architecture is illustrated in Figure 1 THE DISTRIBUTION OF PROCESSING 3-Tier UIS Architecture with intermediate often referred to as departmental capability computers Tier 2 where the bulk of the processing needs for that community of users was expected to be accomplished These Tier 2 processors typically thought of in terms of the mini-computers of the day would where necessary communicate with a relatively small number of increasingly powerful mainframe computers or computer complexes Tier 3 for compute-intensive number-crunching processes for access to and often down-loading of portions of very large data bases and for eventual storing of processed data which might be needed by users outside the immediate community Users were expected to have little concern with the mainframes which might be accessed sinre the primary processing would be accomplished by the evermore-numerous Tier 2 computers U The 3-Tier architecture is currently the ac- cepted standard for a large installation-and is pertinent to smaller installations as well principally by eliminating the need for one or more tiers-and is expected to be relevant well into the next century The basis for the Tier definitions has already changed however and will change further as technology evolves U The high-level UIS architecture Figure 1 developed in the early 1980's was predicated upon processing trends at that time and mirrored industry expectations and architectural trends of that period widespread processing distribution with decrease oflarge mainframe processors Users were expected to interact through either smart or dumb terminals Tier 1 principally Processing Distribution Trends Among Tiers U Already the personal computer revolution has exerted great influence There are very few dumb terminals still in use and even smart terminals have been largely supplanted by powerful personal workstations As the power of these workstations increases-already the high-end Tier 2nd Issue 1992 CRYPrOLOG page 16 iEGU'I' 11 UWl Ji JPYIA GOMmT GlWif f Sb8 mas DOCID 4011853 USER INTERFACE SYSTEM User Geogr lphk Community User Service Service Service M lGE oGRAPmc jCO ' - 'i T R ANSPORJr SERVICE ' i ' ' Network ' Iii ifffiiA_GEbG ifiE tBd t' i%ib i ' TRANSPORT SERVICE ' ' ' ' W' ' B A' C K C A BELLMASTER GeogT lphic Community Service K R i ' B 0 Oi' 'N'N r---- --- -- -- -- E ' _ -- - _ _ E GTS BACKLIGHT 1 processors are approaching the power of the early supercomputers--more and more processing will be and is accomplished in user areas This capacity and the increased sophistication of endusers are certain to support the momentum toward dispersing processing functions to Tier 1 It is likely that in a few years most tasks previously carried out in Tier 2 and many in Tier 3 will be accomplished routinely in Tier 1 plexes might migrate to the next substantially more powerful generation ofTier 2 departmental computers serving to replace those current Tier 2 processes which will migrate to Tier 1 U Another possible scenario which appears to be somewhat more likely is that most oftoday's Tier 3 processing will remain in Tier 3 whose computers will also become ever-increasingly more powerful If this happens it is likely that much of -terThat leaves a less predictable future for Tiers the Tier 2 processing that is not suitable for Tier 1 may be redistributed to Tier 3 leaving the very 2 and 3 Some need for Tier 3 processing appears to be certain for the foreseeable future at least for real possibility that Tier 2 will largely disappear There is much in favor of such a coalescence certain highly computation-intensive processes including architectural simplification reduction of such as cryptanalysis and for storage and retrieval of very large databases only a small subset support costs reduction of operator costs and of which might be relevant to any particular set of simplified network management Opposing this users It is possible however that the majority of are the cost of mainframe processors vendor flexibility power and air conditioning needs and the functions now carried out by mainframe com- 2nd Issue1992 ' CRYPrOWG ' page 17 JsX SElCft lft' VJA Gg G JNEbS ONirY DOClD P L 86-36 EO 1 4 c 4011853 possibly elimination of whatever security benefits may be inherent in the use of departmental machines Both industry direction and cost-benefit analysis will influence the choice em It is safe to say however that Tier 2 will remain relevant at least into the late 1990's albeit with a reduced and less important role such as support for Tier 1 rather than processing functions per se Likely functions include those offile servers traffic distribution database repositories and input output support processors for Tier 3 processors MAN-MACHINE INTERFACE The Distribution of Processing r-------------------- Standard User Interface U There has been much discussion both at NSA and elsewhere about the desirability of having a single user-friendly mechanism to permit any class of operator to interface with the equipment An early goal was to define and implement a standard interface for all programmers operators and analysts who deal with computers At first it appeared possible to come close to this goal in terms of a single workstation ASJ W with a single keyboard configuration connected to a single intermediate computer ASH using a single operating system UNIX and a single programming language C Now we know that none of these assumptions has or come to pass We realize it to be a nice theory with more apparent than real benefits and at a staggering cost if in fact it proved to be technically feasible It did not address the problems of freezing on outdated technology in a rapidly changing field Realizable Man-Machine Improvements -teT The planned use of windowing software packages for high performance workstations permits access to many systems through a single keyboard and workstation This in itselfis a major improvement in man-machine interface simply by requiring an analyst or programmer to learn only one machine keyboard and set of procedures for any given job Now that the basic technology is here there is a need to acquire modem workstations with state-of-the-art user interface devices and capabilities and to train our workforce to make use of them Further analysis is required also to 2nd Issue1992 CRYPrOLOG page 18 P L 86-36 EO 1 4 c 1 Y 1 Vh' SBGBBT GOlcffiW GIWH' LB Ol'a JPY DOCl D 4 O 1 1 B 5 3 s e e fHi T determine which equipment operator functions P L 86-36 STANDARDS AND PROTOCOLS In particular there is as yet no standard UNIX nor is there likely to be one in the near term Moreover as hardware becomes relatively cheaper with respect to total system cost system vendors increasingly differentiate their products with value added software functionality As a result the software features available from vendors are not common across the industry and will not be so in the future If one NSA element makes heavy use of proprietary value added features of one vendor or consortium of vendors portability to a different element or application may be lost U The use of standards particularly standard U Although software or hardware standardiza- protocol suites is required by the UIS architecture and is essential for achieving interoperability and commonality among differing functions and systems Even though the Agency is in the process of adopting a much more flexible hardware standards program its principal standards efforts are moving away from the low-level standardization on hardware and toward the higher-level standards available for both software and protocols tion may alleviate some difficulties that hinder interoperability it is difficult to achieve commonality let alone standardization for many reasons While commonality is a rational goal to promote software cost sharing for example the real key to SIGINT production in a joint environment is system interoperability System interoperability can come about most easily when the parties involved are following the same architecture which is heavily dependent upon common usage of a set of standard protocols at all logical levels Unfortunately although NSA and other organizations are moving in this direction there is not yet widespread availability of stable commercial packages that implement the protocols necessary at all levels It is reasona6le however to expect this problem to be resolved over the next few years Problems U Since the mid-1970's there have been in- creasing Agency efforts to provide a degree of standardization on computer hardware in an attempt to curtail the enormous support burden inherent in a multi-vendor multi-component environment and also to try to achieve the obvious benefits of system compatibility by use of identical hardware As the computer marketplace has evolved toward the widespread use of standardized packages ofboth hardware and software the Agency has attempted to do the same It has become obvious however that a single choice for any given purpose cannot handle the multiplicity of uses which Agency elements require As a result standardization efforts are moving increasingly into the realm of software and especially protocol standardization U The focus ofNSA's software acquisitions has clearly swung toward UNIX-based systems This trend can be expected to continue and it is expected to enhance software portability and to make commonality among different NSA systems easier to achieve Nonetheless it is not a panacea Movement Toward Standards U It has been the case and will continue to be that there exists a multiplicity of available standards in any given arena In most cases a choice from among a hopefully small set of acceptable standards will be available to implementers Even in cases in which we attempt to use a single standard the magnitude of the installed base together with the rapidity of technological and standards change make it unlikely that we will ever deal exclusively with a single standard throughout the processing system There must however be an increasingly strong movement away from vendor-proprietary technology and toward the use of open standards and components which can be acquired from multiple vendors 2nd Issue1992 CRYPl'OLOG page 19 SBGRBl' Ih'Y'mI aE lift GOMi I'f'i' GIfMflrf SLS may DocrD 4011853 Even here it is unlikely that large segments of the installed base will actually be converted largely because of the immense investment in software Rather we must attempt to encapsulate proprietary communities minimize the proliferation of proprietary technology to new systems and provide standard gateways and bridges to interconnect the proprietary complexes to the external open UIS world Categories of Standard U Regardless of the type of standard experience has shown that a single usable-in-all-situations standard is probably neither achievable nor desirable Rather small families of acceptable standards should be defined in most instances to balance the conflicting needs of interoperability and lowered support costs with flexibility of choice and optimization for a given environment This leads to a grouping of standards in three basic categories U Thus the UNIX operating system is highly recommended for most applications and will be even more so as the various versions of UNIX coalesce toward a single universally recognized standard There may always be some applications for which UNIX is inappropriate and some especially large mainframe-oriented complexes for which its introduction within a reasonable period does not appear to be feasible technically or economically Still other areas may have an installed software base dependent upon a competing operating system which may preclude any movement toward UNIX U Similarly a relational DBMS is recommended for most applications with the highly available and economically competitive INGRES recommended as the DBMS ofchoice However it is recognized that there are some applications for which this choice would be inappropriate and for which a different relational DBMS or a traditional hierarchical inverted file DBMS would be more U Mandatory Standards These are and must sensible Such choices should not be dictated and remain few and far between Currently there is a system designers should be free to choose an alternative for cause In the same vein certain single mandatory standard for the USSS that of DoD standard protocols e g TCP the transport the network address protocol standard currently layer networking protocol may be recommended the DoD Internet Protocol IP This is necessary as the protocol of choice for most applications It in order that all systems throughout the SIGINT is recognized however that the associated capasystem can address one another and mutually bilities and often overhead may nof be justifiinteract but even this requirement is less than able for some applications and also that emerging absolute IP must be made available for excompeting e g Open Systems Interconnect protochanges between communities but it is not manmay be an appropriate alternative even in cols dated for use within a local community of users Moreover some complexes historically dependent advance of acceptance of the full OSI protocol upon a given vendor and that vendor's proprietary suite protocols are unlikely to provide even IP as a U Selectable Standards There will continue canonical in the foreseeable future relying instead to be many instances in which selection from a upon their own internal protocols and on gateways small set of approved standards will be acceptable to make the necessary translations to the stanpermitting system designers the flexibility necesdard world outside sary to make an informed choice for a particular application while still adhering to the require U Recommended Standards In addition to ments for meshing with a controllable required standards there will be an increasing number of standards of choice recommended for supportable architecture For example one can select freely from among the various higher-level use unless there is strong rationale for using a competing standard Reasons for such recommen- standard DoD protocols and this freedom of choice will undoubtedly continue after OSI is embraced dations may include both technical and resourceoriented rationale to include economies of scale bulk-buy discounts user familiarity and ease of portability U With respect to languages we will attempt to reduce the number which are used and thus must 2nd Issue1992 CRYPI'OLOG page 20 SHEJIlB I' H Mf BLB JPYfA eO PiHN'f eUAfflffiLS ONfl DOCID 4011853 be supported but will not try to impose a single will arise if they select a different choice for cause 'I I language for all purposes It is likely that use of c will continue to increase within the class of applications for which it is appropriate FORTRAN use though likely to falloff will not disappear in the near future and use ofAda will increase especially for new applications eventually supplanting many of the older languages such as PLII For networking IEEE 802 3 Ethernet technology is expected to predominate in the near future yet IEEE 802 5 Token Ring will likely increase within its own realm of applicability By the mid-1990's FDDI Fiber Distributed Data Interface is expected to be the dominant technology D Regardless existing proprietary technologies will continue for the foreseeable future within existing local complexes due to their widespread use in many of our current applications They will however become increasingly encapsulated within the complexes where they predominate and standard commercially available interfaces will be used to connect them with other elements of the processing system Their use will gradually wither and die out as they are replaced by new or updated systems in which standard open protocols are built in Summary U The use of standards of all varieties is key to our architectural planning and is essential to meet our requirements within acceptable cost However standards should never be viewed as absolutes and must be used in a sensible rather than dictatorial manner In particular practicalities indicate that the most promising road to compatibility and easy accommodation of technological change is by emphasizing well-defined protocol and interface standards which are widely supported and available Secondary benefits will result from use of common software to the extent possible Standardization of hardware while very beneficial in areas where it can be accomplished is very difficult in a broad sense It is incumbent upon system designers to be aware of the various standards available to consider carefully the benefits which they offer then to make a reasoned decision accepting in advance the problems which CONNECTIVITY UIS Connectivity Goals D The requirement to provide connectivity among the various systems and personnel within the USSS is the most basic goal ofthe User Interface System architecture The strategy for achieving this is once again to follow the lead of industry wherever possible and to concentrate on use of commercially available components and standard protocols Our connectivity requirements are very similar in concept to those faced by a broad array ofindustrial and governmental organizations although the number of connections anticipated data volumes and geographic dispersity of our environment present particular problems faced by only a few Three illS connectivity precepts affect all network planning All-ta-All But Not Any-ta-Any U To meet the basic illS goal of providing any user at any location with whatever processing and data resources are necessary to accomplish its job it is necessary to provide the technical capability for any person or system to interconnect with any other person or system At the same time it is necessary to pr vide the capability to restrict such paths to those that are permitted and mecessary for operational and security reasons Networks Must Accommodate Devices The USSS processing system does and will encompass a vast array of systems and devices from many vendors representing a variety oftechnologies and technical generations It would be unreasonable and unrealizable to require each potential vendor to make special modifications to its standard product line to interface to our networking structure As a consequence our networks must be designed to accept unmodified connection of this array of devices This requires once again use of standard network technologies standard protocols and standard interfaces all compatible with the technologies and protocols supplied as a matter of course by the majority of systems being produced by the various vendors 2nd Issue1992 CRYPrOLOG page 21 SEeB fi ItMfDikl VTA eOMUff' CltMfrfBLS Offill DOCIO 1011 B5-3 Management and Security m Although network management and security are problems faced by all large organizations our needs are and will remain more severe and more stringent due to the size and diversity of our system and especially to the unique security requirements which we must meet It is in this area probably more than any other in which we will have to push industry to meet our needs fund special developments and utilize much in-house expertise to satisfy our requirements Network Segments and Technologies U Global networking in the 1990's and beyond can be thought of as consisting of five categories of networking segments four UIS one GTS each with different requirements and each at least potentially using a different type of technology These segments are the following after OSI is within NSAW in the very near future and ultimately throughout the system This technology will be upgraded to FDDI as soon as it is available and might well be upgraded again to enhanced FDDI probably 200 Mbps during the course of this decade assuming that this technology is defined and becomes available Even the basic FDDI version should remain adequate into the next century however and a replacement of markedly different technology is unlikely principally for cost reasons This technology can be expected to be standard consistent and long-lived throughout the system Inter-Complex Segments User Area Local Area Networks m These LANs will interconnect user workstations and small departmental computers or servers within a local area often defined in terms of a community ofinterest Typically these are 10 Mbps ETHERNET IEEE 802 3 LANs today and this technology is expected to dominate in the near-to-medium term By the mid-1990's we will likely still have ETHERNETs in place but 100 Mbps FDDI fiber optic LANs will become dominant as fiber is extended into user areas throughout the system This technology should remain adequate for such use well into the following decade and cost considerations indicate its early replacement to be unlikely Interactive Backbone Segments m These high-speed LANs are used to intercon- nect user area LANs with one another with intermediate i e Tier two processors and with central processing complexes BELLMASTER 1 5 SO Mbps backbones are being installed to satisfy this requirement today under Project CLOVER and BELLMASTER is expected to become the ubiquitous controllable supportable architecture For example one can select freely from among the various higher level standard DoD protocols and t'h'Q o 1 ' o unll oo oo 2nd Issue1992 CRYPrOLOG page 22 SECRET HAlfflLE TJiA eOMHf'f e L8 ONLY P L 86-36 EO 1 4 c P L 86-36 EO 1 4 c P L 86-36 Dcltt c 4011853 Systemic Balance G GGO The historic and on-going qu stion of proper balance between collection and processing both with respect to capability and resource commitments is certain to continue indefinitely Wide Area Network WAN Segments eer The final category of segment is the WAN which will be defined under auspices of the Global Telecommunication System GTS These segments will be used to interconnect instances of UIS i e campus processing complexes at each facility throughout the world Represented by a variety of communication links today packetswitched dedicated circuits Tl and T3 carriers fiber satellite etc these segments can be expected to embrace a variety of rapidly changing technologies at any given time Once again standard protocols and interfaces including use of carefully selected gateways will permit GTS to pass processing data from one illS network to another in as transparent a fashion as possible thereby effecting truly global processing connectivity The WAN segments must be capable ofinterfacing to both the interactive backbone LANs and the inter-complex backbone LANs as well as embracing new communications technologies PLANNING AND BUDGETING CONSIDERATIONS Cost Considerations m Cost ramifications both dollars and person- nel are necessarily an inherent part ofthe planning process regardless of the issue being examined Consequently cost considerations will not be separated in this plan from other planning considerations It must remain obvious however that affordability is the sine qua non for any aspect of planning and that all cost factors must be considered in the planning process including those e g facilities which are affected in the aggregate but which are difficult to quantify as a function of a specific decision - 1 U Equally important are considerations of having sufficient i e neither too few nor too many personnel with the proper skills and training in the proper locations to fully utilize the technical collection and processing assets available Moreover we must have sufficient and robust communications to ensure that whatever amount of data are collected can be moved in timely fashion to the point of processing These considerations must be balanced also against such aspects as provision of the proper levels of support for equipment and personnel regardless oflocation proper facilities to include power and cooling needs and more difficult to assess the cost-effectiveness of modernization and asset optimization against the constrained investment resources which we must face Cost Drivers of Modernization 0-000 We must continue to move the SIGINT system forward to meet the ever-increasing challenges which it faces but any fonn of progress has cost ramifications Some of these cost-drivers represent simply the cost necessary to achieve a new necessary capability Others may represent the initial investment cost necessary to achieve improvements which have the potential of reduc- 2nd Issue1992 CRYP1'OLOG page 23 8J3GR8f HANB'b YY OOMHff OIWiN BLS mHrY DOCIO- 4011853 XiIilLB iur Ci gIven ievel of capabIlitY m future years Still others represent potential cost savings short-term at the expense of some operational capability All of these must be examined on a cost-benefit basis based upon relative potential to fulfill our mission timing ofcosts incurred and the time-value of money degree to which various resource expenditures equate to relative fulfillment of requirements and to the timing of when given capabilities must be available Typical cost-drivers to be so considered include the following o survivability in all its aspects o computer security to include networks and systems o need for pushing industry via funded efforts to advance technology in given areas to achieve the benefits of that technology earlier instead of waiting for the technology to develop on the basis of market forces o benefits and drawbacks of attempts to achieve commonality e g use ofa single programming language such as Ada and the extent to which such commonality should be mandated o cost-benefit analysis of various productivity enhancement programs including personnel training and use of such technical developments as artificial intelligence and expert systems Organizational Considerations m It can be argued easily that the current organizational structure of the Agency not changed significantly in the last decade and predicated upon 1970's concepts of missions and roles is already highly inefficient in dealing with our current problems If unaltered it is certain to become even more obsolete and ineffective in the years to come Recent trends in technology and processing have had a particularly pronounced blurring effect upon the relevant roles of technical personnel in the Operations R E and Telecommunications Organizations There is already a great deal of duplication of activities across these key components often at cross purposes and with little coordination or mutual visibility and widespread misunderstanding of or refusal to recongize defined organizational roles Ifleft unchecked in the coming years of still more rapid technological change we will be faced with a highly inefficient immensely costly technological anarchy o effect upon facilities of new processing hard--rerGBPareful analysis must be made of the proper ware which often takes up far less room but requires far more power and generates far more roles of each organization with respect to current heat and impending processing technology a partial reorganization may be desirable and the defined o benefits and cost savings of preserving roles must be understood and accepted by all We existing software and the costs and problems of will not be able to afford the levels of organizadoing so compared with the presumed later benefits of new more standardized more easily main- tional autonomy which we have today yet we tained rewritten software must be very careful not to go overboard in defining roles to the extent that we are left with a o loss of capability inherent in disestablishing stifling inflexibility of options Of particular existing outdated systems compared with the benefits to be achieved in supportability space impact in these decisions are the ramifications of and compatibility with newer systems the current trends away from centralization and toward enhanced processing power at each o benefits drawbacks and reasonable extent of use of standards to include use of a common analyst's desk Although there is a potential for undercarriage for diverse systems enormous gains in productivity and SIGINT output there are also potentially many problems o proper investment balance between general processing systems and infrastructure which most of which have been ignored heretofore It support all organizations and the need for immust be realized that analysts are not trained provements in particular operational areas programmers that the efficiency of use of their o benefits versus cost and complexity of multi- workstations as processing tools will be mission processing systems cross-correlation substandard that there will be widespread dupliacross collection types and flexibility for use in cation with little reusability of software provaried scenarios peace crisis and war and 2nd Issuel992 o CRYPI'OLOG o page 24 SEGRB'J fDldfflLE JPYfA eOMHff OIlMfUB'bS ONLY DOCID 4011853 duced and that little regard will be given toward maintainability adherence to standards and other precepts of good programming practice The extent to which these aspects should affect rules roles and organizational structure cannot be foretold without detailed study Nonetheless it is essential that such areas be addressed rather than the current practice of simply ignoring the problem and letting changes evolve without a concerted plan Summary e must recognize that there has been an irreversible change in the way in which processing is accomplished that continued change is certain and that responsiveness in a purely technological sense is insufficient A flexible plan regularly revisited is needed to ensure that our organizational procedural regulatory and personnel practices are responsive to the changes oooooooooooooooooooooooooooooooo Author's Note This paper was written three years ago prior to the breakup ofthe Soviet Union prior to the recent DDO restructuring prior to the current emphasis on downsizing prior to Paul Strassmann's elM initiative and without knowledge ofthe past three years of technological innovation Nonetheless its basic precepts appear to be nearly as relevant today and the crystal ball of 1989 appears not to have been very murky oooooooooooooooooooooooooooooooo Letter To the Editor other reply to the article Where was the Bogeyman The terrorists do not however speak for the majority of Muslims We need to listen to the voices of the majority-not the over-publicized shouts of the extremists Hysteria over the Islamic threat is grounded in gnorance of the Arab peoples their religion and their culture A close examination of all three In any debate on Islam there are legitimate would reveal a society whose basic beliefs and issues to address We need to understand the values are not dissimilar to our own It would reasons for the spread of Islam and the appeal also highlight the individuality of the Islamic that extremist groups hold for some Muslims sects in the Middle East and their desire to estabIt would be a mistake to allow fear or hatred to sh national identities-factors that make a distort our understanding or to let the actions ihad unlikely of a few blind us to the legitimate concerns and problems of the many Only by objective analyIt may be that the word Islam and terrorism sis and by laying aside our Western biases can are so linked in the minds ofWesterners that we we hope to achieve progress in East-West relaare unable to distinguish between the two There tions s no doubt that terrorists have used and continue to use Islam as a pretext for their actions F8tf6 2nd Issue 1992 CRYPl'OLOG page 25 SBORB'F IM NDLE JPYfA OOMHfl' OUANtfELS OUi P L 86-36 DOCID 4011853 P L ______ lposisAO Missions Functions Organizations Personnel V-VVVJ Questions often arise concerning the classification of NSA's missions functions organizations and personnel The following is a list of items and the classification if any of each FACT OF CLASSIFICATION o NSA's missions ofSIGINT INFOSEC or OPSEC training UNCLASSIFIED o Information revealing the general missions and functions of COMINT activities without revealing specific COMINT techniques procedures or targets UNCLASSIFIED o NSA's organizational designators below key component level FOUO o NSA's use of supercomputers as part ofits mission UNCLASSIFIED o NSA's total budget or individual line items CONFIDENTIAL o Total manpower strength of cryptologic community NSAorSCEs CONFIDENTIAL o Individual job title and description that does not contain classified information requiring classification NSA Reg 10-11 provides unclassified job titles and descriptions UNCLASSIFIED o The statement cleared for TOP SECRET Special Intelligence or cleared TSlSI UNCLASSIFIED o The statement cleared for TOP SECRET indocrinated for CAT ill COMINT c-cco 2nd Issue 1992 CRYPrOLOG page 26 eONFJBBN'HAi a HMfBM VJ A GOMH'n' GUi h8 ON hJPY 86-36 DOCID 4011853 ON THE TAXONOMY OFfHE OYSTEr The body of an oyster Adductor muscle Heart Stomach Hinge Mavth anatomy of the oysler The parts of an oyster shell U bo or beck ----t r_--Hinge oldest pert of shelll P L Inner - - - - f ____---lIP0541 shell layer U Turn the clock back to 1 000 000 B C plus or minus three sigmas Our ancestors had just accumulated enough little gray cells to work with U Just as trees and oysters are related so are substitution systems and codes Superficially they appear to behave in the same way U They looked around and marveled at nature They made note of the cycle of the seasons They made note of the progression of the heavens They made note of the tides U But it wasn't until 980 000 years later that U And they noticed that some matter replicated itself and some did not That was a step in taxonomy animate and inanimate though they might not have used those terms U Then they also noticed that some of the replicable matter moved under its own volition and some did not our ancestors learned about complexities in the oyster that distinguish it mightly from a tree It did take a while for them to realize that it was a difference in kind not in de ee U It's not surprising therefore that it has taken us until now to recognize that substitution systems and codes differ in kind not degree Actually it was known years ago but as that knowledge was based on instinct it was set aside as unscientific BASIC CRYPTOGRAPHY U People move at will on land birds in the air and fish in the sea So therefore they must fall in the same class-animate matter that moves at will Then there is also replicable matter that does not move under its own volition For example trees and oysters Our ancestors noticed that trees are rooted in the earth They noticed that oysters are rooted on rocks under water Therefore trees and oysters must fall in the same class-animate matter that cannot move under its own volition U Let us review cryptographic principles In cryptographic systems there are only two possible operations o substitution replacing one character by one or more characters o transposition rearranging the characters It's possible to apply both to a single message -4C SeO SOwe see that on the surface it seems reasonable to assign codes and ciphers to the same 2nd Issue 1992 CRYPI'OLOG page 27 OONFIB HA' lflb lIlA COMai' - GIWRf 8LS OnLY 86-36 DOCID 4011853 family After all the enciphering process is the same-replacement In a substitution system one plaintext character is replaced by one or more cipher characters The process is similar in a code system The plain text called the meaning in codes is replaced by cipher called a code group Therefore the two enciphering processes are similar U Now let us turn to the deciphering process as cryptanalysis was called in the early days c-eeo The critical factor that distinguishes code from substitution cipher is the matter of the plaintext component And this difference is a difference in kind as you shall see a substitution cipher the possible plaintext equivalents for each cipher component are finite the cipher letter can become one of only so many possible plaintext letters It does not matter whether the plaintext component is only letters or only digits or a mix ofletters and digits or a mix ofletters and digits and punctuation It's a closed system -w gQO This is true whether the substitution o a word treaty o a complete sentence This morning I had a long private conversation with the Prime Minister o a phrase under no circumstance o time 1800 hours o a spell -Y- o a family of verb abstract noun agent noun adjective adverb gerund o a selector take fourth meaning of previous group o an oblique form of noun or verb would have signed o a place name Paris o a title HR the Crown Prince ofRuritania o a ship the gunboat MAGNOLIA o an organization 24th Battalion o deciphering instructions beginning with the next group use the PERSIMMON table o a complete message Attack at Dawn CU So as you can see unlike a substitution system that consists of a finite plaintext population in a code system the plaintext possibilities are virtually infinite For that reason the meanings cannot be recovered by purely statistical means though statistics plays a part in the recovery of the meanings system is monoalphabetic or polyalphabetic whether it is based on a hand system or a machine system or even whether there are variants to suppress the plaintext frequencies The possibilities for plaintext values are finite and known If the enciphering process is complex-using shift registers or true one-time pad for examplerecovering the plain text correctly may be exceedingly difficult or even impossible nevertheless the cryptanalyst is confronting a system with a finite plain text population The plain text must be one of the above -CC CeS By contrast in a code system the possible number of plaintext equivalents known as the meaning is very very large for the plain text element may consist of a letter digit word phrase or sentence or even a complete message Therefore the possible plaintext values are for practical purposes infinite and unpredictable as well The nature of code is its high compressibility For example the plaintext component the meaning represented by a single ciphertext element the code group may be 2nd Issue 1992 o CRYPl'OLOG o page 28 CeNt IDmft'IM IIMfflLB vtA eOMHff CIIMf fffiLS OtfilI EO 1 4 c P L 86-36 DOCID 4011853 o study past traffic and combine frequent cliches such as Reference your message Number 364 as one group Except you have unexpected subjects like Boris and Gorby and narcotics interdiction and Bosnia and Hercegovina and GATr U And then there is the famous case ofa French World War I compiler who organized the codebook meanings by whether or not the nouns were derived from verbs-and this for a trench code THE FREE-WHEELING CODE CLERK U To compound the difficulty of recovering the meanings of code groups any rule the compiler sets for encoding plain text the code clerk may undo Take for example a codebook that is in caption order-when the meaning is filed under the most important word in a phrase If the caption is 1234 TREATY thereafter you may see phrases incorporating TREATY WHY IT MATTERS c gOer In the past we used to peruse hard copies of the system descriptions to learn about the past cryptographic habits of our targets But now we're into information retrieval Ifwe're looking for a specific type of system we must type in the appropriate term to access the database And if we're looking for code we will not find it for codes are tagged as substitution systems It's a loss of information that once was known m How did this state of affairs come to pass I believe it was an excess of zeal in taxonomy and misguided at that with a large dollop ofpedantry laced in U And this ties is in with recent practice Consider the term pentagraphic used of a code or cipher group consisting offive characters In parallel construction one of twelve characters would be called dodecagraphic We have puzzles enough set by our targets without concocting our own and confusing cryptanalysts who have not had the benefit of a classical education m 1245 treaty of peace 1267 commercial treaty 1278 to sign a treaty 1289 the Queen signed the treaty m The code clerk who is not familiar with the codebook very likely will encode the equivalent of 1289 as three groups or even four It's time we dropped that clunky usage and reverted to that of an earlier more elegant time when a dodecagraphic group would be written as 12-character or -digit- or -letter as the case may be The important word in the compound-the length-is instantly recognized And that's as it should be Let our documentation be self-evident Eschew polysyllabics m And on that note let me invite everyone to drop forthwith the indiscriminant use of Queen superenciphered Super means on top of A code Sign past tense is enciphered using an additive let's say Treaty Period Only if there is still another encipherment over the first should the term be m And if the code clerk is looking for the number superenciphered 1800 to measure distance 1800 kilometers and o A 4-letter code enciphered with polyalphabetic runs into the military time 1800 hours first that's substitution what will be used The decoding clerk at the other o A 4-letter code enciphered with polyalphabetic substitutioll and superenciphered ore reend knows to decode it as 1800 kilometers enciphered by transposition THE SYLLABARY m Isn't that neater m Then we have the case of the syllabary a gray em Let's get rid of the overburden ofpolysylarea these like codes should be catalogued lables separately as syllabaries not as substitution systems for syllabary systems are open-ended and U Keep it lean mean and clean as unpredictable as codes 2nd Issue 1992 o CRYPl'OLOG page 29 CONFIDE J PIMJ DocrD 4011853 l L Agency too long and he assigned me to work with Glenn Stahly in G4 By then Lyndon Johnson was president and the world situation continued to be bleak 86-36 o afet ictory POU01 After about a year in G4 Mr Raven told me I had a choice stay in PI and move to A5 or if I wanted to stay in G4 I would have to transfer there Since I was in the middle of a major project I opted to stay in G4 and my association with PI became a lot less for the next 20 years 1l -_---- ---- - - -1z ' Formerfy Chief Pi CRYPrOLOG Pu6 islier 'Emeritus CPOUO For a while Ted Leahy helped me on my projects Then he stopped coming by and one day seeing him in the hall I asked him where he had been He said he was helping people who needed the help more that I didn't need him anymore This was not exactly true but it does reflect the PI spirit and what it represented Address at a luncheon on 31 January 1992 to commemorate PI's 30 years of service U We were all saddened by the recent death of i OUO In the meantime George Vergine and Art Levenson-neither of whom unfortunately could be here with us today-were the 3rd and 4th chiefs of PI for about 7 years between 1966 and 1973 Nixon was President and Watergate consumed the nation Bill Lutwiniak but we know he would have wanted us to commemorate the occasion for which we are gathered So today we dedicate this proceeding to his memory and I ask that you join me in a moment of silence gc PI came into being as nearly as I can roUO I had gone to GCHQ in 1971 on a G4 billet which was created because PI would not let me be on their billet which at the time was reserved for CMP graduates I retu ed to the states in the summer of'74 and my family and I watched the hearings on Nixon from our motel room on Route 3 in Bowie and later listened on the radio to Nixon resigning while we were sitting in our empty house awaiting our furniture ascertain on Flag Day in 1961 as part of an NSA reorganization John Kennedy had been in office only a few months and there was a mood of excitement in the country tempered by an apprehension caused by the cold war and the potential Soviet menace After all part of the reason we reorganized was because of Martin and Mitchell No more would we be ADVA ALLO GENS or MPRO but we would be A Group or B Group or G Group How mundane POUO At that time Bill Lutwiniak was in the second of his eight years as Chief of Pl This was POUO I was a young analyst in GENS which the longest time for any of the chiefs of Pl I hope some of the speakers will talk about those had no direct dealings with PI during all the ye between1965andW8 became A4 and very little changed for me I had never even heard of Pl here with us today was appointed the first chief I hOiSyearsbeca JI'El y h a e 'I ww he helped G4 get a computer He was succeeded by Frank Raven Once of the things that happened while Frank was chief of PI was the start of the cryptomath program in 1963 just as I was signing up for a new job in PI I tried to get into the CMP but Frank said I had been at the except through the CMP I FOUO When Bil1J etired he assucceeded by lasfhe sixth Chief of Pl Ford Carter and now Reagan occupied the White House and the country's cold war continued without much change 2nd Issue 1992 o CRYPl'OLOG page 30 PeR OFPIGIAL USB 0NI 0JPY P L 86-36 DOCID 4011853 _During the 80's there was tremen- P L 86 '36 oooooooooooooooooooooooooo - 1 L VUVj dous growth in the Agency both in dollar and manpower allocations I was still in G4 and experienced this enormous influx of new people into our already cramped quarters It was JJ yerth eless an exciting time and we were accomplishhJ i aiot of riririicles as 1 liked to say So it was with some reluctance and misgivings that I accepted the job as the 7th Chief of PI in 1987 Soon after George Bush became the 7th president in the same 30 year period I am not trying to draw a parallel between anyone chief of PI or any president just that there were the same number I Ifflfflt'ff - J Now the world scene was radically changing and austere times were in store for almost everyone The demands for restructuring were everywhere and with the recent collapse of the Soviet Union the reorganization which had been planned for two years raced towards completion As the restructuring took shape it became clear that a PI-like organization was not appropriate in a functionally aligned DDO 1j Q Now we enter a new era and from my perspective it will be exciting About half of the old PI will be going to the new Z Group with some of the functions and individuals dispersed The CMP will remain intact and unchanged in Z5 which I am heading Other parts of PI will survive in various other groups and the new operations staff I J J It is wonderful to see so many people both present employees and retired Pl'ers who have come out today to pay tribute to this wonderful organization It is now time to hear from some of the other people who made PI what it was in the past FOUO In closing while we salute the old I leave you with the thought that it is people who make institutions great We have not lost that magic ingredient with the demise of PI so let's go forward and remember o oo o oo oo ooo e o o ' o To the Editor oo o o o o o The Government Code and Cypher School was formed under the control of Cdr A G Denniston In 1938 Denniston was preparing for war and encouraged the Foreign Office to buy the property at Blechtley Park It had o several advantages 45 miles from London and 50 miles from Oxford and Cambridge where Denniston hoped to recruit most of the o mathematical and linguistic specialists When o o war broke out everything was in order and oo scholars from the universities were rapidly recruited For a short while the staff could be accomodated in the large country house but o soon it became necessary to build huts where various sections were housed-Hut 5 for the Italian Army cyphers and others for weather and similar subjects Many of the sections and huts have long been identified but there were a few where top secret work was done so secret that it was not until 1973 that Group Captain Wmtherbotham made it known that teams at o Bletchley had broken the German High Command cyphers-the Enigma The chief hut for this work was Hut 6 and to this day it still stands But Bletchley Park is slated for development and the remaining huts will be bulldozed o o As many of those who worked in Hut 6 and elsewhere in Bletchley have objected to the disappearance of Hut 6 the Bletchley Park Trust has been formed Already much has been done It is likely that the Government will halt the demolition o o The Trust would like to hear from those who have an interest The Trust may be reached at Suite 8 Denbigh House Denbigh Road Bletchley Milton Keynes Bucks MKI lYF The chief executive is Ted Enever D These are the good old days William Filby former member GC CS oooooooooooooooooooooooooooooooooo 2nd Issue 1992 CRYPI'OLOG page 31 FOR OFFIGIAI ygg 0 Il LJPY DOCID 4011853 A Visit to TIME P L 86-36 Reported by IL - lp05 et al For some time agency reporting specialists have suggested adapting the organizational structure and publication techniques used by print journalism In an effort to see which of these techniques can be used in agency reporting a group of senior reporters visited TIME magazine on 17 and 18 June 1992 L 86-36 P 'sons Visiting I We were very openly and hospitably received by senior officials of TIME Inc the TIME Magazine Managing Editor the Deputy Managing Editor the Editorial Operations Director two Senior Editors the Picture Editor and others From the very first TIME insisted that we spend two days with them They wanted us to see the progression of articles through the editorial process from one day to another On both days we were privileged to sit in on the 1000 a m editorial session that the Managing Editor held with his key staff members to discuss ideas for the weekly issue of TIME Magazine the status of articles already underway etc At these one-hour meetings the senior leaders of TIME in concert with key staff members decide what story lines will be pursued by the magazine This is an intentionally free give-and-take regarding both domestic and international topics The morning session is followed by a series of additional section-level editorial meetings It is at these meetings that TIME managers direct the reporting effort channeling and focusing the various story lines Once the actual article is written there are quality checks throughout the production process by researchers fact checkers and copy editors In separate sessions the Managing Editor and his deputy told us that what we witnessed in these editorial sessions was not always the procedure for TIME Not too many years ago it was a top-down structure with the top telling the bottom what articles should be written and even at times the ideological thrust of those articles This was particularly true when Henry Luce owned the magazine Now stress is put on bottom-up collegial communications This has not been an easy process for TIME and both the Managing Editor and the deputy emphasized that one has to have patience in implementing such a process They told us that some managers and employees feel more comfortable with a top-down structure There is still some hesitancy among the staff to be openly contributory But they say that the bottom-up approach is definitely the right one because for the magazine to compete successfully these days it has to have the fresh ideas and the proprietary involvement ofits employees We were struck by the similarity of the problems facing TIME and our own evolution toward a lower-echelon decision-making and employee-involved Operations 2nd Issue 1992 CRYPI'OLOG page 32 POR OPPlOFA D YBB 9HlX DOCID 4011853 Directorate and Agency TIME has about 400 persons to carry out its weekly publication this does not include another about 200 who are involved in the business endsubscription handling etc Junior reporters start at about $35 000 There aree 28 bureaus with about 55 correspondents outside New York Henry Cabot Lodge was their first correspondent TIME does not have a formal training program for their reporters They expect individuals whom they hire to already be trained journalists with experience gained elsewhere Their staffis composed of experienced individuals trained in the journalistic style of reporting They are required to follow the TIME style of writing The deputy editor kindly gave us a few copies of Writing for TIME which is their internal style manual There is a Wednesday afternoon critique session but there are many quality checks all along the line throughout the production process Researchers are used to assure that the facts are right in an article TIME does not have a formal training program except for training some persons on new equipment received They do not train their reporters and writers The deputy manager told us that it is still a buyer's market with TIME having their pick of young people anxious to come to work for TIME He said this stems from the interest in journalism generated by the Watergate case Their final product is disseminated to numerous CONUS e g Philadelphia Atlanta Wisconsin Illinois Dallas Los Angeles San Francisco and four overseas locations for printing The international edition is different from their domestic edition and ads are matched to the various regions of the world We were told that within the U S even individual subscribers receive different ads in their issue of TIME Depending upon facts related to a specific subscriber a computer determines what ads to include in a particular issue things like age sex ZIP code salaries etc are used as the basis for determining the make-up of the individual copy This amazed all of us and we were wondering if there were some application of this software that we could apply here TIME offered to discuss it further with us if we desired Most of their non-management employees belong to a union the Newspaper Guild so like those of us in the federal service TIME is not as free to hire and fire as one might think we were told A large number of their employees work three days a week twelve hours a day and like it This has to do with the natural production pressure that builds on Thursday and does not end until about 0300 Sunday morning when everything has to be electronically sent to the printing and distribution points About 80-90% of the work is done on Thursday and Friday with the crescendo coming between 3 pm Saturday and 3 am Sunday The goal is to have the magazine on the street early Monday morning When asked why Monday they told us that this is just the way it has evolved over the years Long pressure-filled hours are normal during this time About 80% of the foreign bureau personnel are U S nationals with foreigners hired to handle the clerical and administrative matters As for technology the Managing Editor told us that the TIME goal is to have Macintosh terminals for everyone with the ability to'do article preparation on the terminal and transmit to a large number of other terminals At present they use a variety of computer terminals Eventually all the Macintoshes will be part of a single network so that article preparation and transfer of material can easily be accomplished from reporter to editor to photo desk and so on They plan to abandon their present ATEX system a company name we were told which does not expand We were told that a company named Stratus was great for preventing any down time in their vital operations As in nearly all news magazines photo and graphic design playa large role in each issue Once a story line is decided upon photo taking and selection is usually done before an article is completed TIME has independent photo journalists under contract as well as access to an electronically delivered photo service similar to the 2nd Issue 1992 o CRYPI'OLOG o page 33 FOR OFFlOItH USB 8Nt y DOCID 4011853 AP or Reuters wire TIME is implementing a new high-resolution National Digital system which will enable them to quickly receive and incorporate photos into their magazine The art director decides the layout Three or four of the picture editors are also photographers Magazine covers are planned as long as eight weeks in advance unless some special event like the Los Angeles riots occurs We met the individual who is responsible for TIME covers Just 12 years ago he said TIME was a black-and-white publication In 1981 they went to six pages of color and finally in 1984 went to all color Actual readership is a lot higher than the sales number because they know from surveys that copies are read by approximately four persons beyond the person to whom the magazine is addressed They estimate a 30-million readership weekly Every article in TIME is carefully read by company lawyers to see if the magazine could be liable for something written The managing editor estimated that the magazine has about one serious suit per year We will be reviewing our trip to TIME and to other publishers to determine what we might use to improve our reporting at NSA We are particularly interested in generating greater interest in the editorial-board approach to production Electronic Publishing QUiz E iU1 1 U12 f1Er Il lm 1 AFP 11 JES 2 APA 12 MICR 3 CALS 13 OCR 4 CRT 14 OGL 5 DJDE 15 PDL 6 DOS 16 PSF 7 DPI 17 RFT 8 FDL 18 SGML 9 GML 19 WYSIWYG 10 JCL 20 XES Answers on page 14 2nd Issue 1992 CRYPI'OLOG page 34 FOR OPPICtkL USE ONLY DOCID I 4 01 1 B 53 N- PHl Bfl -H'li_ l 0 8_ - Conference Report ' - o oo I U 1992 National Association ofBroadcasters Convention and Exhibition The Annual Broadcast Engineering Conference and HDTV World '92 Conference and Exhibition held concurrently in Las Vegas Nevada 12-16 April 1992 P L 86 36 itepoTtedby I IP054 S IA November 1991 Information gleaned from NAB seminars and exhibitions will assist me in developing and maintaining a state-of-the-art understanding offacility planning and design facility management reporting production techniques storing and retrieval techniques archiving techniques and technical equipment integration e primary purpose of this trip was to provide me with the opportunity to assess current and emerging broadcast video teleconferencing and multimedia technologies associated with the Operations Directorate dissemination initiatives d other associated publication projects I Second the trip provided the opportunity to assess new broadcast related digital encryption and laser technologies and continued escalating foreign interests in visual information production hardware and distribution technologies U P L 8 6- 3 6 CONFERENCE OVERVIEW This annual convention and exhibition continues to be the largest ofits kind in the world and continues to grow with a record number of over 52 700 registered attendees and exhibitors and a record international attendance of more than 8 651 International attendance reflected an increase of over 20% from NAB '91 New technologies software developments recording formats and multimedia applications continue to be exhibited U CONFERENCE SEMINARS JQ-CC91- Third the knowledge gained from this trip will further assist the Operations organizations and Agency support organizations in developing new intelligence dissemination opportunities in video and multimedia technologies and aid in the assessments affront-end SIGINT collection opportunities fi'OUO Fourth the NAB provides a unique assessment opportunity for those Agency employees involved in implementing the NSAlCSS Visual Information Services Working Group Report of NAB seminars were divided into Engineering Radio Television and HDTV categories In an effort to get a flavor of diversified areas ofinterest I attended the following seminars U Engineering Sessions o Interactive Video The Birth of an Industry TV Answer Reston VA o The In-Touch TV System A Technology Description Interactive Systems Beaverton OR 2nd Issue 1992 o CRYPl'OLOG o page 35 C6NPfBBN'HJ L IIMfflLE Vi A eOMT1ff eH2 it NM 8 ONLY DOCID 4011853 OOPWIBHN'HMs o Production Processes for Interactive Television Interactive Network Inc Mountain View CA U Jean Gard Entertainment Industry Segment Manager for Digital Equipment Corporation DEC o Pay Per View-Video on Demand Jerrold Communications Hatboro PA U Comments Ms Gard shared insights for o Update on New Interactive Television Applications ofT-Net Radio Telecom and Technology Inc Perris CA Video Production Post Production o Driving Towards PC-Based Post Production Autodesk Inc Sausalito CA o Bridging Computer Graphics and High Quality Video Tektronix Video Products Operation Wilsonville OR Television Sessions o Controlling Our Future Broadcasters Cable Telco o All-Industry Luncheon with addresses by Eddie Fritts NAB President CEO Topic Strategic Planning and Innovation The Armed Forces Radio Television Service was honored and by Former President Ronald Reagan Distinguished Service Award Recipient Ironically I witnessed a second attack on this President at a Hilton Hotel this time as he delivered his NAB keynote address o Law and Regulation Conference Luncheon Alfred C Sikes Chairman of the FCC Topic The Future of Broadcasting Merging of Technologies Need for Adaptability future business areas and related technologies that DEC is actively pursuing related to multimedia technologies and services U Geoffrey S Roman Vice President Technol- ogy and New Business Development for Jerrold Communications a division of General Instrument Corporation Comments Mr Roman is a former MITRE principal In his public NAB engineering presentation Roman mentioned that TCI and US West are using cable distribution technologies Roman predicted major domestic US telecommunications implications in six to nine months Although Roman did not go into detail with this statement I believe Roman alluded to a potential new business enterprise for US Cable Companies in the telephone and data business in direct competition with the Baby Bell companies After his formal talk he gave an interactive video engineering presentation FOUO Representatives from the National Cryptologic School's Visual Information Services Facility E23 Research and Engine ring RB and the DIA J2 Defense Intelligence Network DIN executive production staff il UO Comments This provided an excellent opportunity for informal management and technology networking between NSAlCSS elements as well as interagency liaisons CONFERENCE ASSESSMENTS PERSONS VISITED U David Lyon Chairman CEO Basys Group U Comments Basys Automation currently has a system installed in the T5 Media Center area Basys CEO provided insights for future business areas and technologies that the company is actively pursuing in information management systems and services related to video production databases library and archive and multimedia technologies Telecommunications Policy and New Technologies m The FCC through the dynamic leadership of its Chairman and proactive Commissioners is continuing to drive the US into world leadership in digital communications Featured at this year's NAB was the first over-the-air broadcast of digital HDTV on experimental TV Channel 15 as authorized by 'the FCC For those of you who have not seen demonstrations ofHDTV the image quality 2nd Issue 1992 CRYPl'OLOG page 36 GONFm IWIDLB VfA OOMRW OHMftfBL6 OfHrY DOC I D --4-Q l l-B will knock your socks off It will revolutionize the broadcast and entertainment industries not to mention what it will do to the telecommunications industry because of digital technology video compression techniques and non-video digital services The FCC anticipates announcement of a US digital TV broadcast standard in 1993 Shortly thereafter US broadcasters are positioning to rapidly enter this new market opportunity Digital AM and FM radio has already been developed Digital data services are positioning for the radio market and new receivers will soon be installed in US-sold autos to provide alphanumeric information POUO Many international NAB attendees continue to assess and procure US developed video production and distribution technologies for a variety of entertainment business government and military applications The NAB has reached such an international scope that the US Department of Commerce and Foreign Money Exchanges maintain sizable on-site convention presence Hardware Video Production U Off-the-shelf hardware technologies continue to develop for both linear and non-linear video production Linear technology allows only ordered video production i e item tn then #2 then #3 and so on Non-linear technology allows immediate random access broadcast quality video production much like a word processor Analog motion video analog still images and analog voice narrations are first digitized in non-linear editing and then randomly manipulated to produce a final video product You can cut paste and transition combinations of both audio and video using computer technology control software an icon control screen and a special-purpose mouse developed to expedite video productions lieve non-linear video production is the way to go in the future However because this developing technology is expensive I recommend using linear video production technology until the market settles a bit and there is a wider selection ofvendors and established vendor standards and interfaces Video Recording U New tapeless video production formats in- clude rewriteable optical disc and rewriteable magneto-optical disc Manufacturers exhibiting this equipment at the NAB included Panasonic Pioneer and others These video recording formats support both linear and non-linear video production storage and retrieval and archiving requirements These optical technologies are currently limited to approximately 55 minutes to 2 hours of record playback time depending on the manufacturer as well as analog digital and compression formats The optical recording media is quite expensive when compared to videotape Example a one-hour Beta-CAM SP broadcast quality videotape costs about $25 versus a one to two hour magneto-optical disc which costs about $1200 Duplication U A high speed VHS duplica ion system was exhibited for the first time by The Sony Corporation Other manufacturers including Sony Panasonic and others exhibited a number of realtime duplication systems some with auto cassette feeds Teleconferencing U Fully assembled teleconferencing units and integrated teleconferencing technologies were not featured at this convention However a number of manufacturers and system integrators exhibiting at the NAB offered such systems U Television cameras and video recorder devices are both moving to digital technologies with approximately 1000 lines of resolution As these digital devices develop they will expedite the nonlinear video production process because raw materials for video and multimedia productions will already be in a digitized format I certainly be- Software O Software development is not currently a strong point for video hardware producers Although a diversity of off-the-shelf hardware solutions exist for video SIGINT production storage and retrieval of motion video and still images 2nd Issue 1992 CRYPl'OLOG page 37 G9NFI9BN'HAL IWR1L VJ A GOMlN'J' GIMN 'f ELB O H y DOCID 4011853 and long term archiving at this point I anticipate some limited software machine control development with a hardware vendor or related broadcast software developer to maximize automation capabilities of existing commercially available off-theshelf hardware systems POUO There is currently no software standard for desktop video production As in word processing a number of software suppliers provide unique applications programs Important to note is the fact that mouse hardware technology is being reworked for video production Video editors using computer mouse technology for video production are not happy with current mouse designs for speed hand-eye coordination and physical durability POUO Several vendors are working toward software that will combine scripting database searches production and machine automation control This software development will greatly increase the flexibility of video production facilities Encryption A number of companies are currently manufacturing and selling commercially available offthe-shelf analog and digital video encryption devices These companies include Leitch Macrovision Scientific Atlanta and Sony Sony among others now offers an international teleconferencing network Transmission U Canon Broadcast Equipment Division intro- duced a new Optical Beam Communications System called the Canobeam The NAB Broadcasters Daily News reports that this laser devise using a high-speed modulated optical beam can transmit up to eight video signals and 18 audio signal channels in both directions Canobeam transmits wide-band signals up to 500 MHz Data can be sent at speeds up to 156 Mbps Transmission distance can range up to 2 kIn and digital data transmission distance can be increased without signal degradation by using the Canon 3R function relay Can you imagine this type of technology in the next war CONCLUSIONS AND RECOMMENDATIONS EO 1 4 c P L 86-36 c J The NAB and the US Department ofCC rii merce maintain records of overt convention participation by name and country Commercially available off-the-shelf technologies supporting visual information M production teleconferencing and encryption continue to be exported to foreign countries for diversified multimedia information flow capabilities The Operations Directorate should continue representation at conferences conventions and expositions such as the NAB which will enhance the knowledge base required for the management production engineering exploitation teleconferencing and multimedia development of cryptologic visual information M activities U In order to adequately cover the growing magnitude of the NAB Convention Exhibition and Engineering Conference I recommend expanded Operations Directorate representation and I encourage representation of other Key Components at future conferences in support of expanding requirements for Agency Visual Services U New for 1993 the NAB announced a premiere conference and exhibition called Multimedia World Merging Video Audio and Computers Beginning next year this new annual event will be held concurrently with the NAB Convention and will provide a new forum for post-production video business video computer professionals and broadcasters 2nd Issue 1992 o CRYPI'OLOG o page 38 C6N Fft Hi'1c'NBYS VI f OOMRW OItMHffiLS OffIll P L 86-36 EO 1 4 c DOCID 4011853 c l29I In early November 1977 the political situation between Ethiopia and Somalia had reached a crisis At the same time the American Embassy reported that Ethiopia was considering breaking relations with the United States though Ethiopia needed the food and equipment that the US had been providing 2nd Issue 1992 CRYPTOLOG page 39 P L 86-36 EO 1 4 c 'FOP 8B JRB'F UMBH iii EO 1 4 c P L 86-36 DocrD 4011853 Editorial Towards a Viable Thch Track act as technical mentors to interns and to seasoned analysts as well who develop courses Towards a viable tech track I hereby offer some who teach and lecture who develop the professuggestions sional examinations and grade them who write philosophical papers on futuristics in their discio Establish a program ofpost-professionalization pline who compile lexicons and working aids of rotational tours all kinds They provide the hard core oftechnical and historical continuity o Sharply increase the proportion of promotions awarded to tech trackers It is the traditional o Honor the technical types who manage peopleway the more opportunities for advancement the less projects rather than people Let's take a more enticing the prospect for remaining a tech look at our own resources in these tight budget tracker times We are wont to make heavy weather about managing a long-term big-bucks project o Add to all promotion boards a voting tech But we already know from the write ups of the tracker who will look out for other such productivity awards that we have gifted people o Convert some of the SCE slots to SLEs We will who can knock a complex project off in no time be needing even more technical experts in the flat and on a shoestring at that Reward them future suitably To whom it may concern o Be generous about sending tech trackers to professional conferences The insights that can be gained are invaluable In our closed society a breath of outside air is absolutely necessary What's more the attendees can share the wealth by writing up a conference report for CRYPTOLOG o Allow tech trackers time to document their projects at their conclusion Documentation should be an integral part of any professional's duties o Do not overlook the technical staffers These are the people who fight the dragons that besiege our gates for secrets We depend upon them to devise an unimpeachable defense for retaining classification o Be very very careful about overriding an evaluation of a tech tracker that is done by a technical board IT you think it necessary to do so explain your reasons to the technical advisors o Last but not least about TQM cumulatively o Give due credit to people who pass on the torch our professional cadres have a lot of smarts These are the people who document their knowlimagination and ingenuity Sound them out edge and techniques for coming generations who Use them to solve problems 2nd Issue 1992 CRYPI'OLOG page 40 POR 9PPIOfAL USH 6NLJPY DOCID 4 p11853 CRYPTOLOG Editorial Policy CRYPTOLOG is a forum for the informal exchange ofinformation by the analytic workforce Criteria for publication are that in the opinion of the reviewers readers will find the article useful or interesting that the facts are accurate that the terminology is correct and appropriate to the discipline Articles may be classified up to and including TSC Technical articles are preferred over non-technical classified over unclassified shorter articles over longer Comments and letters are solicited We invite readers to contribute conference reports and reviews of books articles software and hardware that pertain to our mission or to any of our disciplines Humor is welcome too Please note that while submissions may be published anonymously the identity of the author must be made known to the Editor Unsigned letters and articles are discarded If you are a new author please request Guidelines for CRYPTOLOG Authors How to Submit your Article Back in the days when CRYPTOLOG was prepared on the then state-of-the-art a Selectric typewriter an article might be dashed oft' on the back of a used lunch bag But now we're into automation We appreciate it when authors are too N B If the following instructions are a mysteJY to you please call upon your local ADP support for enlightenment As each organization bas its own policies and as there's a myriad of terminals out there CRYPl'OLOG regrets that it cannot advise you Send two legible hard copies accompanied by a floppy disk or cartridge as described below or use electronic mail In your electronic medium floppy disk cartridge or electronic mail please heed these strictures to avoid extra data prep that will delay publication o do not type your article in capitaIletters o do not right-justify o do not double space between lines o but do double space between paragraphs o do not indent for a new paragraph o but do paragraph classify o do not format an HD floppy as DD or viee-versa-our equipment can't cope o label your floppy or cartridge identify hardware density of medium software o put your name organization building and phone number on the floppy or cartridge The electronic mail address is via PLATFORM cryptlg @ curator or via CLOVER cryptlg @ bloomfield CRYPTOLOG publishes using Macintosh and Xerox Star It can read output from the equipment shown below If you have something else check with the Editor as new conversions are being added SUN 60 or 150 MB cartridge XEROX VP 2 0 2 1 5 1 4 floppy only WANG 3 112 DD disk only Macintosh IBM Compatibles 3 112 DD or lID 5 1 2 DD or 1m ascii only Stand-alone or Alliance Please furnish a copy in TEXT as well as in your software as we may not have all the software upgrades Please furnish a copy in ascii as well as in your software as we may not have all the software upgrades 2nd Issue 1992 o CRYPl'OLOG o page 41 P6B 6FPIetM J USB 9NLJPY DOClD 4011853 -taP-SEEREf- 'fIllS BOEURUl IEN'f EURON'fAINS EUROBEWORB MATERIAL -f9P-SEEREfNOT RELEASABLE TO CONTRACTORS This document is from the holdings of The National Security Archive Suite 701 Gelman Library The George Washington University 2130 H Street NW Washington D C 20037 Phone 202 994-7000 Fax 202 994-7005 nsarchiv@gwu edu