WGBPDVVlDWGBPD11 lrUlWU l5 lDWVU W I5UJ lSlDW 1 Il OOI UllDl f OOUlW I1 LlmlD 4th Issue 1987 1 ROW TO 'fIN HITROUT PLAYING THE GAME o o o o 1 1 LETTERS o o o o o o o o o o o o o 5 10 PROJECT TIPONI o o o o o u'u u u 6 TliJLUPECTSOPOE' r TIONS NMETEOR SCATTER o o N C Gerson o o o o 9 6ULLE'l'INBOARD o o o o o o o o o o o o o o o o o o o o 10 P GBP e d i r y rij TWQTIPS ON HRI'1'ING o o o POINTERSONG o o o o o o o Lei _ o o o o 16 o Ralph Jollensten 17 TRE HERO OF THE SENTENCE GIMME REWRITE o o o o o o o o o o Va BOOK REVIEH MACHINE TRANSLATION L I PROM THE PAST o o o o o o o o o o BUSMAN'S HOLIDAY A BRITISH CIPHER OF 1783 'fIllS BeEURHMHN EUR9N'fllIN8 J09 WQIUJ A gKIAL I 111 s o o o o ----II o 18 19 20 21 o o 22 CLASSifiED 8'1' MSAlCS5M 123-2 DECLASSIFY ON OFigiFlatiAg AgeFlEY'S DeterR'-liFliti9F1 Reqlolired NOT RELEASABLE TO CONTRACTORS Declassified and Approved for Release by NSA on 10-17-2012 pursuant to E O 13526 MDR Case # 547713 DOCID 4010011 Published by P1 Techniques and Standards DESERT ISLAND BOXES Vd XIV No 4 o 4th Issue 1987 PUBI ISi- ER r------- The scenario below was inspired by the program on o 1 aOARDQF EDITORS Editor o o 1 t963-1103 Collection o ' o 963-5877 963-1103 Computer Systemi Cryptanalysis ' ' ' J 963-5238 CryptolinguistiCS J 963-1596 Index o 1 963-5292 963-3456 Information Sciel l e j Information Security o George F Jelen 859-1211 b Intelligence Research 1 963-3845 1 963-3057 Language Mathematics 963-5566 Puzzles o o 1 963-6430 SCience and Technology 1 963-4958 Special Research ' ' ' Vera R Filby 968-8014 Traffic Analysis obertJ Hanyok 963-4351 'J i1lu o 3- '7 1 tl-----l 3 2 To submit articles or letters by mail send to Editor CRYPTOLOG P1 HQ 8A187 If you used a word processor please include the mag card floppy or diskette along with your hard copy with a notation as to what equipment operating system and software you used via PLATFORM mail send to cryptlg@bar1cOs bar-one-c-zero-five note no '0' Always include your full name organization and secure phone also building and room numbers For Change of Address mail name and old and new organizations to Editor CRYPTOLOG P1 HQS 8A187 Please do not phone Ipublic radio Desert Island Disks where guests select five recordings and one other item to while away the hours when they are castaways on a desert island The guests also discuss their choices and philosophize on their careers and on life in general We thought it might be fun for CRYPTOLOG readers to think out a version ofthis fantasy You're shipwrecked on a desert island on your way to a field site You've got everything with you to carry out your mission packed in boxes that are 12x12x15 And thanks to miniaturization a pc and printer fit in one box Fortunately in the desert island there is food shelter clothing and standard supplies aplenty as well as electricity Also your team can communicate with the main frame at HQS via a pc for a total of an hour a day But you will have to take along the references and handydandys in the form of paper floppy disks or other media that you will need to accomplish your mission Each member of your team is allowed to take along six boxes five with materials for the mission that are selected and packed by that member and one for personal use relaxation or otherwise There's a catch however the team may be evacuated to another island in which case the team leader may be ordered by HQS to jettison one mission box per person What would you take with you How would you pack your boxes All your dictionaries say in one box or lesser ones in the jettison-able box And why What would you put in your personal box and why It will be most interesting to read your responses Contents of CRYPTOLOG should not be reproduced or disseminated outside the National Security Agency without the permission of the Publisher Inquiriei regarding reproduction and dissemination should be directed to the Editor All opinions expressed in CRYPTOLOG are those of the authors They do not represent the official views of the National Security Agency Central Security Service FOR OFFICIAL USE ONL 1 I fl - l DOCID 4010011 How to Win Without Playing the Game L 86-36 On International Protocol Standards and Security Features for DoD IR536 U As the International Standards Organization ISO and by extension the American National Standards Institute ANSI continue to specify inter-computer protocol standards for open systems more pressure is being put on the DoD to adopt these standards Upon completion of the study of the National Academy of Engineering comparing DoD's Transmission Control Protocol TCP with ISO's Transport Protocol Class 4 TP4 for use in DoD the Assistant Secretary of Defense for Communications Command Control and Intelligence stated that DoD will transition to the international standards once they have been shown to be as reliable and resilient as current DoD protocol standards NATO has already endorsed for NATO systems the use of protocols developed from these standards U A benefit advanced for adopting such standards is procurement off-the-shelf as part of a vendor's standard product line But some major pitfalls lurk in this strategy or desire which we will address later PITFALLS OF IMPLEMENTING STANDARDS L - I U Significant attention is being given to the ISO reference model and the standard protocol $p cifications that derive from it Manufacturersare selling their future products with the words weare compatible with the ISO standards or our system is compatible with 4th Issue 1987 o CRYPTOLOG o page 1 IIA lBbE SECKE VJ A eOMef'i' CIIAPHH3LS OnLY EO 1 4 c P L 86-36 DOCID 4010011 SBGRBT the ISO reference model Purchasers seem to be of the belief that products purchased with the ISO caveat will allow those equipments to intercommunicate automatically with products with the same caveat from other vendors U However few people at high levels seem to be paying attention to what really matters which is the implemention not specification of the standards The American and European continents have different approaches to implementation Additionally implementations are jealously guarded usually with comments like don't worry how I implement them internally as long as I provide the standard interface you shouldn't care are at best a smoke screen to obscure what they are really doing and at worse misrepresentation of their products U The original intent of the DoD ProtOcol Standardization effort was to preclude the need to translate protocols by requiring all to implement a standard protocol suite The ISO standards are based on the same premise But as things stand now let the buyer beware Each individual vendor's approach needs to be carefully examined IMPLEMENTATION OF PROTOCOLS WITH SECURITY MECHANISMS U The pit becomes larger when security features are included in protocol implemen U I believe that the European market is so tations There are three elements to the small that any single vendor cannot capture a inclusion of security features in protocols the large enough market share to remain profitable It definition of the security mechanisms the is therefore to their advantage to be able to implementation of those mechanisms in a interoperate with each other's equipments In trusted way and the assurance that the the US however the market is sufficiently implementation is in fact correct and safe large that a number of vendors can each Security mechanisms are normally included as capture a profitable market share thus part of the protocol specification However affording less incentive to interoperate with do not address either protocol specifications other vendors' equipments These vendors are trusted implementations or assurance factors adopting the ISO approach but they are The ISO Security Addendum deals primarily doing this either in conjunction with their own with security mechanisms and their placement proprietary approach or by adding proprietary in an overall protocol architecture the ISO functionality which can result in nonReference Model It mentions that trusted interoperability Moreover each vendor functionality is expected whenever security attempts to add its own little enhancements features mechanisms are implemented but it which try to lock in their customers to its doesn't say much more than that And for its equipments and prefers to implement its own purpose it probably shouldn't No documents proprietary protocols and network architecture deal with the assurance issue for security By contrast European vendors predominately mechanisms in protocols though one that does are planning to adopt ISO and to discontinue address assurance is the Orange Book their own proprietary approaches U Some US vendors are adopting the International Standards as a second capability to try to meet the requirements of some of their users who are adamant about the use of commercial standards But what isn't said is how they will go about providing them In addition to incompatible enhancements to the international standards some vendors have privately indicated that they may provide invisible translating gateways between the international standards and their proprietary protocols Effectively they are providing an ISO interface into their proprietary network The interconnection of their machines runs only their proprietary protocols For those US vendors who take one of these approaches statements of compatibility with ISO protocols U It is clear I hope that trusted implementation and assurance issues are in part implementation issues Security mechanisms need to be carefully implemented understood and evaluated Yet it is not clear that vendors fully appreciate the impact of these statements on their freedom over their implementation One can't allow a vendor attitude of leave the implementation to me It is also not clear how much the vendors intend to do about satisfying the need for trusted implementations and satisfaction of assurance requirements U Will off-the-shelf vendor products be good enough It is no inconceivable that the DoD may have to modIfy and or adapt off-the-shelf 4th Issue 1987 o CRYPTOLOG o page 2 SECRE GlIhN I 8 ONI JPY HAl'iDlrl3 lJI 60MDn' DOCID 4010011 SECRET products to meet assurance and security evaluation requirements as well as to adapt them to a trusted implementation This is independent of the presence or absence of adequate security mechanisms in the protocol AUGMENTING THE INTERNATIONAL STANDARDS _ _-- -- ----_ _---II_ In general it is not clear that this approach can be effectively utili2 e d' It is applicable for simple functionality substitutions such as replacing one checksum with another But it breaks down quickly with more complex functionality especially with the more complex counters against denial of service and integrity attacks U If one changes the functionality of a protocol for example a reassembly scheme to protect against playback attacks one actually has defined a new protocol It really doesn't matter how you refer to it the odds are it won't interoperate with its parent An excellent example of this name game taken to an extreme is the International Standard Transport Protocol TP and its versions TPO through TP4 In actuality they are five different protocols that do not interoperate with each other Yet because purists in ISO will allow only one protocol per layer they defined these different protocols as versions of the same protocol Some even claim that the National Bureau of Standard version of TP4 differs from ISO's TP4 Its not clear whether implementations of the two will in fact interoperate standards arena This method of operation can continue We next look at NATO's method of operation They first try to get ISO to adopt their military requirements in the commercial standards But they recogni2 e that not all military requirements are needed by the commercial world and that they won't all be accepted They have Standard Nation Agreements STANAGs which are basically NATO augmentations of commercial standards for NATO's use U U Now the questions become can a similar bureaucratic approach be used by the DoD and can implementations be structured in such a way as to make the approach acceptable to all parties We propose that DoD develop the equivalent of NATO STANAGs for DoD purposes as part of the DoD Protocol Standards Program Where the augmentations or changes deal with sensitive security issues there is nothing to prevent the DoD augmentation from being classified It appears that some bureaucratic mechanisms are already in place with a DoD Protocol Standards Steering Group report identifying the development of Mil-Spec supplements for the commercial ISO protocol definitions DEVELOPING THE IMPLEMENTATION U The more difficult questiop is can off-theshelf implementations be easily modified And how can the new version be structured so that the old version could still be utilized if necessary U We will start by examining what the conditions are if the desire of using off-the-shelf products containing security mechanisms were U However there is an excellent concept in achieved It would have to be a trusted the previous suggestion That is the implementation It would have to be evaluated modification and or augmentation of off-the-shelf and therefore understood by independent protocols We can combine that concept with a parties and it would have to meet some assurance level of correct and safe operation few others to define one possible approach of utilizing international standard protocols to the All of these criteria argue that the implementation would have to be well maximum extent possible without active involvement in the standards arena structured most likely in a modular way We begin by first recognizing that any protocol functionality of a non-sensitive and unclassified nature can be presented and argued by the National Bureau of Standards They are the Government's agent in the commercial U EO 1 4 c P L 86-36 U Our approach where a DoD product is derived from an augmented off-the-shelf product would still require it be modular But it would not necessarily require that the offthe-shelf implementation be a trusted one Additional security functionality modules could 4th Issue 1987 o CRYPTOLOG o page 3 SECRECf HANBLE Vi A eOMHiCf OW PfPfBJ 8 OPlJ JPY I DOClD 4010011 SEORE'f' be added and calls at the appropriate places in the off-the-shelf version could be inserted Where functionality has been changed the module calls in the off-the-shelf version would be changed to call the new modules one could define a different header format from the basic one and distinguish between them through the use of a protocol version number in the header CONCLUDING REMARKS U Alternatively a conditional test of whether or not one was operating in secure mode could be added with a call to the new module This would allow continued operation of the original product as well Using the terms of protocol implementers the security functionality would be viewed as optional functionality which is not invoked for nonsecure operation but which would always be invoked for secure operation This approach also allows for a more complex implementation whereby the security system may back off from some of the features provided if environmental conditions warrant it and if policy allows it An example of a security back-ofT is BLACKER's Emergency Mode This is not to suggest that back-off be a user-selectable option U The only modules that would need to be protected and evaluated would be the new ones Thus a trusted computing base could be designed around only the new modules leaving the rest of the protocol to run as untrusted code The trusted computing base would also have to ensure that the secure functionality is invoked for secure operation We believe this to be one approach which would satisfy the contradictory desires to use off-the-shelf commercial protocols as much as possible and still prevent transfer of security technology potentially inimical to DDO's mission This approach does not require active participation in the commercial standards world We take what they produce and modify it appropriately through the DoD Protocol Standards Program for the specifications and through vendors or cleared contractors for the implementations Contractors and vendors could be restricted as to whom they provide the DoD secure implementations U What is required of the international standards is only an extensible header and a protocol version number in the header What is required of the implementation of the standard would be modularity U It has also been argued that the US commercial world is also in need of security protection This is a major argument for NSA's active involvement in the international standards process By using the above approach commercial vendors could be provided the security modules DoD specifications or implementations under a license arrangement which could restrict dissemination of the DoDsecured product in any manner DoD required It could be restricted to continental US use to US-owned companies etc This approach is analogous to the COMSEC Commercial Endorsement Program licensing arrangement for encryption implementations U One must recognize the performance risk with modularly implemented protocols The overhead associated with module calls and returns can be significant and may have a negative impact on the ability of the implementation to meet high speed performance requirements Careful design in this area is required A proof of concept demonstration in this area would be reassuring C324 is addressing the issue of protocol implementations U The serious technical implementation in ADA ADA alone forces a high degree of challenges of this approach are a modular modularity in an implementation implementation which would be required anyway for a trusted implementation with U Of course one does not change the functionality of a protocol without changing the adequate performance the manner with which the security modules are integrated into the headers If one adds functionality to the basic commercial product and the ability to run an protocol then the header must be extensible ofT-the-shelf implementation as untrusted code with option fields being used for the new functionality If a functionality change requires on a trusted computing base One conjectured a header field change two approaches could be approach for doing this utilizing ADA has already been raised used One is to use the option field approach where the presence of the option field overrides the correspondent field in the basic header Or 4th Issue 1987 o CRYPTOLOG o page 4 SECRE'f' HAUDLE YrA SOMeR' OII-A l lEbS Q bJPY DOCID 4010011 I SHORB U This approach does not satisfy the desire by some defense personnel to actively participate in the international and national commercial standards arenas for security enhancements Some seem to have confused the means with the ends 14 Private conversation with vendor personnel at the National Academy of Engineering Transport Protocol Study which resulted in 3 15 DCAlDCECIR130 letter dated 21 Aug 86 subj Record of the 27th Meeting of the Protocol Standards Steering Group PSSG 24 June 1986p L 86_36 ACKNOWLEDGMENTS FOUO I wish to than - JC322 an jR536 for their helpful comments on an earlier draft r - - - _ _ _ _ _ _ _ _ _ _ _ SOURCES AND REFERENCES 1 ISO Draft Standard 7498 Open Systems Interconnection OSI Protocol Reference Model 2 ISO 7498 Part 2 Draft Proposal Security Addendum to the OSI Reference Model dated Nov 85 3 Transport Protocol for De artment of Defense Data Networks National Research Council Committee on Computer-Computer Communication Protocols --- _ 11 To the EdilDr 1 - I U As I walked into the office early this jmorning and saw our latest intern hard at jwork these thought flooded my mind With 6 MIR serial C3-160-82 dated 2 Nov 82 subj National and International Standards DoD Use jonly one or two exceptions the 15 or so interns the division has enjoyed in the last 3 Thereof Network Security Issues and Classification years have been super performers We've given them responsible jobs and challenged 7 Q41 Memorandum dated 3 Jul86 serial Q4Ithem to excel They have taken up the 637-86 subj Final review of Network Security jchallenge and produced results far above our Classification Guidelines jhigh expectations 8 MIR serial C32-003-84 dated 18 Jan 84 subj Status for NSA Response to USD R E CCCI I U I've heard expressions of concern for this on Use of Commerclal Standard Protocols in jAgency's eroding experience base I've also DoD jobserved that long experience per se isn't 9 DIRNSA letter N1743 dated 18 Oct 84 jalways relevant and is sometimes an impedisubj Data Communications Protocols ment to solving current problems If the 10 Department Defense Trusted Computer interns we've had are representative of the System Evaluation Criteria National Computer intern popiulation as a whole and I have no Security Center Ireason to doubt that they are then the jAgency's future is in quite capable hands 11 Q41 MIR no serial dated 6 Aug 86 subj jThese young folks may not always do things OSD Initiative re Participation in Standardsjthe way we old-timers would and that may be making Bodies good 12 C Memorandum C-301-84 dated 31 Aug I 84 subj Data Communication Protocol The j ADCS contribution to 9 5 DDO contribution to 9 3-page paper undated and untitled or In6 L _ _ _ _ _ _ _ _ _ _ _ _ _ _ 13 C Memorandum C-367-82 dated 9 Nov 82 subj Classification Guide for Network Security EO 1 4 c P L 86-36 P L 4th Issue 1987 o CRYPTOLOG o page 5 SECRE'f' IIMlBLEl VIA 60MRW 6IIMlNELS O lLY 86-36 I DOCID E401'OQ ll P L - 86-36 eaNFfBEN'ffAL Project TIPONI P L Description of the Task 86-36 U We also require the capability to optically scan hardcopy data in any native script for input to the workstation in the native script EO 1 4 c P L 86-36 4th Issue 1987 o CRYPTOLOG o page 6 IJ MfB eONIi'16EN'ffAL V1A OOMHR OIIAl'HfElLS 01' LY D9 6-3 010011 G9NFIBBN'FIAL SelectiftgR portableMessages U As he works the linguist may wish to jot down notes on either the message text or the translation The notes should be clearly distinguishable from the text itself and should be deleted easily when the translation or report is complete if desired For example the linguist may wish to send notes along with the document to another linguist noting an area with which he had difficulty and where he requires assistance POUO During the process of preparing a message for publication the linguist needs to consult on-line working aids such as gazeteers dictionaries special purpose glossaries databases etc to obtain further information about a name location or another message referenced in the message in question These on-line tools should be easy to reference and easy to display in a small window where it does not obscure the rest of the text During the process of preparing the report the linguist may add footnotes to the end-product The footnotes should be automatically placed at the end of the document and in the appropriate format In checking his work the linguist may wish to examine the footnote at the same time that he is looking at the text which is footnoted POUO The Translation Process After the translator has completed a report or translation he prepares a cover sheet for the end-product and sends both to a senior linguist or checker who verifies the accuracy of the coversheet and of the endproduct The linguist may wish to display the source text translation and cover sheet on the screen at the same time U The checker may consult on-line working aids such as dictionaries and glossaries and P L 86-36 4th Issue 1987 o CRYPTOLOG o page 7 eONPIDBN'fIAL IIAf'fDLI l V1 A COMH 'f CUA HfflLS O LY DOCID 4010011 P L 86-36 EO 1 4 c CONFIBHN'fIAL should be able to look at a footnote as he is reading the footnoted text The checker may wish to jot down notes on the document in order to direct the linguist's attention to errors to be corrected or to train a junior linguist These notes should be clearly distinguishable from the text and easily deletable If the text is returned to the linguist for corrections it is then returned to the checker who will look at the notes and corrections to verify that the document is now correct When the document is approved the checker sends it along to another workstation for publication or a camera-ready final copy is printed Summary of Desirable Features of a Translator's Workstation Screen Size U The display for a translator's workstation should be large enough to allow a linguist to look at large portions of both the original text and the translation so that he can see enough information to understand the text There should also be room for small windows to display working aids such as dictionaries This requirement probably calls a 19-inch screen with bit-mapped graphics display The ll-inch or even larger raster displays are not large enough or flexible enough to offer a linguist an alternative to paper and pencil Fonts Word Processing U The foreign language displayed must be in a form i e chatacter strings rather than graphics or pictures of text that can be edited sorted indexed pal'$ed etc The user must be able to specify how the text is to be sorted or indexed including the direction forward or reverse nulls etc On-line Working Aids I FOUO Linguists require softcopy working aids such as dictionaries gazeteers ists etc which can be referenced from the workstation The dictionaries should include both personal dictionaries and master dictionaries available to all linguists New additions to the dictionary would be added to the personal dictionary and be stored in a file for review by a senior linguist before being added to the master dictionary The dictionaries must be capable of including grammatical information usage examples English translation and descriptive narrative in English the foreign language or both The grammatical information should be stored in concise linguistic format to save space but be expandable by a user interface so that the information could be used by language specialists who are not familiar'with the formal notation system Translator's Editing Features U In some cases for display of the text and in many cases for display of softcopy working aids linguists require foreign language fonts in approximately 30 to 40 foreign writing systems The foreign fonts must include punctuation and special symbols necessary to represent the foreign language accurately and the display screen must have sufficient resolution to represent the native scripts clearly and accurately The fonts required include difficult ones such as Japanese Chinese and Korean as well as the easier European languages A variety of font sizes and faces should be available to highlight portions of text U The checker requires capability to show on the screen the suggested changes to the product in order to assist in training junior linguists This would include such things as pointing out omissions in the text errors in the text suggesting deletions in the text indicating with highlighting pointers lines or color dependency relationships in the text which the translator has failed to understand 0 The author solicits comments from linguists and reporters who might be concerned with any aspect of the process described 4th Issue 1987 o CRYPTOLOG o page 8 60NFIBBN'fIAL lIAUBLEl ViA COMiU'f eUAIH EL8 OULY DOCID 4010011 The Effect of Detonations on Meteor Scatter o man made during artificial modification of the E layer or during high altitude nuclear detonations Meteors create discernible ionization trails at altitudes between 60 and 140 km above the earth On the average the trails last for seconds or less The electron densities in the trail usually range from 106 to 108 per cm3 These figures represent nominal means of distributions extremes are larger in all cases Effects of the latter are discussed below High Altitude Detonations Meteor scatter systems offer low data rate circuits at hiliCh reliability low power and some privacy I J L- If sufficient ionization arising from any other physical mechanism is produced in the altitude range near 100 km the meteor system operates continuously as a normal system This may occur under either of two conditions o natural when sporadic E is present or during periods of high sunspot activity N C Gerson W3 A high altitude detonation may be defined as one that occurs above 40 km where atmospheric number densities approach those in vacuum - or discharge - tubes The detonation releases huge amounts of energy both radiant to gamma rays and corpuscular ionized metallic and atmospheric species lumped together under the term debris The injection of this energy at this altitude profoundly alters the normal ionosphere It should be recalled that energy of this magnitude injected into the ionosphere produces complex hydrodynamical and magnetohydrodynamical disruptions and changes the ambient chemistry and ionization reactions After a single detonation the entire affected area remains violently disturbed for 30 to 60 minutes I EO 1 4 c P L 86-36 4th Issue 1987 o CRYPTOLOG o page 9 CONFIInlN IAL DOClD 4010011 I OONFIB8N'I'IAI a BULLETIN BOARD ANALYTIC SOFTWARE FOR PCs POUO A new work center in G33 is being created specifically to provide support on the pes for the exploitation of manual systems in G Group This center will offer personal help as well as generalized and customized programs All it takes is a phone call In addition it will maintain a library of cryptanalytic programs for the ASTW In the future it will call for programs written in G and elsewhere that could be used by a wider audience OUO The new center will be headed byD 19 3669 Other pc needs will continue to beprovjded by G33 wh chief is erplJ #en r is al 963- I I3669 OO'h P L 86-36 GERMAN TV HOUR If there is sufficient interest we will resume showing TV news and documentaries in German from the FRG Austria and the GDR In the past they were shown once a week at lunchtime on a walk-in basis Let us know how often you'd be likely to attend once a week once a month etc Please respond I DP16 HQS8A187 963-1l03 p L 86-36 U Finally it must be remembered that the ionosphere is self-healing After a period whose duration depends on the total energy and on the altitude of the detonation the ionosphere slowly returns to normal under the action of sunlight 0 MOVING Mail a change ofaddress to HQS Pl Editor CRYPTOLOG Include your Name Old and New Organization and Old and New Building EO 1 4 c P L 86-36 ----- - To the Editor I enjoyed Dave Chizum's article on Ethiopia CRYPTOLOG 1st ISSie 1981 lJO very much It's a thrilling and exciting story - and a firstrate piece of writing It's too bad it's clQS lified and can't be publi tled outside P L 4th Issue 1987 o CRYPTOLOG o page 10 C6NFIBHN't'IAL 86-36 I DOCID 4010011 ALAS WADE GILES I KNEW YOU WELL puy me a t'ik'et and I will t'ake the first poat leaving p'ort So he t'ook off for K'athay His letters t'o his mother reflek't the choy he felt in t'raveling from p'lace to p'lace He mate reference t'o the many intichenous t'ype nat'ives he had pump'ed int'o and the cheokraphik'al ottit'ies he had seen In any k'ase as may be kauched py it's witesp'read usache t'otay Wate invent'ed his syst'em and it is seen on map's and all k'inds of swell st'ull all over the p'lace On the pasis of the k'arefully kathered tat'a p'rovided apove one k'an easily tecite how t'o p'ronounce that p'art of a Ch'inese p'lace name that has an ap'ost'rophe in it and one which toesn't - also p'rop'er names poys or kirls and telek't'aple Ch'inese tishes such as K'ant'onese st'yle pean k'urd THE PINYIN SYSTEM Wonce upon a taim there were tu little Chinese boyis named Pin and Yin Won night Pin was studying Chinese karakters when he suddenly threw down his book and said to Yin Anyone hu studies this zhunk has to be a zherk What du you E Leigh Sawyer Ret say tu our killaborating on a niu systeIl We'll kall it the Pin-Yin' system Yin said Don't hand me that zhaive That would bi qiting on the In the December 1972 issue of the now defunct system It's ok to qit at ma zhoung or at kads but publication DRAGON SEEDS there appeared an not in skul Bisides if we went ahead with a niu article of mine on the Wade Giles system This system we should kall it Yin-Pin system for the romanizing of Chinese once used by NSA and throughout the intelligence community Well they qiued this over for a bit and finally was cast aside on 1 January 1979 in favor of the agreed on Pin-Yin The mein reason being Pin PINYIN system For historical purposes it might kud easily punch out Yin bikause he was a lot be well to reprint the Wade Giles item if only for bigger The top of Yin's head only kame to Pin's auld lang syne By the same token a PINYIN item qin would also seem appropriate so as to permit a Anyhow Pin said to Yin If we are tu get on better understanding of how the system works with this we'd be waise to du it on a full stomak THE WADE-GILES SYSTEM So after a big meal offish head and ternip chaoder and also raiding the kukie zha they set to work For the p'erson who has had little ex p'erience with the Ch'inese lankuache the p'ronouncing of Working dei and night they zhenerated the p'lace names p'eop'le's names art'ifak't's and system in short order When they were finished even the inkretinent's of Monkolian parpek'ue is they went tu the zhuish deli around the kourner for often k'onfusing An unterst'anting of at least beigles and a kup ofzho Ji whiz said Pin tu Yin Wate Chile's ap'ost'rophic usache aft'er cert'ain won't Wade Zhailes be mad when he learns of our k'onsonant's chust might enable one t'o atchust system He may try tu send us to zheil may be himself t'o this esot'eric linkuist'ic area A little even to Xing Xing pak'ground on Wate Chiles might pe in orter Well Pin and Yin did not go tu zheil The Wate Chiles was porn in Ch'ik'ako and lat'er moved to Cheorchia At that t'ime his mother moral of this story being there's more than wun atvised him You ought t'o invent something wei to skin a kat so qin up Wade Zhailes dai Why ton't you ko t'o Ch'ina Wate and invent the hards and bi of good qir 0 Wate Chiles syst'em He said Poy oh poy mom 4th Issue 1987 CRYPTOLOG page 11 FOR OFFICIAb f8K ONbJPY DOCID 4010011 HOPEFULLY FOR ALL OF WE _ _ _ _ _IDSC P L 86-36 In my article How To Write A Memo published in CRYPrOLOG 1st Issue 1987 I promised some pointers on grammar Here they are Grammar is such a problem these days that I have about given up on it and I debated with myself about tackling it in this article But I'll give it a try I must say something about a few errors at least the ones that bother me the most and interfere seriously with my ability to understand memos and almost everything else I read as well and ungrammatical Say most important since that is what you mean Apostrophes are everywhere these days and mostly where they shouldn't be Take its and it's for example It's means it is The apostrophe represents a contraction It's too bad you eon't come It is too bad you cannot come Its is a possessive adjective and never has an apostrophe One of my pet peeves is the misuse of hopefully If you want to make me cringe just use that word Until recently hopefully was just another common adverb meaning with hope He eyed the candy jar hopefully with hope He approached the task hopefully with hope But at present it's commonly misused One doesn't really mean this Hopefully with hope heU make the next touchdown Its place was already taken Other words of the same type are his her our your their and whose H -s was already taken Hers is a pronoun denoting relation or possession as are his ours yours theirs and whose In this case I can understand why it's confusing Ordinarily you make something possessive by adding an apostrophe and an s In other instances however you do indicate the possessive by adding an apostrophe and an s But this I hope he 71 make the next touchdown I urge you not to use hopefully at all if you can't use it correctly If you mean ''1 hope that then say it and do not misuse the adverb hopefully meaning with hope instead Don't be a lamb following all the others into this error And while I'm talking about ly words here is another most importantly That also sets my teeth on edge Only rarely is it the correct phrase Most of the time the ly is redundant Different books provide differing guidance on use of apostrophes for forming the possessive But the following simple rules are quite acceptable and will take care of most needs 1 If a word does not end in s add an apostrophe and an s girl's man's men's people's 2 If a singular word ends in s add an apostrophe and an s Lois's James's Perkins's 4th Issue 1987 o CRYPTOLOG o page 12 FeR eFFICIAL tJSE 6NLT I Dd'Crd -34010011 3 In plural words ending in s add an apostrophe only girls' boys' wives' Which brings me to two common misusages of apostrophes The first example naturally upsets me becauseit involves my own name which happens to end in an s I can't tell you how many times I have received memos with it written this way office Of course this changes my name to hiCh it is not A simple ''Mrs office will do nicely thank you Or else you can leave off the final s and write loffice instead I I The second example is even wilder It shows that the writer simply does not know the difference between singular and plural Some hilarious things can result of course The best one I've seen and I see it every night on the way home is a sign in front of a housing subdivision which reads Settler's Landing Every time I pass it I think about how lonely that single pilgrim must have been 11 Will we ever get affect and effect sorted out It's really easy but lately one so rarely sees or hears these words used correctly Affect means to influence It is a verb only Your answer affected influenced my thinking Effect means to bring about or to achieve as a verb This proposal could effect bring about some profound changes Effect means the result as a Q m The effecu results of all this are obvious 11 And what about lie and lay Lie means to recline it never has an object Lay means to put place or set down it always has an object when referring to present time A man can lie down to rest but he must lay down his head to do so When the action has already taken place note the past tense of lie whose principal parts are lie lay lain 11 And then there are quotation marksl Judging by the epidemic use of quotation marks these days a lot of people are saying a lot of quotable things But that's not the case of course What is happening is that quotation marks are misused for emphasis One of my favorites is a sign I saw recently which said in large letters NO TRESPASSING Now what original wit said that In general if you are not quoting someone directly or giving the title of something don't use quotation marks 11 To conclude I'll say something about the current misuse of I and me It used to be that only uneducated people made mistakes with these two words and also with shelher helhim we us they them Everyone knew about parts of speech and sentence structure and correctly used I she he we and they as subjects the ones taking the action and me her him us and them as objects the ones receiving the action Nowadays one's ears are assailed daily by misusages from all directions radio TV newspapers Just about everyone seems to have jumped aboard this ungrammatical bandwagon The problem is most acute when the objective case is called for as here I gave it to him He wrote it for my sister and me She told all of us about it The problem seems to be those little prepositions to for and of Whenever a preposition is involved people get tangled up and produce whoppers such as these It was designed for we mothers Jim said it to her and I I heard these examples on TV The second one is puzzling since it is inconsistent in mixing an objective her with a subjective I But it is also particularly popular Whenever there are two people mentioned the one farthest away from the preposition is usually in an incorrect form Is it any wonder that I have about given up on grammar 0 A man lay down to rest and laid down his head to do so 4th Issue 1987 o CRYPTOLOG o page 13 FOR OFFICIAL USB ONL t DOCID 4010011 THE MATTER OF STYLE P L 86-36 A CRYPrOLOG article on writing memos 1st Issue 1987 prompted me to ask which letters or memos that pass through the Executive Registry are truly memorable Not many and if they are it's usually for the wrong reason unnecessary parts This requires not that the writer make all his sentences short or that he avoid all detail and treat his subjects only in outline but that every word tell What an essay on the nature and beauty of brevity Most writers don't think about style It's hard to explain what style is much less how to achieve it What causes a combination of words to leap from the page to be recalled later Consider restating this sentence ''These are the times that try men's souls as Soulwise these are trying times Not quite the same Strunk also advises writers to revise and rewrite Few writers get it right the first time Have someone else check your w rk others can spot your errors much more quickly that you can To hold your reader's attention use definite specific and concrete language And be clear There is no substitute for clarity Ambiguity may amuse but humor is not the aim--clarity is D Before you can think of style you must know the principles of English composition Learn to recognize and use properly the eight parts of speech Diagram a few sentences for practice Learn the rules of punctuation improve your spelling by looking up doubtful words in a good dictionary and concentrate on the fundamentals of a plain English style Brevity will help most to develop or improve style OMIT NEEDLESS WORDS shouts Strunk He says Vigorous writing is concise A sentence should contain no unnecessary words a paragraph no unnecessary sentences for the same reason that a drawing should have no unnecessary lines and a machine no Solution to NSA-Crostic # 65 3rd Issue 1987 Ronald Reagan DEDICATION Plaque on Ops 28 You ofthe N ational S ecurity A gency are part of a proud tradition which has its beginnings with George Washington and the American Revolution And you make history quietly - silently - aware that your greatest victories ifever known at all will be divulged only many years from now 4th Issue 1987 CRYPTOLOG o page 14 FOR OFFIOIi L USB ONLY I DOCID 4010011 How to Measure Clarity in Writing whose third syllable- is ed or es as in created or trespasses 3 Add the results of steps 1 and 2 and multiply by 0 4 This will be the Fog Index It corresponds roughly with the number of years of schooling a person would require to read a passage with ease and understanding Fog Index 5 7-8 9-11 1R451 Inspiration for this article came fromlL-_ _ J nd her article How to Write a Memo CRYPrOLOG 1st Issue 1987 Guideline 4 in her article address ed how to write simply and clearly and it included some helpful rules But the article did not explain how to measure the clarity ofWriting I found a simple way to measure the readability of a piece of writing It's a guideline called the Gunning Fog Index first published by Robert Gunning in 1952 The guidleine does have two minor drawbacks First its simplicity results in a measure that is not exact And second the index says nothing about the writer's style However the value of the guideline's simplicity far outweighs the disadvantages for making it a useful writing aid Here are the steps for making the calculation 1 Using a writing sample that is at least 100 words long find the average number of words per sentence Divide the total number of words in the sample by the number of sentences This gives you the average sentence length 2 Count the number of words having three syllables or more in a 100 word passage Don't count a words that are capitalized b combinations of short easy words like bookkeeper or c verbs of three syllables 86-36 fairly easy standard somewhat difficult 12-15 difficult 17 very difficult Isn't it interesting that the standard level of writing corresponds to the comprehension level of an 8th grade education Sentence length and word complexity are the major factors that determine readability But don't go overboard while trying to achieve a lower Fog Index Writing that contains all short sentences with one syllable words would read like a first grade primer Again let me say that the result of the guideline is not precise But a sample having a Fog Index of 10 certainly will be clearer than one with an index of 15 Practice using the guideline on the first 8 sentences in this article You should get a Fog Index of about 8 6 1 Average sentence length is 13 5 2 Number of words equal to or greater than three syllables is 8 3 Fog Index is 13 5 8 x 0 4 8 6 Don't skip over this part go back and do it Now try the Gunning Fog Index on some of your own writing you may be surprised at the results Is your writing too complicated for others to understand Perhaps you can improve it by breaking those long sentences into two separate thoughts Substituting simple words for long ones should help too How does your writing measure up 0 4th Issue 1987 o CRYPTOLOG o page 15 P L Readability FOR OFFICIAL USH 6NLY DOCID 4010011 When the TCB exports an object to a mutilevel 110 device the sensitivity label associated with the object shall also be exported and shall reside on the same physical medium as the exported information and shall be in the same form i e machine-readable or human-readable form Machine-readable and humanreadable explain what is meant by the same form they are not examples of several kinds of same form Two Tips on Writing r f t 11 Two common abbreviations that are often misused are 'Le and e g They are especially nagging since you may not even find them in yoqr desktop dictionary Most people know that one of them i e as it turns out means tha t is and the other e g means for example o The problem is remembering which is which o The result of guessing wrong is that you may say that is and give an example or series of examples instead of an explanation or vice vel'$a H you don't mean for your list to be exhaustive be sure to say for example e g On the other hand if you mean to explain or amplify use i e How tx tell the difference The easy way is to remember the expansion of one of them i e stands for the Latin expression id est which means it is or that is Even if you didn't stud y Latin or a Romance language this one should be easy to remember Id and est look enQugh like it and is to give you a hint By the way Freud used the Latin id when he e up with his terminology Of course if you prefer you can remember the expansion of e g which is exemplia gratia Examples Note the form of the abbreviations a period follows each letter but there is no space between the first period and the second letter A comma typically follows the abbreviation 11 Does a sentence-ending period precede or follow the parenthesis or quotation mark This is the bete noire of many of us When a period is used next to a parenthesis the rule is completely logical if the parenthetical expression is PART of a sentence the period follows it H the parenthetical expression CONTAINS the sentence the period precedes the final parenthesis When a quotation occurs as the last element of a sentence the rule differs Even though the quotation is completely within the sentence the period precedes the final quotation mark The reason for this is that typographers feared that the little period would be lost if it fell completely outside the sentence For the same reason a comma is placed with the quotation marks An exception is made for technical text wherein a distinction must be made between period-quotation mark and quotation mark-period Examples The controls are discretionary in the sense that a subject with a certain access permission is capable of passing that permission perhaps indirectly on the any other subject unless restrained by mandatory access control The TCB shall define and control access between named users and named objects e g files and programs in the ADP system Files and programs are two examples among several possibilities for named objects Any discussion of computer security necessarily starts from a statement of requirements Le what it reall means to call a computer system secure Furthermore the TCB shall use a protected mechanism e g passwords to authenticate the user's identity Passwords are among several possible mechanisms The result of executing substr alph 3 5 is abc If the period preceded the hyphen the result would be interpreted as abc which is incorrect 0 4th Issue 1987 See the Convert Channel Guideline section CRYPTOLOG page 16 FOR OFFIEURIAL USB ONLY P L 86-36 DOCID 4010011 P L 86-36 '-- Pointers on Grammar r 12 P L 86-36 The article on mem osl mpelled One reason why many NSAers misuse partial me to pass on some pointers on grammar that quotes is that they subconsciously follow British usage which is to put quotes inside the period or readers might find helpful comma That's understandable What isn't About hyphenation understandable is the inconsistency many memos have it both ways -- sometimes in sometimes out My grammar teacher gave us a rule about hyphenation that works The reason isn't becctise o Hthe sentence makes sense using each of the modifiers separately don't hyphenate Example e have a large red barn It makes sense to say e have a red barn and it also makes sense to say e have a large barn ' So we don't hyphenate The reason is because is incorrect because the verb is requires an adjective or noun clause following The correct form is the reason is that The reason why is o Hthe sentence doesn't make sense using each ofthe modifiers separately hyphenate Example He had a broken-down car ' You might say He had a broken car but you wouldn't say He had a down car People say I like that kind of a meal when they should say I like that kind ofmeal Use of the indefinite article a limits the field to one thing In this case there is only one thing to choose from meal not a class of things It' not 1dDd of a Misuse of quotes It' not very unique My biggest pet peeve with regard to grammar is the misuse of quotes especially partial quotes The rule is almost always that commas and final periods go inside quotation marks Other punctuation marks should be placed inside the quotation marks only ifthey are a part of the matter quoted Here are some correct examples When people say I had a very unique experience they are probably confusing unique with unusual Unique means one of a kind It is an absolute that cannot be qualified Call it a ftgentlemen's agreement ' Why call it a ftgentlemen's agreement Change teat' to read tdog' 4th Issue 1987 Editor's Note Discussion on writing in these pages is now closed We refer reaaers to the many works on grammar usage and style that can be found in libraries and bookstores We will however consider articles on SIGINT terminology and usage CRYPTOLOG page 17 FQR QFFIGIAL US8 ONLY DOCID 4010011 THE HERO OF THE SENTENCE The Food and Drug Administration Extract from Getting Your Ideas Across Through Writing Training Manual No 7 1950 US Department of Health Education and Welfare enforces five statutes people administered by people yet often we write like this Employment in manufacturing recorded increases while there were declines in trade and domestic service The number of people working in manufacturing increased while the number in trade and domestic service decreased The hero of the sentence is the hero of the story that each sentence tells The subject is what the story's about For some strange reason we often pick the wrong hero as in this sentence When we do admit that people exist we often treat them as mere appendages to abstract The function of the Food and Drug ideas Sometimes we seem to go out of our way Administration is the enforcement of to keep from making them the subject of our statutes to insure the honesty and purity sentences like this of foods drugs devices and cosmetics This is the opening sentence of a The protection afforded industrial workers pamphlet is far from complete Many industrial workers are inadequately protected Who cares to read a story about a function The Food and Drug Administration is the hero General Assistance or relief accounted for of this story so let's make it the subject The nearly all the remaining recipients of Food and Drug Administration Let's make the public aid Nearly all the other people getting public aid were receiving general verb work too enforces five statutes assistance or relief Of course we cannot take time to fiddle that Here are some more sentences in which the way with each sentence Many of us need to real hero has been subordinated Just to change our habits however because we focus your attention on the advantages of habitually go out of our way to select the keeping people and relatively concrete things as wrong subject People like to read about people the subjects you may want to revise these and concrete things yet often we twist the sentences You can make the verbs work more sentence around so that the story appears to be effectively too about an abstraction Most of all people like to read about people Yet readers of Government writing must get the impression that no human beings exist anywhere Our sentences tell stories about functions policies grants conditions factors abstract ideas and it - especially it Everything but people Ours are programs for Responsibility for satisfactory working relationships within their organizational units rests with operating supervisors Refusal of employment of women workers is common on the part of employers There has been much opposition to the measure on the part of the educators 0 4th Issue 1987 o CRYPTOLOG o page 18 F9R 9FFIOIAL aSH ONLY I DOCID 4010011 Gimme Rewrite Va This is a note to prospective contributors and to What to do others who might have to write at NSA At a minimum have someone else do your copy We hope that the articles on good writing in this editing and your proofreading It is well known issue do not discourage you from contributing that you cannot do either for yourself because you because you're afraid of making one of the errors tend to see what you expect to see But it's still mentioned As writing errors go most of the ones better to hand over your bit of writing along with touched on are minor lapses which are taken care sharp blue pencils to at least one other person to edit for clarity And preferably that person ofby the copy editor should come from the population of the intended Self-respecting publishers of books magazines reader This is what experienced writers do and and newspapers all have copy editors as a matter of often they expect their text to undergo changes in course In many newspapers this tedious job is each of several passes by different'individuals assigned to the reporter most recently hired a The wider the distribution of the document the lesson in humility longer its life the greater the care that you must The usual sequence of preparation for books and give to preparation periodicals including newspapers is If you're doing any writing at all in this Agency text editing you should have at your desk some good references copy editing on grammar usage and editing Order a desk copy for yourself Browse in libraries and bookstores to proofreading determine which ones best suit your needs Don't Each should done by a different person In overlook handbooks for college students - some of CRYPTOLOG a low-budget rag the Editor merely them could have been written specifically for NSA changes hats Practice editing - a good way to begin Try your The point that the authors of the articles are hand at editing or rewriting an awkward passage making is that the finished product should not one that you had to reread to understand Practice contain such errors They detract from and tend to writing A good exercise in writing clearly is cast doubt on the substance Think of it as good writing up a suggestion grooming However elegant and stylish your clothes may be spots and tears are what people Hardest of all practice editing and rewriting your will notice not the couturier styling or bespoke own text Set it aside overnight or a week or a month then read it cold Now what do you see 0 tailoring 4th Issue 1987 o CRYPTOLOG o page 19 FOR OFFICIAL USE ONLY DOCID 4010011 Hutchins points out- that some prefer the term computer translation to machine translation or MT but declines to part with tradition He then surveys all or SO it seems the MT systems for which information is available His account of the Georgetown MT project was of particular interest to me since it covers ground that I became aware of while a student there His account is accurate and it contains many details about the various approaches tried and gives samples of the output produced Included also are wryly humorous anecdotes about the rivalries that developed His accounts of other MT approaches are similar Then came the infamous ALPAC Automatic Language Processing Advisory Committee report which all but ended funding of MT projects in this country Hutchins systematically demolishes the Machine Translation Past Present and Future premises and methodologies of this report by W J Hutchins University of East Anglia Ellis Horwood Limited Chichester 1986 He then turns to the modern or post-ALPAC era distributed by Halsted Press a division of John of machine translation describing a wide variety of projects and their applications in places such as Wiley Sons P 308 H85 Oak Ridge Many statements in the ALPAC report and elsewhere ring hollow in the face of successes Reviewed by 1 --1P12 for MT in actual use This book lives up to its comprehensive title The past begins even before the 1948 memorandum A chapter on Artificial Intelligence approaches to by Weaver from i which all machine translation MT concludes with the curious observation that its projects have sprung directly or indirectly with a feasibility in full-scale MT must however remain discussion of 17th-century numerical codes as doubtful eerily reminiscent of earlier dismissals universal languages and a 1933 Soviet patent of MT as a whole that the author himself has shown to be unfounded which went nowhere Weaver's memorandum justified the concept of machine translation in four ways one of which will be of special interest to readers of CRYPI'OLOG During the war a cryptanalyst had deciphered a Turkish message without knowing any Turkish Weaver verified that this incident had actually happened Hutchins explains that the relative frequencies of letters digraphs and so forth are sufficiently similar in English and Turkish to have enabled the cryptanalyst to succeed Weaver was mistaken but in no way does that vitiate the concept ofmachine translation Although there are many translation applications for which MT is inappropriate Hutchins's survey justifies his concluding remarks the future ofMT is secure it satisfies a genuine urgent need it is the subject of worldwide research and development and it is becoming a commercial product like other technical aids and office equipment the application of the computer to translation is a reality for many it is already as much a part oflife as the computer itself 0 3rd Issue 1987 o CRYPTOLOG o page 20 P L 86-36 FQK QFFICIAb USB ONbY DOCID 4010011 'FOP SEOREl'F UMBRA FROM THE PAST I 1 4 c L 86-36 u Provided by L 4th Issue 1987 o CRYPTOLOG o page 21 'FOP SElORE'F UMBRA P L 86-36 DOCID 4010011 009 006 0 a #00000 I 3 000 as 00 0 00000000 e22 ouunHOLIDAY A British Diplomatic Cipher of 1783 Recently a collector was going over some documents he had purchased some time before and discovered among them a 1783 cipher dispatch from Lord Manchester the British Ambassador to France The collector men- tioned It in passing to someone who by some strange coincidence had been at GCHQ in wartime who passed it to a current who sent it on to in the hope that one of our clever readers could break the system The prospects for success through analysis are not good There isn't much text We don't know what of those corres- pondents used during that period All that our NBA intermediary could dredge up was a cipher similar in appearance that Thomas Jefferson used to correspond with James Madison 0000 Q000 uh coo 009 Ill0000000000 0 0 a 000 0 000000 000 0 000030 0 00000300000000 a 4th Issue 1987 0 u 0 a DOCID 4010011 The Thomas Jefferson Cipher I 4th Issue 1987 o CRYPTOLOG page 23 FOR OFFICIAL USB ONLY I DOCID 4010011 The British Cipher of 1783 Jtr ' v tr- a-k 4 2- 7 GBP1 ' d t t4 n j la -oe - e 9 t d k f ' L- ' J rG et't yc--- e ' d p 7 ' ' c 1tr 'd rn-t r d- n 'L ha - Ja----n- y j Pb 77' 7GBP GBP - - r7- n u c'a -iJ 9 9 i' F 7 5 ' dA 2 3 176 tJ# '1 e - 3##t Z oF 'J 3 GBP ry 6'6 f GBP' 9- 2 41 J 8 97d 1'6 9 369 -9b 5696 4i' l F J8t4 I 3' GBP 4 f6 J' 6 7 r 21 b' jrzJ4 4 sy q 7 J5tIJ yj'S l' 53 7 1' 5 f d4 l'6 9 #' 44 2 6'94 3428 7f i'9 40 9 R119 1194 901 ' 8b' b f h 4 JJ J6tfj ffJ 1 4 47 3 4 tb94 21 1 ' 9 9 bi5 '2 7GBP bo 3 l B4 9 39 37- i' ft f'6 1 1 ' 9 61 4 1a TitJ I 2 4 J fJ J 3q J 97 4 f1 l'J i4 1 64 # 791 GBP 232311166 9-25 '1 5 '24 br z 9 t'1 444 21 9 4 74 J3tJ 3 ttrl g 2 to 1 2 t1dlJ 1 21 1J tJ 118 6 J 56 327f 61i Z5 4 j4 f3i 2 39 2424 J 5 3128 ff r SS64 4 J4 f ji afl6 'ItjCj 8- 24ftl B 06t1 t1S oS 4 tl 1 ' StlGGBP 1#'4 GBP42 f 4 '1 7 1'6 1 ' f3 ' 6 8 fl1 f4 7 4 4 of ' l' 6 5'8- 4th Issue 1987 CRYPTOLOG page 24 i'OIt O i' i'IeIAL tJSE ONL i DDCID 4010011 Mm MK w My 7 579 Mf g iff 2 52ng 79 445% 2mg 3435 55% 3 4 57 27er 2444 9392 545% mm way 24 5151 7752 xyxz 575 7516 02% f5 77 WM 55 fy 9am M Agar 44 M4 4 1 44 if L Mgr 6 2 J LX201 - iv 1 4th Issue 1987 page 25 DOCID 4010011 IIIS BOEURlfl lEN EURON'f2 INS COBEWORB rd2 'f RIAL -feP--SEEREl-NOT 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