OOqIO I 1 I 400 384t I II A27 _ _f - P L 86-36 GBPDUUlDGlGBPD11 $ l JrnUU GBPDl $Gl 'lD V l5lD 15 Ij WI5Ul 15 J' WUl l1aJGJ Sm urnl1rnrn February 1982 86-36 NATIVE SCRIPTING OF LANGUAGES U ooooooooooooo o ---l ooo oo oooooo ooo o o ooooooooooo o oooooo 41 A TIME FOR CHANGE U AAAS 1982 TWO REPORTS U ------SOFTWARE U ooooooooooooooooooooooooooooooo ooooo oooo ' 9 GAYLER U ooooooooooooooooooooooooooooooooo ' 11 NSA-CROSTIC NO 38 U oooooooooooooooooooooooooo David H Williams ooooooooooo 14 THE INTERNAL PERFORMANCE EVALUATION i o oo 16 FRIEND OR FOE U oooooooooooooooooooooooooo A WAIL A COMPLAINT AND A MELANGE U Samuel S Snyder et al ioooo 18 KRYPTOS A NEW SOCIETY U ooooooooooooooooooooo i o oooo 20 J Vooooooo o CORRESPONDENCE U oooooooooooooooooooooooooooooooooooooooooooooooooo I ooooo 23 SIMPLICITY IN COLOR U ooooooooooooooooooooooooo C Garofalo ooooooooo 'i' oooooo 24 SOME ADVICE FOR USERS OF UNFRIENDLY SYSTEMS U 26 ooo BUT WHAT DO I DO WITH MY PAPERS U 27 r i l 'fIllS BOeUMEN'f eON'fAINS eOBEWORB MA'fERIAt eLA 3SIFIEEl BY NSA eSSll 123 2 REVIEW aN 16 Feb 2612 Declassified and Approved for Release by NSA on '10-'1 2- 20'1 2 pursuant to E O '135 26 vl DR Case # 54778 DOCID 4009841 l 8R 8l E'IeIAb BE 81lbY Published by PI Techniques and Standards for the Personnel of Operations VOL IX No 2 FEBRUARY 1982 Editoria' PUBLISHER BOARD OF EDITORS Editor-in-Chief 1 1 7119 8322s '--- L production 1 1 3369s Collection I k8555s Cryptanalysis oooooooooooo 1 I1 5311s cryptolinguistics oooooo 1 5981s Information Science 1 3034s Language oooooooooooooooooooo Machine Support 1 8161s 1 A man named Jim Wilson from Moscow Idaho told a story the other day It seems that a certain community had a high cliff that was the cause of repeated accidents People kept falling off the cliff The town council decided to take action and passed a resolution to use whatever funds they had to buy an ambulance and keep it at the bottom of the cliff One member of the council who had been away on vacation returned and questioned the decision Why don't we build a fence at the top 1 5084s Mathematics - - I r 8 laf Puzzles oooooooooooooo David H Williams 11038 Special Research ooooooooo Vera R Filby But he was overruled We decided to do this thing right So we took a poll Everybody who had gone over the cliff said they didn't need any fence -- what they needed was an ambulance So that's what we bought 7 19 Traffic Analysis ooooooooooo Don Taurone 3573$ r---------- ------------ What are you working on -- a ambulance For individual or organizational subscriptions send name and organization to CRYPTOLOG PI or calli 1336915 P L 86-36 To submit articles or letters via PLATFORM mail send to cryptolg at bar1c05 note no '0' in 'log' l 8R 8FFrerAb HSE en Y fence or an DOCID 4009841 eetfFIBfJlf'fMIo Native Scripting of Languages U by 1L -- P L lp16 MT MAT Coordinator In Collaboration witb 1 -- O f B L 86-36 lpt6 Language Processing Coordinator I l hat is the value in practical terms of providing an NSA linguist with worksheets in native script This question has been asked and vari U ously answered for as long as there have been NSA linguists We may not be able to put the question to rest but we hope here to surface the factors which need to be weighed in order to make an informed decision --- U The limiting factors in producing material in native script form are technological and economic As technology has progressed the availability of the script has increased its quality has improved and the cost of producing it has decreased By contrast the linguistic and cryptanalytic reasons for or against having it have remained -and may be expected to remain reasonably constant U What are the linguistic reasons favoring use of native script They are so basic that no one would question them were it not for the difficulties and expense of effecting the result In the discussion which follows ideographic languages which pose the greatest challenge to technology will be addressed specifically However general statements are intended to hold as well for other languages whose writing system is not based on the Latin alphabet Feb 82 U A second linguistic reasOn supporting native script is that almost all academic training is based upon it PerSonnel arriving for duty are familiar with it and those returning for advanced training must know it Language acquisition requires extensive practice Once learned the skills are maintained through practice If after a thorough grounding a person fails to use the language or some aspect of it thst person can usually reacquire the skills relatively quickly through practice just as a person who once learns to ride a bicycle can after a lapse of years relea n the Skill relatively quickly CRYPTOLOG glolFIQIl n r Page 1 EO 1 4 c P L 86-36 Ih IBIoEj 'lIA Selllll'f 8UMUll l'M 81ft 1 86-36 DOCID 4009841 eOMPIIlir ItTIAL However the person who has not learned the skill sufficiently well in the first place whether it be riding a bicycle or reading a foreign language will not enjoy this advantage of quick recall of skills Providing native script routinely on materials processed daily gives the continuing practice needed to develop such a long-retained skill or to refresh a dormant one U The accompanying article was written in 1978 as part of the Language Processing Forum activities The Forum itself is now dormant but the remarks in the article seem to be current It is worth noting that NSA is now acquiring a hardware and software update of the IBM 3800 printing system Included in the software package will be a set of dot matrix codes for Chinese Forean and Japanese as well as an interactive cha - acter design program Expected delivery date is 1 June 1982 D Finally the operational need for a flexible work force demands that a qualified language analyst in one element be readily transferrable to another element As priorities shift or as attrition occurs within the work force it should be possible to rotate an analyst with the appropriate language skill into another element with a minimum of onthe-job training Exclusive use of coding Schemes by some elements and native script by o thers adversely affects this tr osferability Feb 82 CRYPTOLOG EURQ FI EUTlAL EO 1 4 c Page P L 86-36 U'Nhrb HI' EURQUINT SlMtUNEb6 8tibY DOCID 4009841 EO 1 4 c P L 86-36 e8UFIB 6N'flfrh EO 1 4 c P L 86-36 response while it could be quite tolerable for an operation being covered on one eighthour shift especially if the delay can be scheduled for a nonworking shift Overall time enhancement resulting in more timely product may be expected when the technology selected for scripting does not inherently introduce excessive delays during prime duty hours The technological reasons for or against native scripts as stated previously change with the technology There was a time when there were no automatic means for representing a large and complex character set Since then a wide array of offerings have been advanced including cumbersome mechanical impression printers driven by paper tape high-quality high-speed plotters and electronic devices such as full-graphics CRTs cathode ray tubes and electrostatic printers Each component considered must be carefully assessed for its impact upon the analytic effort The three major questions to be answered regarding adoption or rejection of a system are I 1 Are the characters of suitable quality for easy reading 2 Will the system's dollar sive and 3 What are the relative time costs cost be exces- These questions are subjective and should be made with as much valid information as possible In answering the question regarding cost in money the less tangible but very important linguistic impact as delineated above must not be discounted In other words the decision not to provide native script on the grounds of system costs must take into account the high cost of language training the potential resultant decrease in the quantity and quality of SIGINT product and the additional time if any it will take to train the language analyst in use of the alternative system In a period when the supply of personnel with language skills is diminishing collection technology is improving and the work load is expanding it is incumbent upon the manager to select methods which put those language skills to the most productive use In regard to time costs both the time delays imposed by the additional computer processing required to provide native script and the time enhancements accrued by increased productivity on the part of the linguist must be considered Computer processing delays must be measured in conjunction with the time sensitivity of the problem and a conscientious assessment of the delay For example a delay of two hours could be excessive for a problem requiring 24-hour coverage and immediate Feb 82 The following persons provided information which contributed to the report or concurred in the language of the final report organizational designators reflect those at the time of concurrence CRYPTOLOG L QtTFIB6tTTIAe Page 3 86-36 DOCID 'PElP 4009841 9Ee E'F tllIBR 'i A Time For Chan bylT53 W 'th apologies to Gertrude Stein a rose is not always a rose at least not in the Intelligence Community Over the years each agency in the U community and many subordinate organizations has developed and implemented its own unique formats for intelligence reports Now we find that something as basic as the serial number by which a report is known has no common form name or location NSA calls it the serial but puts it on a line with no key CIA calls i t the Report Number and keys it accordingly State Department calls it the ''Message Reference Number and puts it on the Classification line while DIA calls it the IR Number and puts it in two places the Subject line and the first line of Text U Even within the same agency some vital report elements may vary from one vehicle to another For example the standard CIA electrical report has the report number in one place while the reports released by the Domestic Collection Division have the report number in two places and often in modified forms Also the attention to format standards at the time of preparation leaves much to be desired in most agencies including NSA Key report elements are often misspelled misplaced or arbitrarily abbreviated U Such quirks may have been inconsequential when everything was being scanned and distributed by hand The human eye can easily search for and recognize whatever element of information is needed and will accomodate reasonable variations However now that each of the major agencies in the Community is building or planning to build major automatic distribution indexing and retrieval systems little differences add up to major design problems Feb 82 U The Intelligence Community is now at an important crossroad The automation projects we are going to have to work with for at least the next decade are still in the development stages If we are to resolve this format mess it must be now - before we are bound up by the implementation of systems designed for the individual formats If not each Agency will have to build complicated software to diagnose and index all of the report forms and the potential variations Unfortunately the increased complexity of such software will add operating overhead and require more human intervention in the form of editing U Any variation in the way similar elements of reports are presented adds significantly to the difficulty of retrieving information If an analyst does not know that the serial number of a DIA Information Report is found on the Subject line or that CIA reports contain a Subject not a Title he may well miss information vital to a project Finally if there is to be any significant sharing of bibliographic data bases throughout the Community it seems reasonable that we speak the same language U A study done by T5 of electrically received collateral intelligence reports revealed that while there are unique elements in the reports of each Agency there are also a significant number of common elements of information Therefore a proposal to standardize electrical reporting formats was recently submitted to the Information Handling Committee IHC of the Intelligence Community IC Staff The solution proposed is relatively simple all members of the Intelligence Community should agree to and comply with certain broad format conventions for electrical reporting This does not mean that all agencies will have to confine their CRYPTOLOG Page 4 'Pep 9BeltB'P tJIlllltA P L 86-36 - -- ' - _ - ___ I l - DOCID o _ 4009841 Feb 82 CRYPTOLOG e8UF 18EU'fVrI Page EO 1 4 c P L 86-36 _ DOCID 4009841 6Qt1FI881l'fIAh reporting to a single mold Where there are common elements in the various reports we should present them in a common way and where there are unique elements we should use a standard pattern PROPOSED MODEL FORMAT U U This proposal addresses only the information portions of an electrical intelligence report It does not attempt to cover communications format problems There are obviously however some areas that are of both information and communication concern Where such an overlap occurs an effort has been made to minimize the communications impact U In the proposed model format the information portion of a report is broken into three sections Address Contents and Management The Address section consists of information about when the report was sent who originated it and to whom it was addressed It includes the Precedence line along with the From To and Info lines The Contents Section contains the body of the report including the classification accounting information references serial number title and text among others The Management Section contains information about the acquisition preparation and disposition of the report This final section would include review and declassification instructions enclosure information project numbers etc U On the preceding page is an example of a DIA report in both the current and proposed formats On the following page is a proposed list of the various lines that could be included in a report No single report would ever contain all the lines listed as options but every report would contain all those identified as required lines Also included are examples of various other reports in the proposed format IMPACT ON NSA REPORTING U U The IHC has formed a working group to look into the problem and make recommendations on the proposal NSA CIA DIA State Department and each of the Services have been asked to participate T5 VI and Tl represent NSA on this working group U The proposed format is not one used by anyone Agency currently It would require Feb 82 changes by all members of the Community but represents a compromise employing facets from each of the current formats No element of information being reported by any Agency would be omitted of6 - If the proposal is adopted as submitted the expected impact on the current reporting format used by NSA should be minimal We would flag certain lines with keywords Serial Accounting etc and probably change the Title to Subject Other changes may be worked out by the working group as the effort progresses However since NSA has been reporting electrically for so long and using SOLIS for automatic indexing our format is already well adapted to the current environment and is fairly consistent SAMPLE NSA REPORT P 281906Z JUN 81 FM DIRNSA ITO NSA OSCAl LIMA NCR NMCC WHITE HOUSE BT EO 1 4 c P L 86-36 SEC RET CODEWORD SECTION 1 OF 2 ACODEX ENPOIF0081179 SERIAL XY 12345-81 SIGINT ALERT DUMMY TAGS ZDN ADEF AOPS LE SY SUBJ ZENDIAN AIR FORCE DEFENSIVE PATROL REQS TZ419 TEXT REVW 28 JUN 11 BT 113768 NNN FUTURE DIRECTION U U The proposal to the IHC also suggests that since format consistency and input quality are so important to the automatic indexing and retrieval systems future on-line report preparation systems should be designed to do automatic quality control and verification of critical elements In the case of NSA we could have any product report being released checked by an abbreviated version of the SOLIS system Thus instead of editors correcting product reports after they are released the reports would be returned on-line to the author prior to release with the errors marked Obviously such a system would have to have overrides for CRITICs or in the case of system failure CRYPTOLOG 88f1FIBEIl'fIA h Page 6 DOCID 4009841 eSIiFIBI K'fIAL Proposed Format ADDRESS SECTION precedence dtg FM TO INFO BT or ZEM CONTENTS SECTION classification caveats CITE ACODEX SERIAL PASS WARNING DIST EO 12065 CTRY TAGS SUBJ REF DOl ORIG REQS SOURCE SUMMARY COMMENTS TEXT Cite numbers SIGINT accounting information Report number Passing instructions Literal warning notices Distribution date State info on conformity with 12065 Country codes Tag codes Title References Date of the information Originator other than the FM address Requirement numbers Information concerning the source of the material Summary paragraph Preparers comments on the report The actual body of the report the Intelligence MANAGEMENT SECTION PROJ COLL INSTR PREP APPR EVAL ENCL ACQ DISSEM REVW DECLAS BT Project numbers Collection management codes Special instructions Prepared by Approved by Evaluation requirement Enclosures Report acquisition information Field dissemination Classification review date Declassification date denotes required fields Feb 82 CRYPTOLOG eSflFIBBIi'flAh Page 7 DOCID 4009841 Tap SEeRET HlIBM Feb 82 CRYPTOLOG Page 8 Tap SE8RET f R IBlb EO 1 4 c P L 86-36 DOCID UNCLASSIFIED 4009841 American Association for the Advancement of Science 1982 LOOKING AROUND wit IP13 OTwo Reports o o Software Gayler P L rogramming languages and operating systems were analyzed from a human factors viewpoint in the 1982 AAAS American Association for the Advancement of Science meeting in Washington by Dr Ben Shneiderman of the University of Maryland One of the surprising discoveries was that comments make a program harder to understand Indentation also impedes understanding which is opposite to popular belief Most operating system messages are hostile and accusing e g FATAL ERROR RUN ABORTED or ILLEGAL PASSWORD as if the user had committed a crime This apparently reflects an innate hostility toward other humans by systems designers P 86-36 aid comprehension Shneiderman reported that the Begin-End blocks of PASCAL were harder to use than the When-If of ADA In the design of terminal operating systems there are three categories of users to be considered viz novices knowledgable infrequent users and experts Each needs a different operating system The novices need the utmost in clarity and simplicity with a small number of meaningful commands tnowledgable infrequent users need simple commands which are easy to remember Psychological issues are An interesting technique was developed for controlled experimentation with programmers to test human factors Programmers at some skill level were given a program to study which used certain combinations of language comments indentation mnemonic names control structures etc Then they were asked to reconstruct the program from memory Independent variables were altered and statistical methods used to verify hypotheses o o o short term human memory load displays should be kept simple closure the desire for completeness anxiety called computer shock or network neurosis caused by the fear that the machine will dominate the user o The cardinal reason why comments and indentation interfere with comprehension is that they take space and cause the program code to be spread over more pages The page flipping impedes short term and long term memory and creates noise in the understanding process If the comments are printed to the right of the computer code the resulting program script is much more compact Indenting two spaces is better than four spaces It is important to avoid dangerous cluttering Higher level comments which describe the problem domain rather than the program domain Feb 82 the locus of charge control a desire to be in Response time is a complex matter A lag of more than 15 seconds may disrupt thinking Apparently it is beneficial to reduce the variation in response time within the system On line assistance may be more confusing than paper manuals The screen formats should be predictable so the users will remember where to find information Paper is a separate medium with wide bandWidth nonvolatile memory portability and no outages CRYPTOLOG UNCLASSIFIED Page 9 DOCID 4009841 UNCLASSIFIED Error messages are usually hostile and frequently uninformative e g guard mode error 2 System messages should be brief positive constructive specific comprehensible Usually the system messages are an afterthought assigned to the least experienced programmer The operating system should be designed around the system messages Currently no one keeps track of the errors that users make The mark of a good language or operating system is user satisfaction not designer delight o Shneiderman proposed that academic and industrial computer researchers should introduce controlled psychologically oriented experiments to develop research data on com puter human factors Commercial program and system designers should create standards and metrics for software quality develop guidelines for interactive system design and promote the use of pilot studies McKeeman from Wang Institute noted that software was much too cheap and was produced by untrained amateurs whose low quality code caused expensive problems for the users Programmers lack professionalism because they lack the power to say NO to a bad task or decision Programmers at TRW have about one tenth the training in their skill as GM auto workers Where software is used in public safety matters e g medical air traffic law enforcement matters it should be produced by licenced professionals who can refuse to produce or certify inferior or untested code Note the FAA is starting a $8 5 billion automated air traffic control system to replace most of the human controllers by computer software so the issue is not academic Stucki of Boeing observed that the direct cost of producing software may be much less than the indirect cost of the errors it makes He cited the famous failure of the APOLLO 8 moonlander software which failed from input overload 30 meters from the moon's surface The pilot then did a manual landing Shuttle I was postponed for a software reason The French have a system of 100 weather balloons half of which were lost when the command software issued a self destruct order Currently operating system documentation is word of mouth It is efficient to write the documentation before the coding A number of publications were cited viz Software Psychology a book by Shneiderman and IBM Systems Journal There was a six article series in ACM Computing Reviews March 1981 on Database Language Research At NBS in March 1982 there will be a meeting on Human Factors in Computer Society o I Irland of BTL said that 12 years ago the first SPC stored program control switch was designed for low down time but failed due to severe software bugs that require reinitialization about 30 times a day when a 40 year run between reloads was expected The effects of real traffic were unforeseen Now SPC switch software does not fail severely when it goes into service but it still takes hundreds of switch-years of operation to discover and remove residual bugs In 1950 BTL produced no software while now 50 percent of BTL staff work on software They now produce software at 100 times the rate of 15 years ago but the team experience is lower o There has been major growth in software effort and in real time applications and in life-critical applications space medicine air travel There has also been growth in the complexity of systems which network micros minis and macrocomputers - Maxims of software quality o Feb 82 CRYPTOLOG Quality must be designed in it cannot tested in UNCLASSIFIED Page 10 be DOCID 4009841 o Programs which are good are of high quality UNCLASSIFIED correct or o The earlier in the process the defects are eliminated the better o Bugs are never all eliminated o The most costly bugs are in ments o There is a paradox in trying to control software quality when it cannot be measured the require- A critical point is that all requirements must be testable Many metrics for software quality have been found wanting Some bugs are much more costly than others so that mere bug-removal counts are not a guarantee of reliability Initial bug density depends on requirements experience of design teams development environment and methodology GAYLER ON ARMS CONTROL U The elimination of manned aircraft tactical nuclear weapons chemical weapons and small surface warships were among the recommendations made by Admiral Noel Gayler USN Retd former DIRNSA at a AAAS meeting on Arms Control on 8 January 1982 The AAAS meeting was concerned with setting goals for defense R D to improve the security of the u S and other nations Admiral Gayler was joined on the panel by William Perry former DDR E and Richard Garwin of IBM a noted Arms Control expert There was a surprising amount of agreement among the panelists about the uselessness of many expensive weapons systems particularly the MX Since Soviet manpower costs are only one fourth of their military budget while these costs come to more than half of the U S defense budget the point was aptly made by Dr Perry that a numbers race against the USSR would be impractical because any expansion of numbers of weapons would increase the manpower demands and hence leave even less money for necessary hardware Under Lanchester's Law if the USSR has four times as many tanks U S tanks must be sixteen times more capable This also is hopeless The only answer to the Feb 82 Soviet threat according to Perry development of superior technology doesn't require manpower or money is the which Dr Garwin criticized the MX and argued that the central problem in setting R D goals was to search for better alternatives to unsuitable weapons He drew attention to the bureaucratic and political obstacles to considering alternatives He then described the SUM system in which a fleet of small conventional submarines operate within 600 miles of the U S coast and carry a few MX missiles in capsules The resulting secrecy and physical security would make a Soviet first strike impossible and would cost much less than the MX So far the Air Force appears uninterested in this alternative Garwin advised looking for things that didn't fit into a bureaucratic niche because new technology won't fit Admiral Gayler then developed his argument about new technology from first principles The USSR and Warsaw Pact he noted had all the resources and physical means to maintain their war capacity already within their geographical domain They had no global needs The West by contrast was a loose transoceanic alliance All the Western countries are in a deficit status in raw materials they must import China is a makeweight There is no China card and no alliance There are uncommitted countries many of which have a surplus of resources The national security missions in this context are complex since economic and political factors as well as military factors apply What do we want to do The primary aim according to Admiral Gayler was to avoid coercion by the USSR i e where the USSR could split the US from its Western allies or make the resources and raw materials inaccessible to the West CRYPTOLOG UNCLASSIFIED Page 11 UNCLASSIFIED DOCID 4009841 The West can succeed in preventing Soviet coercion by concentrating its defense efforts an reorganizing its defense to try leapfrogging the Soviet numerical buildup and geopolitical expansion the military tasks are Keep sea and air communications open This requires the capability to defeat air and sea forces and to control information Hold ground e g in Europe and Korea The West must be able to defeat armor and infantry and to control information Project power in remote areas world with or without bases of the To accomplish these goals requires weapons control Enemy weapons must be hit precisely not just bracketed by saturation fire The effectiveness of modern precison guided weapons is great and the total costs of an engagement will be reduced by accurate fire To illustrate the value of precision application of force Admiral Gayler contrasted the mining of Haiphong harbor which was accomplished without casualties and brought the North Vietnamese to the bargaining table to arrange a cease fire with the combination of saturation bombing and massive infantry op ra- tions in the jungle which were inconclusive and produced heavy casualties on both sides The new defense strategy would depend on modern target acquisition systems imaging and signal processing and detecting emanations The primary applications of the new technology would be for the missions of search surveillance and intelligence Admiral Gayler noted parenthetically that most useable military intelligence comes from the electromagnetic spectrum not from agents or informers Presumably this means SIGINT imagery and radar A central information spider with all data flowing into an overloaded central node for decision making would not be suitable Continuous control of information was vital Note Admiral Gayler did not define the term control of information but apparently meant a broad concept of acquiring and protecting all kinds of information Possibly he meant control of technological information as well as operational military information Feb 82 Space gives line of sight access for information and surveillance Space will be the dominant factor in naval war in ten years e g by the application of ocean surveillance systems and weapons guidance systems that will allow over the horizon targeting The evolution of space to a combat zone is inevitable Admiral Gayler gave the example of the early use of aircraft in World War I in which the French and German pilots on scouting missions waved to each other at first but soon turned to battles for air supremacy Space was a good place for a contraption war in which heavy battles between machines could be fought without producing human casualties or collateral damage To establish a U S military position in space the first step was to get a Space Commander similar to the land sea and air commanders now in control of those military missions Robotics will be an important factor in future defense systems Current examples of robots are satellites and nuclear power plants where complicated machines perform tasks in environments in which humans can hardly function Manned fighter and attack aircraft are about obsolete in the current environment of surface to air missiles Not only do the USSR and Warsaw Pact have a big SAM defense but even small countries have SAM weapons which are capable of shooting down U S and allied manned aircraft Admiral Gayler interjected a comment that although he was a former combat aviator he saw no point to training and executing manned air missions in which the pilots had no chance of survival or of accomplishing their missions The solution was to go to robot fighting vehicles on land and in the air The man would be in the combat loop but at a safe location on the ground instead of being spun around at 6G stresses while he was trying to hit a target Specialized small tanks which were expendable and automated could also do the ground combat function with the operator safely located The key element was a reliable data link and Admiral Gayler was sure that U S technology could provide that Comment if such remote fighting machines do come into service future battlefield communications will be even more voluminous and critical than ever before Sea battles in the future will depend on acoustic warfare At the present time the U S has developed its electromagnetic warfare and intelligence to a high degree A similar development in acoustic surveillance intelligence countermeasures and security is needed CRYPTOLOG Page 12 UNCLASSIFIED - - - - - - - - - - - - - - - - - - - - _------------_ o DOCID 4009841 UNCLASSIFIED People are a problem in the Armed Forces because there are too many of them not used efficiently Admiral Gayler said that based on his experience as a task force commander a modern large carrier could be operated with one half of the 2400 people by better use of technology He gave an illustration of a catapult operation in which four men are used in a human chain to decide when to fire the catapult which depends on the pitch and roll of the carrier where one man with a reliable command control link could do the job faster and better By reducing the size of the armed forces the military services can be more selective in who they hire and retain In training military personnel the u s should use technology like the arcade computer games rather than expensive and old fashioned training methods Maintenance should be based on no failure devices Earlier Dr Perry had proposed throwaway maintenance as currently practiced with handheld calculators Personnel policies in modern war should be based on motivation selection and allocation of tasks Admiral Gayler remarked that the main element in a sea battle was which side had the smartest Admiral Weapons of mass de truction have no military utility as theater or strategic weapons If tactical nuclear weapons are used the nuclear conflict will immediately escalate and the damage to allies will be extreme Therefore the only need is for central strategic nuclear forces to deter any use of nuclear weapons and these strategic systems should be survivable The primary point of vulnerability in a central war is not the nuclear forces but Washington itself The USSR will develop and maintain both fixed and mobile ICBM's Chemical and biological weapons have only a deterrent utility they are not effective combat weapons Unneeded programs such as the Bl and MX should be cut The Bl seemed to Admiral Gayler to have little use while he stated that the MX was destabilizing There is very little payoff in security from nuclear weapon development Most of the surface navy is not needed particularly destroyers The Sea Control functions of protecting shipping and clearing or blocking sea lanes can be done only by submarines supported by long range aircraft and surveillance systems Science should concentrate on payoff areas and should seek step function advances In response to questions Admiral Gayler noted that the proliferation of precision guided weapons PGW's to small countries had raised the level of military violence in a way that could threaten US forces His proposal for the elimination of all tactical nuclear weapons was predicated on reductions by both sides rather than unilateral action He agreed that the transfer of the Atomic Weapons program from Department of Energy to the Department of Commerce which had no security role or competence was not a sign of a well thought out reorganizaton When asked about the apparent indifference of the man in the street to the massive and terrifying buildup of long range nuclear weapons Admiral Gayler cited the psychiatric concept of denial in which a man told he has terminal cancer replies that he feel fine refusing to even consider the facts of his condition This in Gayler's view caused people to shut out the facts of nuclear annihilation which are too awful to face When asked specifically what kinds of wars he thought the U S should prepare to fight in the future Admiral Gayler said they should be non-atomic should not involve large infantry commitments and should be focussed only on specific military targets like the mining of Haiphong harbor which would be decisive rather than blindly destructive Money is important for defense but currently most is spent on people The answer is to reduce the number of people not only in the uniformed services but in the civilian bureaucracies which are associated with each Service and have much duplication of staff and function The elimination of the the three civilian Service Secretaries would in Admiral Gayler's view be useful Feb 82 CRYPTOLOG UNCLASSIFIED Page 13 DOCID 4009841 UNCLASSIFIED Ac c oJr cUng to a tong- tancU ng ltow NSA-Crostic NO J8 by dhw b ineM tJu1cLi t on a newe omeJt whOH name the Mme M thiLt 06 an e 6 tab-Uohed peJt60 1 meJt m t adop t a new name HaJVt y MOILgM and T un Co lWJa y Me two example 6 06 t t Jta cU t o n Me two WOItcU M and V oth Jayne Meadows' husband full name B au rhum C Asian capital 2 wds O See Word F 3 wds 256 I9T 232 134 i l l ill 229 TIT ill 176 m 149 E American singer and actor former Rhodes Scholar and instructor of English at West Point 2 wds F Play and film starring Word Q 3 wds foIl by Word OJ G Uawa iian word denoting H Lap 3 i1 non-Jlawaiia n person wds I Customary practice J Spur urge K Caribbean nation L Keen feeling for what is appropriate M British actor real name James Stewart 2 wds N Winner of 1981 Liherty Bowl 2 wds O Delicate exactness subtlety P Pierced with a sharp instrument Q Late erican actor former English instructor at Harvard Thornton Wilder was one of his students 264 i l l 253 made his acting debut in Word F full name 206 226 ill 99 R Percussion instrument Feb 82 CRYPTOLOG UNCLASSIFIED Page 14 193 2T ill 28 245 6 241 240 DOCID 4009841 UNCLASSIFIED s Tending to cause discontent or envy T - Sleeveless garment ness 1 11 Condition of being most satisfactory V British actress real name Jennifer Jones W Sharp repeated rapping or knocking comp X St Francis of -3- 249 mmm 126 Y American actor M B A once employed by the Budget Bureau of Connecticut 2 wds Z Made helpless ZI' Too wise you are too wise you be I see you are too wi se for me 2- R 3 -- X 4-rC -r-5 A 1 6 --- Q - 7-CE 18 --'Vm--- o'9 -nO II o 'Fl'TTITI-- E'II 12r- Qnt I i3 Z r I'TT14-iiT ' IIC S'-- K - E'17 r 1--rMn I '- I 19 o 35 M21 A 20 U 36 F 37 ZI S3 Y S4 ZI Q'22 i o o J 68 '-' - G I o I T03'X ia E l n A i ti 56 105 C Y 72 o i 41-- ti 42 z fs7' T - - B 26 21 j27 I' GI73 G 60 V 61 N 74 106 H 107 E o Q 75 A 76 M 77 J 78 I Y 79 123 0' 143 S 144 0 145 A 146 U o ' x'ls '-6'-K- -l-S7-C- - S-8-A-I-lS-9-N -- -1-1-60-iT6IA 162 0 1 o r 231 C 232 188 i -- -t-1 8 9 A- t 1 9 0'f i '1- 9'1'0 -1 921OM t---t'19 3 -nQt-l - 9 4 - yt-I 9 S 1l o 218 W 219 Q N A 234 N U 81 L 82 ZI 83 A 84 8 o o 114 I lIS L 116 S 117 Q 118 X 147 M 148 S 149 D ISO NISI D o I 180 A 181 D 182 P o 183 K 1l4 A '1 ' 9 - -b rt- 17 7'Ar -ln98 -I1 -- 19 9 - 11 12 '0 '0 A -l - --- -- --- -- -- -- --I----1I---I o o - __- - i 208 W 209 G 210 Z1211 P 212 Q 213 W 214 H 215 A o o o 220 A 221 W 222 E 223 Q oo n 233 o 224 IJ I 22S 2 1 o 26 Q 22' A 228 0 229 D 1 ' 230 P' 2 5 U '23-0 237 Z '23 8-5 239 E 240- Qr---'- '2 - 4 l'-Q -2 - 4- 2'Y -- 43 - P 24 4 'Cet -2d - 5 -' Q --- 2 - 4 6-D 2 '4' ' ' ' R ' o 248 S 249 X U 67 - - -tl - 6 - 3 y'1 1 6 4- H t-- --- '-17f 5 'F '-16 6 CttI6 7 - Eetll - 6 '8 A-i 187-V 202 A 203 Y 204 T 205 Z 206 Q 207 ' 216 E 217 H 66 o 40 A 141 Z 142 F o 1 A 124 F iis li-6-XI-1-2-7-N- -1-28-A-I-12-9-K --- -13-0-I 1-3-1-S- -1-32-U-I-13-3-At-1-3-4-D -I-3S-W i 1-S4 ' I-S-5 186 A o Y 80 IDS C 109 A 110 L III FIl12 Q 1130 o ---I--1f- - -4- -- -- - - o--- 2Ul Q 4 R 65 II o 121 W 122 N o o Z 'I-ilm --'A j-i ' 510 --' M h-5'I-- Hetl5 Z -7'ri Q 4 1 7 U h 4 '0' 169 J 18S II o o 5 M r- - - 4 6'--i' r I T 59 A 30 ''''W ''3'' 1-- A f- 3 '''2-- R ---t-'''33''''Z'-It-3 C4 -''''Et---i i Z 4 l Q '4'4' E '-4 58 P 28' R 18 o ifs-C h8 '9 -A-- -'t-- -If 9 0 y t 9 1-- M t 92 Ct 9 ' 3 -' yt 9 - 4 --- 111 9 S-A 9 76'Z -1 - --t c97 --- A 9 ' S - Z-1r 9 9 --Q I ' 0 0'H l- 071' ' Y7'I o I S2 R lS3 25 I 71 86 V87J 119 A 120 E E Z4 I -QfZ9--'E o I I o -l----J'--- - - _ __- - -- - - - - - - _ -- - --l---- -f- -- - - -- -- - --l 1 I o R 23 38 ZI 39 Ii 40 55 69 ZI 70 SSZI 1025 o I 'F 253 Q o Feb 82 CRYPTOLOG UNCLASSIFIED a 254 Z 2S5 SI256 D 257 C 258 lJ 2S9 V 260 P 261 Y 262 EI 263 21 4 Q 265 A Page 15 dh - DOCID een FIBlllfllAh 4009841 The Internal Performance Yuendot g 9 U Evaluation P L M at do the USSID System the Alternate Intercept Coverage Plan the Vital Records Program and Agency Staff Responsiveness have in common As you may have deduced from the title of this article each is the topic of a recent Internal Performance Evaluation -- IPE for short It is readily apparent that IPE's cut across a wide spectrum of Agency operations so it is probable that you will one day encounter the process In this article we hope to take the mystery out of the IPE by defining it by examining how it comes to be and by looking at the results that come from it In so doing we hope to convince you that the IPE is not a foe but a friend -- a friend that has the potential to improve your operation U To begin we should put the IPE into its proper perspective The IPE is one of several elements of the Performance Enhancement Review Program PERP Established by USSID 520 the PERP is a management system designed to provide DIRNSA CHCSS Key Component Chiefs and SCE Commanders with information needed to make decisions concerning the operational performance of the US SIGINT System The objective of the PERP -- and consequently the objective of an IPE is to improve the efficiency economy and effectiveness of the US SIGINT System U The Director of Performance Evaluation manages the PERP which includes not only the IPE's but also Field Assistance Visits Field Evaluation Visits and a PERP Panel Overall the PERP examines systems programs plans or activities that support the US SIGINT System Specifically the IPE evaluates systems programs or activities internal to NSA and under the direct operational control of DIRNSA CHCSS The PERP Panel an executive forun for coordinating and evaluating results of the PERP may be used to discuss potential IPE topics the status of an ongoing evaluation or the results of a completed one U How does an internal Agency program or activity become the subject of an Internal Feb 82 Performance Evaluation It begins as a proposal from Congress DIRNSA Key Component Chiefs SCE Commanders or individuals assigned to the Agency or to its field activities Many of the proposals however are generated within the Performance Evaluation Directorate itself Q2 U The proposal process begins by defining a potential internal evaluation topic and ends with a Management go no-go decision There are two major decision points along the way The first of these is the evaluability assessment A potential topic is assigned to a research Action Officer who first validates it through a cursory examination of the topic and the variables involved and then makes one of three recommendations o to continue the evaluation o to terminate it or o to put the proposal into a data bank of projects to be conducted at a later date The decision at this point is made by the DPE U If the decision is to continue the evaluation then the Action Officer will conduct detailed research meet with concerned activities or elements and conduct briefings or meetings as required While doing research in-depth the Action Officer also develops the evaluation methodology the road map that will be used to reach the most valid and reliable conclusions within available time and personnel constraints The methodology developed is a large part of the second major decision product the formal Evaluation Proposal U As you might expect the Evaluation Proposal really digs into the topic It looks not only at the methodology but also reviews the background of the subject and tries to assess what contribution can be expected when the evaluation is completed It defines the approach to the evaluation explains what data will be collected and how it will be analyzed and tries to estimate the cost of the CRYPTOLOG Page 16 8SNFIBf l'PI tb ------------------------ - 86-36 _ _ _ _ _ _ _ _ _ _ _ _ _L _ _ --l DOCID 1 4009841 88IfFIBBIf'FIz'tL evaluation The objective is to help management decide whether the proposed evaluation is both feasible and worthwhile U If you have the impression that the Evaluation Proposal process is time-consuming and methodical then you are correct but we keep it that way fqr good reason we don't want you to view the IPE negatively In fact we think the most important parts of the proposal process are the communicating and coordinating that go on At every step of the way the Action Officer and the elements involved work together They participate in meetings and briefings on the topic and they jointly develop alternatives options and recommendations All this is to ensure the integrity of the evaluation and that it works for you not against you U Here perhaps is the proper point to underscore our relationship with the Inspector General IG Simply put we are not an IG we do not function as one and we do not do their staff work We do not as a matter of course coordinate all our proposals through the IG If however our workup of a topic leads to an area where IG involvement is appropriate then we must coordinate our efforts with them For example fraud waste and abuse are currently high-interest areas in the Federal Government Should an example of fraud surface during a proposal process then we would be remiss if we did not act to bring it to the attention of the IG In fact this has never happened but if it did we would certainly make sure that the action was fully justified and that the organization being evaluated was fully aware of the situation U Once an Evaluation Proposal is approved and depending on the nature of the topic we may go all the way to the Director with a Decision Memo the implementation process begins First a Team Chief is assigned to the evaluation This may be the same person who developed the proposal but that is not always necessarily the case It is entirely possible for example that one Action Officer would do the proposal and another perform the evaluation itself The Team Chief together with the management and technical experts already identified begins the data collection reduction and analysis process Whatever format this takes questionaires statistics interviews etc -- the methodology is followed to end in a draft report The draft report is coordinated first with the team members then through Q2 and finally is distributed to appropriate elements Agencywide for review and comment D The IPE Team collates all external inputs and incorporates them into the final report It is important to note that the Feb 82 report will make every effort to present all points of view Of course there may be occasions when some opinions must be abridged or omitted from the body of the report In these cases they will be appended to the background material forwarded with the final report for management consideration The report usually does not stand alone rather a briefing accompanies it Yet even then the process is not complete Each report leads to conclusions and recommendations and the task falls to the Team Chief to monitor the implementation of the approved recommendations until each has been resolved In this way we try to ensure that an IPE is not a one-shot action but a continuing process that leads to a system improvement U Suppose that you would like a problem evaluated but you do not want a detailed formal evaluation Or perhaps you would like to conduct your own internal evaluation but do not know exactly how to start Q2 as the NSA CSS central authority for coordinating Agency performance evaluation activities and conducting performance evaluations will be happy to assist you on either a formal or informal basis We have a staff with a variety of backgrounds and experience in SIGINT and COMSEC operations as well as engineering and operations research skills We are prepared to offer assistance in any way that will improve efficiency economy and effectiveness U To recap IPE's are a management tool that can be applied to any aspect of internal Agency operations IPE topics can be suggested by anyone and they all follow a structured path through the evaluation and implementation processes Each step of the process is so designed to preserve the integrity of the evaluation and to make of it an exercise in participative management The principles of coordination and collaboration work to ensure that the Performance Evaluation Directorate is not perceived as an IG The entire process is usually lengthy and always methodical the better to guarantee that the results will contribute in a positive manner to the effectiveness economy and efficiency of the US SIGINT System So what do the Afghanistan Crisis the Russian Language Acquisition Program and NSA's Emergency Destruction Capability have in common Like the topics listed at the start of this article they have reaped the benefits of an objective evaluation process the IPE D If you have a problem that you think would benefit froQ a formal or informal evaluation contact Q2 We will be happy to assist you CRYPTOLOG 6811FIBEiIlTlz'tL Page 17 UNCLASSIFIED DOCID 4009841 '- -I 81 A W A Complaint and a Melange U by Samuel S Snyder T54 et al Editor CRYPTOLOG ARLINGTON MELANGE U by William M V Hoffman Herewith a couple of poetic contributions by former associate the late Captain William M V Hoffman Bill was assigned to my section B-II-b-4-a sometime in 1942 when we were first becoming successful in our attack on the Japanese Military Attache system Bill was a former Episcopal priest who was distantly related to Franklin D Roosevelt and liked to tell of having breakfast at the White House with FOR He had an uncanny knack with words but somehow had a terrible time making good guesses when we assigned him to overlap work His frustrations come through in his Wail of a Cryptanalyst We made good use of his talents later when we assigned him to write training materials for new people coming on the job In his Arlington Melange you might have fun trying to identify individuals he had in mind on each line Example the genius dashing about like a flea on a binge is Frank Lewis The other poem about computers by a computer was started by me on a plane late at night while flying home from a computer conference in Detroit in 1951 By the time we landed I had about five or six verses the rest were composed the next day in the office including contributions by Red Lathroum John Rixse and Dotty Blum Samuel S Snyder T542 Feb 82 There are M A 's from Harvard and Princeton Musicians and doctors of law There are all kinds of teachers And even some preachers and pale-faced Ph D 's by the score There are anthropological experts And Swedes who are silent and bland And one of our bosses Is a colonel of horses Who longs for an active command We have graduates straight from Smith College Who are brilliant and friendly and fair We have mathematicians And bright statisticians Who can calculate odds to a hair There's a plentiful sprinkling of screwballs And a few on the lunatic fringe While temperament flashes And genius dashes About like a flea on a binge We're a weird and outlandish collection We're a cockeyed and comical crew But come Nazis or Japs You'll find we're the chaps Who will see that the message gets through CRYPTOLOG UNCLASSIFIED Page 18 DOCTn L 040Q98421-- r UNCLASSIFIED THE WAIL OF A CRYPTANALYST U My rate's a megacycle My cost - a megabuck But if I make just one mistake My friend you're out of luck by W M V Hoffman There are moments when the world looks bright and rosy When the messages are tailing as they should When each fancy little letter Makes the columns come out better And the rows could act no better i f they would Then the heart of the cryptanalyst is merry And he chortles as he sets the traffic down And he thinks the enemy chaps Are such simple-minded saps And the man who makes their ciphers is a clown But alas there comes a day of horrid failure Our cryptanalyst is miserable and blue And the nasty little letters Thumb their noses at their betters And make gibberish no matter what you do You can stand them on their heads and read them backwards You can shift them left to right and to and fro You can calculate and mutter Till your brain becomes like butter But no matter what you try it doesn't go Then you curse the day when ciphers were invented And you kick yourself around for being dumb And the guy who once invented Such a code is a demented Low-down slinky lousy dirty rotten bum P h 86-36 My mem'ry's like an elephant's Each number sticks like glue I don't regret what I forget-I'm filled with something new When coders misinterpret My numbers new and old-It's all the same I get the blame For doing what I'm told The people all around me That come and go each day-Sometimes they fuss sometimes they cuss Sometimes they sit and pray The operator ponders The coder strokes his chin But Oh the loud annoying crowd They stand around and grin My ills are well attended My docs are maintenance men Oscilloscopes and lots of hopes Can fix me up again My access time can vary Whatever they decide But fast or slow they still say No They're never satisfied Sometimes Sometimes For I can A million I am disgusted-there is no joy see that I will be dollar toy Sometimes I get quite weary While pulsing through the night But if I err just hear them swear They never treat me right COMPUTER'S LAMENT U byl My names are always silly With Ack they often rhyme Be dignified I've got my pride Let's joke some other time land Friends They don't appreciate me In spite of all they say Why I do more than fifty score And brother that ain't hay March 1951 My business is computing With great efficiency Though what it's for is more and more A mystery to me My heart beats are electrons My mem'ry works quite well But where I go or what I know-That only man can tell So feed me lots of power Adjust my settings right And if we try then you and I Can help to win this fight Dorothy Blum Leo Lathroum John Rixse Feb 82 CRYPTOLOG UNCLASSIFIED Page 19 DOCID 4009841 SPORE 8BeRB'%' KRYPTOS A NeVi SociefYclI yl P L 86-36 I CA Intern KRYPTOS Society Historian he KRYPTOS Society chartered in October 1981 is a new organization whose purpose is to promote cryptanalytic excellence and to U provide a focal point for Agency cryptanalysts On 9 December 1981 at the Society's inaugural meeting Miss Ann Caracristi Deputy Director NSA spoke on 'Cryptanalysis the Kef Her address was followed by comments bYl Ion the steps taken by the Cryptana ysis Career Panel of which he is the chaiJJUan toward 'restoring the luster to cryp analysis' Attendance at the meeting was outstanding In the interest of accurat ely recording the comments of both speakers this summary has been prepared from J no tfo'1a f the time P L World War II providing tactical decrypts during the Korean Conflict 1950-52 and maintaining successful efforts against the Viet rOng 8 888 But with success Miss Caracristi pointed out problems arise Exploitation leads to production and we find ourselves making less of an investment in new problems The number of cry tanalysts including to-math has shruak Success also has impact on other fields It creates work the linguist the engineer and data syserson and the collector And as a L- 86-36 Miss Caracristi observed that without cryptanalysis the re would be no National Security Agency yet theJe existed no CA organization until 25 years after the founding of the Agency's first professional society the eMI She then addressed the problem of identifying Agency cryptanalysts Everyone it seems knows what a cryptanalyst does but no one knows just what one is The mathematicians in the back rooms of AS and G6 are they cryptanalysts What about the signals people in A5 and B3 And the linguists of G6 A2 andA3 They all perform cryptanalytic functions aren't they cryptanalysts Whatever your definition several names come to mind as eminent cryptanalysts William Friedman Frank Raven Peter Jenks Bill Lutwiniak I L Phil Dibben Jim Thompson j These and many others have through their successes saved countles s lives with the information they provided -- breaking ENIGMA and PURPLE in I P L 86-36 Feb 82 I iss Caracristi put it ringing Q The KRYPTOS Society can help the CA work force by o providing support to the CA Career Panel o identifying new talent force CRYPTOLOG Page 20 SEeRE'%' SPOKE - - - - - - - - - - --- for the CA work DOCID 4009841 8 ECREr P o providing mentors from its membership new cryptanalysts o keeping up with events in the outside world and in the academic community and o for advising the Director on the health of the profession P L 86-36 downgraded from ics nHial grade ofGG-lS or GG-16 Elements of 'M' were proposing that ecl ltlve S hOUld be a grade GG-12 Mr L- J was successful in getting them to upgrade thepQsition to a GG-15 once again a grade level more appropriate to the hroader duties of the office At this time c Jbecame the Executive eventually to be succeeded by I I then talked with office chiefs and obtained their endorsement for regular visits by the Executive to CA offices to see what was being done and what was not c J p He Miss Caracristi concluded her remarks with the following words of guidance We should look to the immediate future as an opportunity to recapture CA manpower through recruitment We must bring new cryptanalysts on board and develop their CA abilities teaching them to appreciate the excitement and the challenge of tackling hard problems to learn to enjoy small successes and to follow their intuitions when the trail gets hot By doing this we can build for NSA's continued success P L 86-36 U Whent I became chairman of the Cryptanalysis Career Panel two years ago his goal was to restore the luster to cryptanalysis to counter losses the field had suffered Toward this goal he set three objectives to provide a place for middle and senior cryptanalysts to seek advice and to learn of potential in other organizations an informal nonthreatening resource for post-professional career development to establish a professional cryptanalysts and to establish a technical excellence for the field of CA society U One deficiency noted by the CA Panel was the lack of a professional society for the central profession of NSA They weren't sure of the level of interest in such an organization but knew of the tremendous popularity of CA-30S an annual seminar series To determine the amount of support such a society might expect questionnaires were sent out Encouraged by a good response the Panel formed a committee to organize the society that eventually received the approval of the Director and of the Council of Learned Organizations U In establishing an award for technical excellence in the field of cryptanalysis the Panel wanted to recognize steps taken along the path to success Frequently those at the end of the road are recognized while early contributions go unrewarded Additionally recognition often depends on the value of the output The Panel wanted to recognize 'coups of the mind' without regard to whether or not vital decrypts were produced The award was envisioned as a pin with the possibility of a plaque for two-time winners Nominations would be accepted as events occurred The Panel would act semi-annually to select recipients presenting perhaps ten awards per year for award U The first objective was a service that could be provided by the Executive of the Cryptanalysis Career Panel since the Panel's responsibilities included much more than training interns Over a period of time the position of Executive of the Panel had been Feb 82 CRYPTOLOG Page 21 SECR E'f' IIVCeO DOCID 4009841 86d6 UNCLASSIFIED decld C panel hoWOUld recipients In closing l ptf red his favorite maxim for NSA cI 'YPtana LYl3ts D The award wasoriginaT1y conceived y I las an Agency award 'Hhenhe took the prorosalt ol I then DDM Mr I suggested that to establish it as an Agency award would take a long time and a great deal oof red tape Offering it as a professional society award however would be relatively easy With this recommendation the award was put on a back burner until a professional organization could be formed Eventually DDM ecame DDA and as replaced by When fuade with the a point of acquainting proposal he was advis ed that it shouldn't be a professional award but an Agenc3 8ward Having waited this 10ng 1 _resisted changing thE proposal again With the establishment of the society well on its way Mr c Jmade an appointment to discuss the new society the award and the objectives of the Panel with Director Inman and Deputy Director Caracristi Admiral Inman on hearing of the award proposal expressed a strong opinion that it should be an Agency award Strong opinions at that level are always considered When the M34 Awards Division was approached with the award proposal they recommended that it should be a professional society award not an Agency award It was then that did a little name-dropping apparently convincing them that an Agency award was entirely appropriate I U Establishment of the award was then hung up in the legal office for over three months The Panel meanwhile to decide on a name for the award went to Q23 with their original suggestion 'Excellence in Cryptanalysis' But Q23 didn't want any reference to CA in the title They wanted to avoid identifying our best cryptanalysts to the outside world and expressed concern that the design of the pin might be too revealing as well And in addition to the legal office and Q23 the offices of Protocol and Heraldry had become involved I IhadJreencorre c t in his appraisal of the situation P l ' do 1m_di e1y the mIraculous takes a little longer U The KRYPTOS Society welcomes new members For information please contact any of the following people President President-Elect Treasurer Secretary Members-at-Large I Program Committee Membership Committee CA Pand Ex o L - ----l I 86-36 I tasked by the Panel with choosing a design and name for the award suggested the 'Gold Bug Award' after Edgar Allen Poe's story involving the breaking of a cipher This suggestion solv d both the name and design problems -- lJ enarne did not refer to cryptanalysis lind suggested obvious design ideas ThE award proposal is now in the 'M' office that provides documentation on the administration of such awards I lexpects the award to make its appearance sometime in 1982 The Panel proposed that nominations go to the CA Career U Feb 82 CRYPTOLOG - I I UNCLASSIFIED Page 22 DOCID 4009841 FElR ElFFlelkh JSr Elllb' programmers themselves Corre8pODdeD e F01 fO You stated in your message I only ask that the material be sufficiently readable so that people who don't work in your special skill area can read it through and come away with some appreciation of how things are going in your territory I feel that everyone working on the Unix systems has some territory in which he has extended use where he might share his specialty with others who may not be as versed as he in that particular area FaBa Perhaps then we UNIX users should start talking to each other and start sharing our ideas Indeed Cryptolog itself might serve as a useful vehicle We have the resources available to us already so Why re-invent the wheel From tfh at CARONA To cryptolg at BARlCOS 1 ---' Tl512 Wayne P L U I received the message sent to Unix users back in October and I am taking this opportunity to express my feelings on the subject of communications of the users on these systems FEl JEl Having worked in the 1 IT88 area for several years I have had the pleasure of working with a number of people who use the Unix Systems There is truly a wide range of people using the systems today and an even wider range of tasks being performed No one has the time or the need to use everything on the system each individual specializes in some particular area which fits his or her own personal needs Each day someone discovers a new method of improving that area The users manuals help with the basics but it would be impossible to document all the routines and working aids that people have perfected and used on the systems PO O In the past I have attended several meetings of UNIX-BASED TERMINAL SYSTEMS USERS GROUP The purpose of these meetings is to give users information on developement of the Unix systems and to stimulate interest in areas which users are having problems The trouble with the meetings now I believe is that they are aimed at the Systems programmer level rather than at the user level A majority of people who attend these meetings are Systems people themselves or deal more with the programming applications of the system The material presented is on a level that is hard to comprehend by those who are not Feb 82 SOLUTION TO NSA-CROS IC I cRYPTOLOG Lan uag e i rftl1e No 37 U News 1-----------' September 1974 The Wisconsin Native American Languages Project is an undertaking funded by the Great Lakes Inter-Tribal Council to involve speakers of Wisconsin Indian languages Ojibwa Potawatomi Menomini and Odeida in the application of linguistics to the analysis study and teaching of these native languages ul RYE i5 5cheduled to be pha5ed out by mid 1983 Effort5 are under waY to et in If YOU are an touch with all U5er5 acti ve U5e r and have not been contacted by now pI ea5e advi5e T1533 exten5ion 40305 CRYPTOLOG Page 23 Faa aFFIeIAh JSB El lh'f 86-36 DOCID 4009841 UNCLASSIFIED from COMMAND October 1971 '% r- ' l i -'-il u _ he conduct of crypto TA studies 'involves the requirement to diff erentiate between quali ta tive lev -_ - - els - or categories and the use of a f'0 simple color scheme facilitates such differentiations Any color scheme to be used in an analytic effort of large volume must be blue purple and brown Ii- _ o simple in nature o skillfully chosen and o systematically applied Black - carbon number 2 pencil has been found to be the obvious choice for normal usage It is of a medium degree of hardness and density suitable for both erasure and longevity properties Green - is the weakest that is of lightest density All of the other colors under consideration superimpose on green quite readily An analysis of the various colors available is made to determine which are practical and most suitable for the task My experience has demonstrated that six different colors are most practical These are black green red For black erasures a medium hard rubber erasure is adequate For color erasures best results are obtained with a typewriter eraser Feb 82 CRYPTOLOG Page 24 UNCLASSIFIED D0 C - I D - --- 4Z-l0u o o 0 o - 9 B 41 UNCLASSIFIED Red and Blue are of equal boldness either can be easily superimposed on both black and green Purple - a bolder color which can also be produced by a superimposition of red on blue or blue on red Brown - is considered to be the boldest of all colors as well as the most exclusive and conclusive Colors manufactured by different commercial firms vary drastically in hardness density and coloration it is desirable therefore that having started with a particular brand to continue with that brand and not intermix brands to ensure that distinctiveness and clarity are maintained for brown where special conditions of clarity or distinction are vital to the problem Primarily useful as a final ordered and oriented intermediate enjoying the same general stature as brown o BLUE - A relative base value having a significant bearing on the state of recovery o RED - A base of lesser value or no relativeness completely arbitrary a first step necessary in any endeavor tt BLACK - To record or log information as it appears in its earliest or original form o GREEN - An envious color which is reserved completely for suspected garbles projected or expected but unobserved values not proven but highly suspected as being correct Used to alter a meaning or information without obliterating the original black form A color ladder may be displayed graphically as follows COLOR LEVEL Qualitative BROWN 5 PURPLE 4 BLUE 3 RED 2 BLACK 1 GREEN a Once a color scheme is established for a given problem maintenance of color discipline is mandatory in order to achieve uninterrupted and unambiguous continuity Discipline is also of great benefit to management in that it ensures that redistribution of analytic personnei can be effected with minimum disruption to the overall effort Postscript Colors Green Another way to express the levels color ladder is by these definitions o o of - symbolized hope to the ancient Egyptians the - in the Middle Ages was be good for the eyes supposed to - to the Mohammedan a sacred color BROWN - The ultimate in degree of trueness not to be questioned May also represent captured compromised information or its equivalent PURPLE - High in degree of reliability may be used as a substitute or companion Feb 82 - in modern times has come to envy signify Purple - to the Tyrians and Romans a purple robe or band of purple signified authority became a symbol of majesty to the Romans CRYPTOLOG UNCLASSIFIED Page 25 DOCID 4009841 UNCLASSIFIED Human Factors Corner ' P L by o Some Advice to Users of uNfriendly Systems U o But What Do I Do With My Papers U have received a number of responses to my article about Unfriendly Systems in the November 1981 issue of CRYPTOLOG Several people have called to express their strong agreement with some of the points in the article and to tell me Borne of their problems with the systems they are using My impression is that in some cases the problems these users are reporting are of a serious and long-standing nature It seems clear that for several large classes of NSA computer users in key applications areas problems with poorly-designed interactive software may be causing real and significant losses in personnel dollars and vital productivity I For example in one area which I won't name users were at one time instructed by their supervisors NOT to make use of a major set of information support files designed specifically for their work by an in-house computer support element in their own organization Why Because it was so difficult to get logged on to frame a request and to complete viewing responses without making an error A user error caused the system to be locked up preventing all users from getting at other things they needed until operations could be called to get the files closed and the user logged out The procedure for getting into this system is still so complicated and unfriendly and its response time is so slow once a user is into it that many people avoid using it on line and instead use hardcopy printouts You may be wondering what terrible programming language this system could be written in to cause it to be so inadequate The trouble lies not in the software system but in the design philosophy of those who developed and implemented the user interface The programming language is an excellent one and provides rich and extensive facilities for composing convenient Feb 82 screen No we can't gramming language it belongs on interface whose like this -II -- J responses If those guys can't use our system it's because they're too dumb There's nothing wrong with OUR system Anybody who comes around here complaining about OUR system is looking for a fight What can you do about a situation like this Those who have spoken to me about problems of this kind seem to have a defeated fatalistic attitude Some are quite bitter and angry They have tried to tell their management about the problem and the managers h ve tried to get something done about it apparently with little success I'd like to suggest an approach that may get you a little further Put a notebook or lined pad near each terminal Call it the trouble log gripe log or what have you Tell everyone who uses the terminals to note every case of a problem with the system that he or she experiences Note the name or initials of the user so that you can find out more about the problem if need be the date time in hours and minutes and as much detail as possible about the events that occurred Note what was typed in or entered what the response was how long it took for the system to respond and any other facts or events or times that show loss of user time due to deficiencies and unfriendliness in the system Don't forget to note cases where the keyboard layout or design of commands needlessly leads users to make costly errors for instance when a user is very likely to destroy hours of work because the ERASE key is right next to the ENTER key or a misleadingly-labeled command or field repeatedly leads users to make mistakes that CRYPTOLOG UNCLASSIFIED Page 26 86-36 DOCID 4009841 UNCLASSIFIED waste a lot of time over and over again day after day or an unhelpful response screen leaves users up a tree so they waste time trying to guess what went wrong After you have collected about a week to a month's worth of the trouble logs on all your terminals study them and make a summary of what you have found You can analyze the raw data in several useful ways For instance you will probably see certain recurrent problems that you suspected already and which together account for the major proportion of your personnel dollars wasted at the terminal by users trying to use the system You can write up a little paragraph on each describing the problems that cause the wasted time stating in man-hours and dollars man-hours times your element's average hourly wage the extent of the wasted resources and recommending a change to improve the situation If you can get your boss and better yet your bosses' boss to do the same thing and combine all the summaries into one memo to your machine support organization or whoever is likely to respond constructively your memo will have even more impact Keep up the trouble logs keep on making monthly reports and keep up the pressure This course of action if followed in a constructive way not just angry griping and backed up with good solid data on numbers of hours and dollars lost should certainly get you a lot further toward a solution of the problems It will also provide the machine-support management with a clearer idea of WHAT is wrong and some immediate ways in which they can improve the system ooo BUT WHAT DO I DO WITH MY PAPERS Reprinted from Human Factors Letter 4-80 published by the CISI Human Factors SIG Many of us who use CRT's may not be entirely happy with the response time or details of the user interface on the system Even for those of us who are satisfied with these aspects of the system we use however there remains a stubborn problem that needs to be solved somehow That is the matter of where to put papers books and worst of all machine listings to which we must refer while we key things into the terminal Various holders and racks are available to sit on the desk or table and hold pages of copy often with a horizontal bar or marker that can be moved or slid up and down to keep the place at a given line These are adapted from aids in use by typists before CRTs came along They might solve the problem at least for page copy if not for books or computer listings In fact I have rarely if ever seen anyone other than a typist or clerk using one of these devices in conjunction with CRT data entry though they look as if they might be helpful Elsewhere in this magazine I have included some drawings of these gadgets as advertised in a recent INMAC catalog of computer accessories l' You may well ask what about the all-toofrequent cases where potential users of a facility or subsystem have simply given up after many attempts and now no longer use it at all Or the case mentioned above where a system is so punitive and unfriendly that supervisors have positively forbidden users to employ it because it constitutes a menace to overall productivity I suggest that you ask some of your potential users of such systems to keep a record of all the instances when they WOULD HAVE LIKED to use it or to get at the information it contains Try to get an estimate of the time and personnel dollars you would save and the increase in quality of your product if this aid were available and user-friendly These are just a few suggestions of ways you can fight city hall and improve matters for all of us if you are a user of an unfriendly system Feb 82 CRYPTOLOG UNCLASSIFIED Page 27 DOClD - 4009841 0 _ UNCLASSIFIED II - Feb 82 CRYPTOLOG UNCLASSIFIED Page 28 DOCID 4009841 UNCLASSIFIED Users of CRTs are much more likely to prop folded bits of paper from which they must copy or extract information above the bank of keys on the keyboard or against the terminal beneath the screen Depending on the shape and size of the keyboard and the height of the screen above it there is frequently some space to hold a small or folded sheet of paper there Listings on size-l2 paper books and thick documents must be laid open on the desk or table to the right or left depending on which side you would prefer to have a crick in your neck that day You crane your head to one side or the other over your hands and arms which are busy typing at the keyboard and reach across awkwardly to turn pages If you are nearsighted your troubles are exacerbated especially with computer listings These vast cumbersome blocks of fanfold paper seem to extend for miles away from your eyes and the line you want to see is all too often way up there at the very top If you fold the listing over on itself it is very likely to tear or else to become dog-eared and disheveled so that it never folds quite right thereafter I have at times resorted to wrapping the listing around the edge of my table pinching it against the table with my tummy and squashing the top down so it sits in front of me just below the keyboard Whatever else happens sooner or later part or all of the fanfold falls onto the floor unravelling itself to create an exasperating mess Feb 82 For users who aren't fortunate enough to have a terminal stand or desk wide enough or uncluttered enough to accommodate listings or papers beside the CRT the lap or knees are the only solution And heaven help the user who needs to refer to several hardcopy documents at once while he is using the terminal I have often seen users who must consult printed working aids and dictionaries trying to cope with five- and six-deep piles of books and documents in layers on their laps the table etc while they paw back and forth from one to another trying to get the information they need to complete a transcript or report on the CRT Perhaps some day a truly useful comfortable and efficient file structure will be available on line for many of the most frequently-consulted documents There are now a few experimental systems that provide users with a spatial organization of different files like the physical arrangement on the top of a very large desk or table Such a system tries to help the user remember where he put away a given document when he discarded it temporarily to consult another which documents or files he is working with and where he is in each one A good example is the DATALAND system developed by Nicholas Negroponte and others at MIT Unfortunately these systems aren't available to most of us just yet In the meantime until the state of the art catches up with this problem we will have to live with it CRYPTOLOG Page UNCLASSIFIED 29 TOP-SEGRET- DOCID 4009841 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