DOCl GJ£iJUVlIlGJ£iJ b lJlBl lrnVU £iJUlIJGJlB L1U1Il1V Ul15U1Il1Ul15 Ul OO15 1 fi 15 f OO 1 Il1 l1lllmfi 4th Issue 1988 6w r n o s t _ 1 1 BULLETIN BOARD •• • • • • • 6 LOTtJS 1-2-3 A FANTAST rCREsEARCH'1'OOL 1 I 7 THE NEW CRAYONS AN A L L E G O R Y I I 12 RESUMES AND JOB APPLICATIONS r 14 THE QUESTION OF LEADERSHIP • • • • • • I 15 SCRIBE • • • • • • • • • • • • • • • I • • • 17 SOFTWARE REVIEW RIGHTWRITER • • • • •• • Leroy Ferguson •• 18 CONFERENCE REPORT AMS CENTENNIAL CELEBRATION David Harris • 20 QUOTE WITHOUT COMMENT • • • • • • • • • • • • 28 LETTERS • • • • • • • • • • • • • • • • • • 28 FROM THE PAST •• ' • • • • • • • • • • 34 BONUS PUZZLE MINICRYPTS • Bill Lutwiniak 35 NSA-CROSTIC plus No 67 1 I • 36 'fIllS tJO€tJ IEN'f €ON'fAINS €OtJEWORtJ r-IA'fERIAb ClASSIFI Q iY SA tCSSM 1 QIiCbASSIFY ON Origina'ting AgQn y'§ QQ'tQFR liRatieR ReEtuiFee NOT R l ASA8L TO CONTRACTORS Declassified and Approved for Release by NSA on 10-17-2012 pursuant to E O 13526 MDR Case # 54 net DOCID 4019712 ' P L 86-36 Published by P1 Techniques and Standards VOL XV No 4 4th Issue 1988 1'--- PUBLISHER '---- ' NSA'S BLACK HOLES BOARD OF EDITORS Editor 1 963·1103 L I Com puter Systems 1 963 1103l Cryptanalysis 963 5238 Cryptolinguistics 1 963 4740 Index 963-4814 Information Science 1 963 56 Information Security 972 2122 Language I 963-3051 1 963-5566 Mathematics Puzzles I 63 6430 Science and Technology 1 1 96H958 Special Research Vera R Filby 968-S041 Traffic Analysis Robert J Hanyok 96 351 Illustrators 1 1 1 963-4989 ' ' ' ' ' ' 1 To submit articles or letters by mail send to Editor CRYPTOlOG Pt HQ BAt87 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@bar1c05 bar-one-c-zero-five note no '0' Always include your full name organization and secure phone also building and room numbers A recent theory about the universe holds that there are little black holes like worm holes In our universe leading to other universes It must be right We have evidence right here at NSA There's one worm hole between Fort Meade and FANX If we ever get into that universe we'll find reams of memos that never got to their destinations There's another worm hole just outside of SAB 2 that captures bundles of CRYPTOLOGs on their way to the Intended recipients There's even one just a pin prick to be sure between HQS building and Ops 1 that attracts proof sheets of CRYPTOLOG articles Strangest of all NSA's black holes don't go In for junk mail You know the kind when every employee gets a memo after the fact stating that 25 December is a legal holiday Curious isn't It Of course there are those who attribute the disappearance of mail to a virus the same kind that affects the US Postal System We don't know which of the two is the right explanation for the missing mail Or if there's still another cause We hope that somebody can find out soon and do something about it For Change of Address mail name and old and new organizations to Editor CRYPTOlOG Pt HQS BAt87 Please do not phone Contents of CRYPTOLOG may not be reproduced or disseminated outside the National Security Agency without the permission of the Publisher Inquiries 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 FOg OFFICIAb l'8K OP'lb¥ DOCID 4019712 I G9NFIB8N IAL PROJECT RELOAD L 86-36 86-36 P04B 'f9U9 Project RELOAD was launched in March of 1985 to improve the U S SIGINT System USSS support to national decisionmakers and military commanders This was to be done by upgrading the SIGINT production capabilities and missions of field sites in particular those operated by the services The initial objectives were to restore technical health in the field and to enhance analysis and reporting in the field thus ensuring the wartime survivability of the USSS U In the earliest stages the project concerned developing short-term solutions to site-specific endemic problems identified during field surveys made in late 1985 of 55 collection sites These surveys gave mission management at the sites an opportunity to document their perceptions problems and needs The findings formed the basis for and provided the spirit upon which RELOAD strategies were subsequently built Problems unique to each site were addressed by appropriate Agency elements which provided quick fixes to pressing problems In fact some 325 corrective actions were implemented by late 1986 resulting in some improvement in front-end production 0000 However by the time the final RELOAD assessment was approved by the Director in April 1986 it became evident that certain problems in the USSS were systemic deficiencies requiring the development of longterm institutionalized solutions U About the time the basic RELOAD Findings and Recommendations report was released the results of three other major studies became known and the recommendations cited therein significantly influenced the course of follow-on RELOAD actions These studies are 'fOUO A Study by the Director's Blue Ribbon Panel consisting of high-level NSA executive-consultants It reviewed the USSS in its entirety to determine the state of its technical health and its ability to accomplish its assigned mission now and in the future It concluded that technical health in the field had to be strengthened the practice of giving the military the SeEs virtually robotic tasks had to cease and that NSA had to share the high tech duties with the field -- not just to enhance the field's capabilities but to allow the time and talent of the NSA work force to be directed toward keeping up with the vast expansion taking place in all aspects of SIGINT Systemic deficiencies were evident within NSA as well as in the field for different reasons G 00 The Future SIGINT Capabilities Study FSCS which examined the capability of the USSS to handle the rapidly increasing volume of signals in the environment as well as its ability to cope with our targets' advances in signals technology It pointed out that the seriousness of the rapidly advancing technologies of our targets poses serious problems requiring strategic action if the USSS is to maintain superiority in intelligence 4th Issue 1988' CRYPTOLOG ' page 1 CJ9NFIB8NTlAb P L 86-36 II A MJJL i JAl A Q TT II UTMi bi Q T bY DOCID 4019712 L 86-36 llatherinll 1 OONF'IBElNCfIAL U Through all of this there has been a very vivid perception on the part of NSA the field and the SCEs that RELOAD has had and continues to enjoy the strong backing of DDO and the Director U The Director of NSA charged the key components and the SCE commanders with identifying specific people to work with the RELOAD Project Management Office to implement the findings and recommendations of the various studies A tiger team approach was selected as the effort is and will continue to be - - - - - - - - - - - - - - - - - - a corporate one The players involved are DDR DDT DDA ADT DDO A B G P and W CMC ESC INSCOM and NSG fflT The working groups chosen to meet and implement near-term less than a year midterm two to five years and long range beyond five years solutions to the issues have contributed immeasurably to the accomplishments of RELOAD as have the innovative suggestions of field sites The SCE HQ are represented on all working groups with the Cryptologic Manning Collection and Processing Working Groups actually being chaired by INSCOM ESC and NSG representatives respectively RELOAD is broader than just the working groups and the initiatives do not come from the working groups alone Nor do initiatives come just from the SCE Headquarters they can come from any element in the process and ofu n spring from the field sites themselves RELOAD has brought about a better bonding between the - - - - - - - - - - - - - - - - - - front end and the back end of the system with FOLLOW-ON ACTION the common goal of doing it more efficiently and effectively U Because three of the four studies focused on operational matters under the Operations U RELOAD continues to use quick-fixes to Directorate it seemed reasonable that follow-on problems wherever practical but is following up with long-term corporate NSA-SCE action should be undertaken by that Key Component Since Project RELOAD had done institutionalized programs serving as the so much on the study initially and was most means by which organizations are brought familiar with the problems it was given the together to fix their own problems It provides corporate go-ahead to attack the systemic the forum and the opportunity through the problems plaguing the system The nature and working groups to actively pursue solutions to breadth of the RELOAD program changed problems with the RELOAD banner serving as drastically as it shifted from problem study to the driving force A corporate entity has been problem solving The reconfiguration and created in which all of the parts recognize modernization of the entire USSS seemed to be essentially one goal to equip and to educate necessary Consequently the scope of Project the work force to acquire the skills required to RELOAD has expanded as has the original list cope with the challenges of tomorrow of systemic issues By the end of 1987 RELOAD had blossomed into a widespread brainstorming P L 86-36 EO 1 4 c 4th Issue 1988 CRYPTOLOG page 2 60NFIBJj NTIAb lIANBhi l ViA eOMHff eIIA1H ElLS ONLY I DOCID 4019712 EO 1 4 c P L 86-36 C6NJi'IBEN't'IAL effort with fresh ideas emanating from many NSA and SCE sources The effort has already yielded some degree of improved field technical health has upgraded target assessment and reporting and has made genuine progress toward expanded front-end crisis response and self-sufficiency capabilities The emphasis continues to focus on improving technical health through enhancing the skills of the work force As a consequence certain RELOAD initiatives traverse the entire USSS and impact on both the military and civilian work forces The achievements recorded in 1987 are many and attributable to the hard work vision and cooperative efforts of numerous personnel at various field and headquarters elements 0 000 During the past year the technical health concerns and issues documented in the RELOAD findings report of April 1986 drew the corporate working groups that were examining the topics into a scope beyond the narrow analysis and reportin structure earlier associated with RELOAD analytic proficiency is on the rise A working group TA Focus Plan calling for a broad range of reforms to strengthen career field discipline mechanisms is in final coordination and appears to have overall management command backing 0 000 Cryptanalysis - A host of SCE field stations have now reacquired basic crvntanalvtic canabilities and missions I teach cryptanalytic I Two courses were conducted to techniques and acquaint --- analysts with target cryptanalytic histories and profiles Stations have been given appropriate software and data base access where possible to aid local exploitation efforts Early results are most promising Productivity at some sites has markedly increased processes producing monthly keys for some targets have been accelerated and most important local target identification and warning methods have been upgraded FOYO ADPlData Bases - One achievement in this area has been the development of a means for all sites to request access to data bases by approaching one office T52 with their justification Group A has developed and fielded procedures to provide access to information requested which also satisfy needto-know controls This system will now be applied to other data bases and made applicable to more field sites Access to data bases from field sites will remain a function of the capacity for communications paths to ---1 support it Other aspects being worked are a directory of information managed by T52 which U The present scope of RELOAD actions becomes evident from the following summary of can deliver a listing of data bases provided by highlights target offices designated for access to field sites and an up-to-date listing of software that FOITO Traffic Analysis - The focus has been has been developed either in NSA or in the on increased training responsibilities and field and is available for acquisition documentation Site technical exploitation initiatives are flourishing Corporate skill standards have been introduced to assist the training and assignment processes and major revisions to existing career pathing and development schemes are under consideration The traffic analysis and SIGINT reporting skill standards were subsequently combined into a SIGINT analyst skill standard Certain field sites have assumed increased target responsibilities and new technical studies e g ITNs are resulting Mission managers are refocusing duties to capitalize on available EO 1 4 Cfperience and skills and as a consequence P L 86-36 P L 86-36 I CONIi'IBEN'fIAL U Manning - With the aim of getting better skills-to-job matches the SCE HQs are now geared to push their current manpower syStems to the maximum One universal method employed was the drafting of better job skill requirements statements and the rollover of these expressions into TDs Also to be employed in this process will be the RELOADdrafted and corporately approved three-tier skill standards Attention is being devoted to tracking and assigning skilled personnel including the addition of trailers to occupational codes to identify specially trained and uniquely capable personnel Most important SCll managers are striving to implement improved career pathing schemes to enhance the cryptologic development of personnel N'SG's A R plan is a trend setting example U Language· This newly established corporate working group has established as its goal a commitment by managers to fix longstanding problems in proficiency maintenance retention assignments and training • It held its first meeting in December 1987 and will interact with SCE and NSA management to gather essential information to forma baseline of existing linguistic proficiencies from which to build realistic requirements that will lead to improved language competencies in the field and at NSA 0 OCO Processing - The main thrust in this area was to make collectors cognizant of why certain signals were important and what the plans and strate es were for future field ex loitation EO 1 4 c P L 86-36 U The challenges of the combined studies have driven what has been done u der RELOAD and continue to direct what is yet to be accomplished The Field Stations and SCE HQ are most enthusiastic and have been the key players since the results of the initial site surveys developed into positive actions The producer community is now convinced that RELOAD is not just a fad that Will go away The front-loading effected by RELOAD will continue it will continue because there is now an understanding of the scope of the problems and an understanding of what must be accomplished under the umbrella of RELOAD The concerted effort includes reaching outside the USSS to obtain reviews and reconsiderations of procedures affecting the very heart of SCE skill acquisition and maintenance to wit training and assignment 0 4th Issue 1988 CRYPTOLOG page 4 eOIU'ItlEY 'fIAL IIA tDU3 V'fA COr HU'f CIIArnn JLS moiL EO 1 4 c P L 86-36 I DOCID 4019712 HOW TO MAKE A MORSE OP P 1 4 c 86-3 P L S OOO NSA was asked to re are SIGINT trainin several years ago The OPI solicited input from several sources for material and training including E23 now E54 for help with training new Morse operators The following is based on a memo prepared by one of my NCS colleagues at the time and to which I contributed We never learned whether our recommendations were adopted or used But we think it worthwhile to publish our findings on what it takes to make a Morse operator The statistics and conclusions are based on our experience at NCS in training morse operators THE RECOMMENDED PLAN U It usually takes a Morse operator trainee approximately twenty weeks to achieve a code copying speed of twenty groups per minute This is not full time -- not all of the 800 hours are spent copying code To achieve 13 GPM usually takes about 165 hours U The usual progression for a Morse trainee is that after about 110 hours the average student can copy eight groups per minute From this level the student will progress rapidly usually attaining 13 to 15 GPM after an additional sixty hours of work At this point a plateau is reached after which it will take another 150 hours of copy to progress to 18 GPM By the term plateau we mean the level at which the student can copy at his highest speed accurately and with relatively little trouble The next higher speed on the other hand seems to present the student with an inordinate amount of trouble That is at the plateau the error rate will be at or under 2% while the error rate at the next higher speed usually starts at about 15% and gradually diminishes The plateau speed 86-36 varies from student to student but as we said above it usually ranges from 13 to 15 GPM U This plateau is apparently the transition point at which Morse code copying becomes a psycho-motor response rather than a cognitive translation of the sounds to symbols After the transition is made code speed acquisition again rises steeply with a gradual flattening of the curve as the natural limits of the students' abilities are approached It will take approximately 400 to 500 hours of copying Morse for an average student to attain a code speed of 20 GPM 6 In designing a course for an intercept operator material other than morse copying of course is taught In designing a course combining Morse training witl the academic material required we recommend that the subject matter is scheduled so that the first two weeks are devoted to learning the Morse alphabet After this most of the time should be spent on the academic material gradually increasing the amount of time spent on Morse until the final weeks when copying Morse is full time The conditions under which the students work particularly during the last weeks should be as realistic as possible This could be accomplished by introducing recordings of actual target transmissions or perhaps by assigning the student an actual mission which is duplicated on cover U Morse learning is a very individual thing with students progressing at different rates reaching their plateaus at different code speeds spending different amounts of time on their plateaus and so forth A student who appears to exhibit good potential then should not be dropped for failure to achieve average progress 4th Issue 1988 CRYPTOLOG page 5 SElOREl IIANBLEl VIA OOMIN GWANNELS DOCID 4019712 BEeRf 3 U One feature of Morse is that accurate copy is much easier if and when the student learns to stay behind the character being sent Three to five characters behind is what the average high speed operator manages but accuracy is nearly impossible if the operator is working right on top of the character being sent This should be stressed by the instructors from the time the students have mastered the alphabet all the way through the course effort on operational material and the rest either speed work if still needed or introduction to other alphabets U From a training perspective it will take twenty weeks or more to make a Morse operator Most people will tell you that at least a year actually doing the work is necessary to produce a competent operator but that's another story that can be told some other time 0 SUGGESTED TIMETABLE U The end of week nine by which time the students would have had 159 hours of Morse training most of the students should be working on 15 GPM or thereabouts Whatever speed they are on it is likely that most if not all will be at their plateau level Here is the time to start to bear down and increase the time spent in the training day on Morse The end of week fourteen after approximately 300 hours of Morse training striving to achieve higher code speeds will be getting old A break here will in some cases provide better results in the long run and in most other cases make little or no difference That is there is a good chance or benefit and little chance of loss by taking a break from the Morse at this point The next two weeks we recommend should be devoted to classroom training in intercept procedures When Morse training is resumed after that all appropriate logging and copying formats should be incorporated U By the end of week 18 some 330 to 350 hours of Morse training will have been given 8-666 Starting with week 19 or thereabouts code speed practice should be interspersed with recording or tar et activit or if resources are available Additi ona1'my - O sr u' ' 'r-e n T ' w o ll - O a' 'l'v o 'e 'a c le 'v e e required code speed and are apparently gifted operators can be introduced to barred letters and special characters be ond what the will have learned as routine BULLETIN BOARD UNIX BROWN BAG SEMINARS FQUQ At the February session of the brown bag seminars on UNIX sponsored by CRYSCOM CRI l'335 will speak on analo es and CA applications In March L -J T335 will give details on data configuration using SCCSIMAKE especially in regard to ther UNIXIUNICOS Software Exchange Proposal Date time and place for the February and March meetings will be announced later Fpr further information please call or write the CRYSCOM Exec 1 lops 1 P13 963-3405 CRYSCO '89 P L 86-36 FQB'O Planning for CRYSgO'89 is underway The sessionswilI tab plate the week of 19 Jun jnHQS 9A135 the Old Director's gonference Room Suggestions for topic t are solicited Contactone of these CRYSC '89 Chairman people 968-8141 CRYSCOM Exec 9633045 or _ _ CRYSCOM Chairman 963-4681 Reminder Material from CRYPTOLOG whether classified or unclassified may not be disseminated outside NSA without written permission Address inquiries to the Editor U The remainder of the course probably through the 23rd week or beyond should be full-time Morse with about two-thirds of the EO 1 4 c P L 86-36 4th Issue 1988 CRYPTOLOG page 6 IIAP'WLB YIA SBeRB eOMIP' CIIANP BLS DOCID 4019712 I GON'FIBBH'I'IAL Lotus 1-2-3 A Fantastic Research and Intelligence Tool EO 1 4 c P L 86-36 EO 1 4 c P L 86-36 86-36 P3 Technical Assistant Lotus 1-2-3 is a commercial software package its creators say was designed for accounting and financial analysis But its integrated spreadsheet database and graphics components also make it an ideal piece of software that can be used in the Agencyby traffic analysts net identification analysis and query counting engineers mathematicians typists and managers I J1 Its ready availability on IBM PCs XTs and ATs that are widely distributed throughout the agency the low cost and compatibility with new upgrades of Lotus make it a cost-effective analyst's tool especially welcome in these days of tight budgets U The menu options include • FILE IMPORT- Inputs to a Lotus spreadsheet any ASCII numeric or textual file up to 240 characters per line inputs to other spreadsheet files DBII DBm Visicalc word processor files • PARSING- Allows user to separate any line s of text into separate columns of numeric or alphanumeric characters The parsing line can be stored for reuse on new data • DATABASE- Allows sorting of columns and rows of data permits querying the database using criteria to extract information U Since Lotus is an integrated package any menu option can be used either independently or together with other options on the same data IMPORTING DATA INTO LOTUS FOR DATABASE OR MAP DISPLAY PURPOSES FOUO Lotus can import into its spreadsheet any ASCII textual or numeric file from any Agency database if it has 240 characters or fewer on a line and can be stored on a diskette That means that reports from the field with mixed narrative and tabular information can be processed through LOTUS to pull the tabular information together Then that tabular information can be parsed as'required by any parameter The parsed results can then be sorted by any parameter or saved to another file for database querying U If new reports require similar parsing you need set up only one parse line store it and use it for the new data Once all the necessary data are extracted and stored on separate files these files can be read back into the spreadsheet and combined and sorted as a single file The data can also be queried for specific criteria and the data matching the criteria can be extracted for any purpose such as input to the Lotus map display spreadsheet PARSING • STATISTICS- Finds sums averages and standard deviations for data entered into the spreadsheet • GRAPHICS- Displays Bar Pie Line XY plots on the screen using spreadsheet-entered data ILEO 1 4 c P L 86-36 POUm If you have old alphanumeric textual files you've created with a word processor or on a terminal such as COASTLINE and you want to create a database without retyping the data 1 Lotus parsing is for you Figure 1 shows how 4th Issue 1988 CRYPTOLOG page 7 eONFIBRN't'IAL IIAUBLI3 ViA COr-flU CIIAUUElLS OULY DOCID 4019712 EO 1 4 c P L 86-36 CeNPIBI N'fIAL parsing works on a message where the parameters are separated by I marks and zulu time latitude and longitude are wanted for the database U Lotus tries to up a parse line first line in Figure 1 where L label space and means continue the label until a new symbol appears If the parse line doesn't meet your criteria you may modify it as shown in Figure 2 where S in the parse line spac Lotus then parses the data automatically and lets you store the parsed data anywhere you want on the spreadsheet as in Figure 3 Fig 1 Original Parse Line and Data Fig 2 Modified Parse Line and Data The data in Figure 3 is arranged in three columns and can now be sorted using Lotus' database sort option combined with another file having similar data previously extracted the same way or input to my Lotus Elint Display spreadsheet for plotting I ' 'IS ' uawoase U In Figure 5 the database is sorted with the from column as primary sort DATABASE SORT AND QUERY AND DATA TABLE SETUP TO COUNT DATABASE DATA U Figures 6 and 6a show how the database is queried for information Lo tus requires setting up a criteria and output header with the same titles as those fQr the database Criteria are input in the criteria line for any of these header parameters Results matching the criteria are then di$played on the output lines 4th Issue 1988 CRYPTOLOG page 8 EO 1 4 c P L 86-36 EO 1 4 c CONFIBElN'J'I L WLW-gU VIA CQMlP'fT CHAP'F fEb8 QP'UfyL 86-36 I DOCID 4019712 EO 1 4 c P L 86-36 OONFIBBN I1 1 a • 000 I have just touched the surface of what Lotus can do For example curve-fitting techniques can be developed on a spreadsheet and the input curve to be matched displayed together with the calculated curve message 6 iii U These steps are continued for all parameters in the Data Table until it is filled Analysts can then scan the Data Tableito determine who talks to whom and how many times thus establishing relationships ivery easily editing hard to do on Coastline can be input to Lotus where paragraph-justifying can be easily done and the results passed back as a Coastline file Files too large tI l transmit to a 4th Issue 1988· CRYPTOLOG • page 9 EO 1 4 c P L 86-36 OONFIBBN IAL K1tNnLl f11 eO fII eK1tI I ELB OI 'LY P L 86-36 EO 1 4 c P L 86-36 I DOCID 4019712 e6NFIBRN't'IAL 1 4 c P 86-36 field site via Coastline can first be read into Lotus split into segments and then transmitted via Coastline At the field site the segmented files are read back into Lotus recombined and saved as the original file It is also possible to convert any program binary file into an ASCII hexadecimal file split the file into segments and transmit it to the field site The field site then recombines the file in Lotus and reconverts it to a binary file and uses the program This saves sending new application programs by mail to the field u FOUO The field can also transmit programs not working correctly to NSA where they can be analyzed and transmitted back This saves mailing time usually 3 weeks per mailing updating time and uptime for using the program Bar pie line or XY charts can be developed in a spreadsheet displayed and plotted The data for these charts could be imported from a diskette keyed in by hand extracted from a query from a database setup in the spreadsheet or a sort o the data or entered from a parsed message U In summary this article shows that Lotus can handle many intelligence chores without special programming Before your organization buys sophisticated hardware or software to accomplish a task check into Lotus spreadsheet capabilities You'll be surprised at the power available Try it you'll like it Editor's Note We are advised that ASPIC has replaced Lotus 1 -2 -3 as the standard spreadsheet software with SuperCalc4 but the latter does not have parsing capability and lacks other features necessary for the applications described in this article Lotus 1-2-3 can still be ordered as non-standard software with justification 4th Issue 1988 CRYPTOLOG page 10 6e lFlQ SN'I'IAL IlAUBbB '1M COMHi't' CHAnnELS OnLY P L 86-36 EO 1 4 c DOCID 4019712 e6Nf'I8ftN f'IAL 4th Issue 1988 CRYPrOLOG page 11 HA lBf K G9NFI9KN'I'Jit b W OO r H'P'i'T' OIIAJ NEJ S onJ Y I DOCID 4019712 Once upon a time there was a special school for young artists Only the most promising students were allowed to attend The school had several boxes of crayons for them to use Each box contained eight fine colors The school did not have enough boxes to give one to each student so the children had to share But most of the time that was okay While they were waiting for their turn to use the crayons they could rub their drawings with tissues to make them look smoother or they could think about what they would draw next Sometimes they wished they could have more crayons but they managed to make do The children learned to make wonderful pictures with their simple tools and people began to come to look at their masterpieces hanging In the school's own gallery The schoolmaster was very proud of the gallery He wanted to help the students produce more and better drawings for the visitors to admire So he decided that each child should have his own box of crayons He arranged to buy more The new crayons would be thinner so the young artists could draw finer lines There would also be 16 colors In each box so the children's pictures could be even more wonderful than before These new improved crayons were expensive so the schoolmaster could buy only a few boxes at a time But nobody worried too much because they knew that one day everybody would have his own When the new crayons began to arrive the children discovered that the boxes had room for 16 crayons but two colors were missing from each box The crayon makers had not yet been able to make new Improved versions of blue and purple crayons But they promised to make them soon and send them to the school As the months went by the students were given more of the new crayons Finally there was one box for each student Some had the old ones and some had the new The new 16-color boxes still had no blue nor purple So the children with new crayons had to borrow the old ones when they wanted to use blue or purple in their pictures The children with old crayons had to borrow the new ones so they could by P L A313 4th Issue 1988 CRYPTOLOG page 12 FOR OFFI6IAL USE OP iUt 86-36 DOCID 4019712 An Allegory draw finer lines and use some of the fancy colors The new improved thin crayons were easier to break so everyone had to be more careful And of course everybody had more crayon sharpening to do The schoolmaster declared Now that we have one box of crayons for each student you can draw more pictures faster to hang in our gallery N0 the children explained We cannot We do not really have enough crayons for everybody to use at once We still have to share We also must spend more time taking care of the crayons The schoolmaster thought the children were ungratefUl and greedy He refused to buy any more new crayons unless the children agreed to trade in the old ones He did not want to make the school board angry by having more boxes of crayons than there were children in his classrooms But the children insisted that they could not give up the old crayons because they needed blue and purple to complete their pictures Soon the children were spending less time drawing and more time trying to explain their crayon needs Yet the schoolmaster would not change his mind the boxes of crayons must not outnumber the children But he promised to provide new improved blue and purple crayons even if he had to make them himself then they would no longer need the old ones The children were not convinced but agreed that it Wf uld be just fine if he could really do it Then they could see about getting a complete set of new improved crayons for everybody They just wanted to make sure they had the new blue and purple crayons before giving up the old ones Meanwhile the crayon salesman said that there would soon be no more 16-color boxes of crayons Instead he would be selling boxes of newer more improved crayons in 24 colors Well at least there would be room for that many The crayon experts were still working on the red and orange ones but the salesman assured everyone that these would be ready soon P L 4th Issue 1988 CRYPTOLOG page 13 FOg OFFICIAL USB ONLY 86-36 I DOCID 4019712 Office symbols and organizational designators RESUMES and JOB Also although the terms COMINT ELINT SIGINT etc when associated with NSA are considered FOUO you must be careful when associating them with job titles duties and performed functions APPLICATIONS '-- IDDOICAO During our employment with the National Security Agency many of us will prepare or have already prepared one or more resumes or job summaries in which we must explain in unclassified language what we have done and are doing on the job at NSA This often is a difficult task since much of what we handle tends to be classified All resumes detailed job applications student co-op reports and th like must be reviewed and approved for release as unclassified The Classification Advispry Officers CAO's throughout NSA have been authorized by Q43 to review and release these resumes applications and r ports The five CAO's within the Operations Directorate are listed at the end of this article Any document intended for public release such as a resume may not contain classified information The following additional information also may not be released NSA missions and functions As far as clearances are concerned you may indicate that you are cleared for TOP SECRET Special Intelligence TS SI without further elaboration or detail You may not include any additional specific accesses Revealing that you hold a TS SI clearance may actually be helpful even if the position sought may not require a clearance The fact that the government granted the clearance speaks well for the honesty and integrity of the applicant The following is an updated listing of the Classification Advisory Officers CAO's within the Operations Directorate Please contact the appropriate CAO whenever any assistance is desired or needed concerning proper classification of information or materials or for review and release of a resume applications etc Employee names names of supervisors are releasable employee charact erreferences should not be attributed to NSA employment but to U S Governmentelllployment Project namf 1sand covernames Salaries other than the originator's Number of people employed by NSA P L 86-36 EO 1 4 c As stated elsewhere indications of one's foreign language capability in connection with NSA duties are considered classified this creates problems in writing a resume Association with languages in an academic or training context is permitted but specific languages should be listed only in the academic section of the resume Specific foreign language ability e g native speaker self-taught etc should be listed only in the summary of qualifications or miscellaneous section You may show any and all language training courses as unclassified You may not associate your language capability with the duties of your position Dick Sylvester B03 4053s B CAO 2S010-2 4th Issue 1988 CRYPTOLOG • page 14 FOR OFFICIAL USIil O fL¥ P L P L 86-36 86-36 I DOCID 4019712 THE QUESTION OF LEADERSHIP P L 86-36 PI6 f'OUO During the early 1950's I was assigned to the staff of Commander North Atlantic and Mediterranean for duty with Communications Unit 32D COMUNIT 32D I was assigned to various destroyer division staffs where I taught ELINT countermeasures to radiomen and radar operators aboard three different destroyers All three destroyers had similar missions and almost identical manning U At that time Greece and Turkey were going through a very unsettled period so one mission assigned to the Navy was to show the flag in Greek and Turkish ports - usually spending three to five days in a port or lying offshore before moving on to the next Observing the difference that leadership made on the effectiveness of officers and crew in carrying out their missions gave me some insights into the practical effects of leadership It may have been only coincidence that leadership varied so much from one ship to another but the magnitude of variation made the picture very clear U Taking the leadership of the first destroyer USS GYATT as a baseline the officers and crew seemed to be of average competence and morale was about average for ships in the Mediterranean at that time Command reacted to problems of training and morale rather than anticipating and controlling them Reaction was quick enough to avoid major problems and well guided enough to get the job done U The Commanding Officer CO was not known by the average crew member except as an authority figure He relayed all instructions and commands through his Executive Officer and down the chain of command That is the norm throughout the Navy and works well when exercises are large and diverse enough to leave little slack time It does not work as well when a great deal of time is left unoccupied while supplies are at low levels from lack of replenishment opportunities and recreational facilities during port calls are minimal U The second destroyer was USS O'HARE which we boarded in Gibralter where she arrived after a stormy Atlantic crossing during which an accompanying destroyer was overrun and sunk by an aircraft carrier Our small detachment was very favorably impressed on our first visit to USS 0 'HARE to survey our working spaces The crewmembers on the Quarterdeck the area aboard the ship at the end of the gangway were noticeably sharper and more courteous and the ship was cleaner and in better repair than GYATT in spite of the rough crossing POUO We left Gibralter and returned immediately to the Eastern Mediterranean On the day before our first port call in Greece aboard O'HARE the CO mustered all hands on the fantail the open stern of the ship and delivered an interesting talk on the area that included local points of interest as well as historical highlights and a brief on local customs He emphasized the point that we the crew were to be ambassadors of good will and desired that we make a good impression on the local population He saw to it that sightseeing 4th Issue 1988 CRYPTOLOG page 15 FOR OFFICIAL USfJ ONLY b I DOCID 4019712 tours were organized for those interested in them and arranged for sporting events softball and soccer for others The difference in crew onduct ashore and in our COMUNIT 32D perceptions of this port were radically different from our two visits aboard GYATT During that visit and subsequent ones to other Greek and Turkish ports all preceded by talks delivered by the CO or another of the officers I do not recall a single serious breach of discipline We had been used to seeing at least a few fights and or drunk and disorderly incidents on previous visits of GYATT We soon became aware also of the reason that the ship was in such good repair and clean condition The CO and his officers had fostered a spirit of friendly competition between the two leading Boatswains Mates those Petty Officers responsible for repair upkeeep and seamanship which resulted in fast efficient repair and cleaning of anything in their areas of respnsiblity A spirit of pride in their work and friendly cooperation existed among the other departments of the ship also resulting in the award of several E's official awards for excellence to the ship U U After such an enjoyable tour aboard O'HARE we were ill-prepared for our introduction to USS KNAPP on a miserable rainy day in Plymouth England The KNAPP U I cannot recall ever having seen the KNAPP's CO but his reputation was well known to all of the crew He was said to prefer drinking with some enlisted crewmembers to spending time with his officers by whom he was not well regarded He was alleged to have been arrested for drunken driving by Military Police in Rome where he had driven a jeep which had been assigned to KNAPP from a Navy motor pool in Trieste for use by ship's officers It seemed to us in COMUNIT 32D that the poor crew discipline and KNAPP'S generally rundown appearance and condition were a direct result of the negative influence of the CO U Differences in leadership in the three destroyers was also evident from the examples of seamanship that we observed while aboard GYATT was generally competent and did a good job of maintaining station and maneuvering during exercises and while coming alongside a tender or tying up to a dock or pier O'HARE's crew distinguished themselves by the precision of her movements whether maneuvering in restricted waterways or tying up to a dock in tricky winds and tides KNAPP generally was able to maneuver competently under favorable conditions but her crew was not able to deal well with unfavorable circumstances U One remarkable example illustrates the sent a whaleboat to pick up our unit and when great difference between O'HARE and KNAPP the Coxswain the Petty Officer in charge of the ships with the two most widely different the whaleboat discovered that he was to take commanders Both spent two weeks in a course aboard equipment as well as crew he refused at the Antisubmarine Warfare School in the Foyle River near Londonderry Ireland During to do so After some discussion with O'HARE's Officer of the Deck he agreed to take both the course each had to refuel at the dock in the Foyle River The river at that point is equipment and crew but we were offered no help in loading or unloading When we finally rather narrow and the current swift The maneuver required an upstream approach a got aboard KNAPP we found that our assigned work spaces had no usable antenna connections 180-degree tum and ended with coming and the antennas that had been installed some alongside the fuel dock from upstream O'HARE time before were not in operating condition made the upstream approach and dropped her We had to do all of the repair and installations port anchor when about two ship-lengths ourselves upstream of the fuel dock reversed one engine and swung around the anchor like a child U It came as no great surprise after that to around a lamp post and ended in perfect see the large number of crewmen returning position to back upstream against the dock aboard drunk from liberty in Plymouth and KNAPP made her upstream approach and turned other places later There was rivalry between without using her anchor which left her factions of the crew of KNAPP but it was not drifting into the fuel dock as she attempted to the friendly rivalry we had observed in O'HARE back alongside Irish linehandlers on the dock KNAPP's crew often were involved in drunken were forced to scramble quickly to avoid KNAPP brawls between factions as she tore up six to ten feet of the pier before stopping 0 4th Issue 1988 CRYPTOLOG page 16 FOR OFFIeIAL USE ONLY DocrD 4019712 Secontis swiftfy speeaing past Jfours only minutes fast Stufaen stop return anaseet for tliat tone -- tliat voia unique Trivia fore anti infonnation Passing 6act anafortli 'tween nations -Sentiing 6act inte igence To tlie Operations gents Letting tliem put out tlie story fu of t nowfeage grim anagory Qy icf y now you liit repeat itntl your ears in for a treat '£ntifessfy repeat tlie plirases-'1 0 it 'tif your vision gfo zes-fJ f ver min justget tlie souna sy a6fes you move arouna 'Ttl you sofemnfy ao swear %Wt you liear is wliat is tliere Tlien you caf tlie otliers over Previous grouna witli you tliey cover 'Ttl you Jinaffy agree Its an 'I' anti not an ' £' j fter a fang typing 6out l ouget tliat 6fastea transcript out g22 1 -_ _ 4th Issue· CRYPTOLOG· 17 Q FIQ TM K Qb VIA 9 U T 11 l Li 9 LY P L 86-36 I DOCID 4019712 OfficeWriter the agency standard Output files match the format of the input files Software Review We found the program easy to use with a better-than-average user's manual It is designed as an aid for business and technical writing and we highly recommend it for that purpose The writing hints in the user's manual would benefit other writers as well In summary a good buy Note the passage delineated by lines was run on RIGHTWRITER The results are shown on the facing page System requirements RIGHTWRITER by Rightsoft Inc 2033 Wood Street Suite 218 Sarasota Florida 34237 GSA price $56 Upgrade $5 additional Reviewed by il Ip13 P13 has been reviewing recent software that DDO analysts -might consider buying One of these is RightWriter which evaluates a document for grammar style usage and punctuation The program creates an output file consisting of the original text interlineated with comments and suggestions It provides an overall critique a measure of the grade level of the writing and the strength of delivery It flags jargon and cliches and concludes with a list of words in the document that should be reviewed There is also an option to provide a frequency count of all words in the document Minimum Internal Memory 384K RAM Minimum Disk Configuration Two dual sided diskette drives or one diskette drive and a hard disk • Operating systems PC-DOS Version 2 0 or higher MS-DOS Version 2 0 or higher OS 2 Computers IBM PC PC-AT and Ps 2 and all MS-DOS compatibles Special versions are available for DOS earlier than 2 0 and for memory sizes smaller than 384K WE WELCOME REVIEWS of books articles software audio cassettes and video cassettes that relate to any of our disciplines or that pertain to our mission_ The manufacturers claim that the program is compatible with most word processors including P L 86-36 4th Issue 1988 CRYPTOLOG page 18 POR OPPl6IAL USB ONL¥ DOCID 4019712 Output from RIGHTWRITER r •••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••••• « » SUMMARY Overall critique for D RIGHT soft2 txt Output document name D RIGHT soft2 0UT READABILITY INDEX th 8 87 6th 8th 10th 12th I I 1 1 1 1 1 1 l th I I SIMPLE I ------ GOOD ----- I Readers need a 9th grade level of education STP£NGTH INDEX COMPLEX 0 92 0 0 0 5 1 0 1 1 1 1 1 1 1 1 1 1 WEAK I STRONG The writing has a strong style DESCRIPTIVE INDEX 0 60 0 1 0 5 0 9 1 1 1 1 1 1 1 1 I I TERSE I ------------ NORMAL ------------ I The use of adjectives and adverbs is normal JARGON INDEX I WORDY 0 00 SENTENCE STRUCTt E RECOMMENDATIONS 15 No Recommendations « WORDS TO REVIEW » Review this list for negative words N jargon J colloquial words e misused words M misspellings or words which your reader may not understand RightWriter 1 compatible 1 critique 1 evaluates M 1 interlineated 1 jargon N 1 reviewed 1 « END OF WORDS TO REVIEW LIST » « END OF ••••••••• It It It • It It It It It It It It It It It It It It It It It It It It It It It It 4th Issue 1988 It It It It It It » S Y It It It It It It It It It It It It It CRYPTOLOG page POR OPPICIAL USB ONLY It It It 19 It It It It It It It It It It It It It It It It It It It It It It It It It I DOCID 4019712 CONFERENCE REPORT worked on as well as an inspiration But traditionalists resist The establishment was clearly trying to show that it realizes applied science has a role to play EVALUATION I think that conferences like this are a convincing argument for the importance of having Agency people travel to at least selected conferences I would hope the Agency would continue to try to keep our program of attending outside conferences alive despite the budget crunch In my view it is foolish to spend money on supporting academic research and fail to gather up the material that is presented to us at conferences for such relatively low cost It would be self-defeating to let our ties to the outside atrophy for lack of money All our previous efforts and expense would quickly be wasted The American Mathematical Society Centennial Celebration August 1988 Providence Rhode Island Reported by SUMMARY David Harris R51 Several Agency personnel attended the Centennial Celebration of the American Mathematical Society The celebration featured invited talks by eighteen prominent mathematicians surveying the present state of American mathematics It also featured some discussion of where mathematics is going a discussion generally dominated at this conference by the applied types interested in improving the ties of academic mathematics to the users of mathematics in the real world be they DoD physical and social scientists or industry The left wing of the academic world winner in last year's referenda on relations with DoD was conspicuously quiet The managers of the society were apparently minded to try to undo some of the political harm from last year's debates by advertising and supporting better relations with applied math and its sponsors The world of mathematics nearly always has known some conflict between applications-driven mathematicians say Herman Weyl and purists say Bourbaki Some of the best math now is being done by non-mathematicians Organizers of this conference made an apparent effort to sell applied areas as a source of good math and good problems Applied scientists can be a source of wisdom on what problems should be 4th Issue 1988 The technical highlights of the conference included an excellent minicourse by Gene Luks on Groups Graph Theory and Computer Science and a related talk by Aschbacher on recent developments in group theory exploiting the classification theory very interesting talks on the interaction of physics and mathematics by Jones Kac Uhlenbeck and especially Witten a talk by Gross on modular forms and elliptic curves a somewhat disappointing minicourse by Assmus and Key on Coding Theory and Finite Geometries and talks by Caffarelli Diaconis Fefferman Freedman Friedman J Harris Howe Majda Peskin Sullivan Tarjan Thurston Bott and Lax In my personal opinion the most awe-inspiring talk was Witten's on the relationship between quantum field theories and the future development of mathematics He argued convincingly that many of the recent breakthroughs in math can be understood in terms of quantum field theories and that theoretical physics can serve as a guide to new directions for mathematics This possibility was illustrated by Jones' talk on knot theory Kac's talk on modular forms and by Donaldson's work on 4-dimensional manifolds For details of the technical talks please see my trip report Perhaps equally important was the political side of the conference The political talk at the meeting was for once generally pro-DoD and CRYPTOLOG • page 20 FOK OFFIGIAb USE ONb¥ I DOClD 4019712 anti the usual academic mathematical establishment Especially outspoken were the talks by Peter Lax a former president of the AMS and Edward David otherwise known as head of a commission that helped to convince the government to increase its support for academic mathematics They both had many critical things to say about the lack of effort by academics to help deal with the current and future crisis in recruiting more American-born mathematicians David stressed the dangers of getting involved in politics Once the math world declares itself political on any issue politicians will cease to view its opinions as based on expertise and will instead judge it as a political ally or enemy This has heavy long-term consequences for mathematics support by government David mostly gave the academics the bad news that they are failing to do their share of the job and that they had better get out of politics and get to work in preserving mathematics as a profession The mathematicians were repeatedly told that math is in competition with welfare and housing the poor So mathematicians must expect to be held to meeting their social obligations if they expect funding Lax gave what the president of the AMS called a blunt speech roundly criticizing many in both the pure and the applied mathematics communities The anti-DoD resolutions of last year have clearly left much hard feeling and the establishment was keeping its mouth shut to try to avoid making further trouble Another interesting event at the conference was a JPBM-sponsored meeting on the MS 2000 program and the abuse of MAs as teachers in academia It was never quite clear however whether the intent was to deplore the practice of exploiting masters holders by using them as underpaid temporary help without benefits or whether the intention was to educate department chairmen on the best way to do exactly such exploitation As usual we engaged in a considerable amount of informal discussion with academics both for purposes of improving our general image of finding out what is going on in academic mathematics and of ultimately aiding recruitment 4th Issue 1988 POUTICAL SPEECHES Opening Ceremonies Perhaps one could look on this opening ceremony as indicative of the lack of understanding by laymen of what professional academic mathematicians do and how they view themselves Neither the governor nor mayor showed up for the opening ceremonies despite the fact that AMS is an employer in Providence The governor sent his policy advisor Valerie Southern The mayor sent Joseph Almagno the Superintendent of the Providence School Department apparently under the mistaken impression that the members of AMS are teachers Brown University sent Phillip Stiles Dean of the Graduate Schoo1lResearch The London Math Society and the MAA e7 changed presents with AMB SIAM accepted a present from AMS but gave none that I know about in keeping with the general skepticism by applied mathematicians throughout the meeting about the attitudes of the pure mathematicians of AMS President Reagan sent a letter indicating his staff has some awareness of AMB's activities and stressing the role of mathematical skills' in mai ntaining our domestic prosperity our national security and our competitive stance in the world marketplace That's a challenge I know you're ready to meet Not if the members of the AMS can help it Governor DiPrete could not come because of the National Governor's Conference His aide showed little appreciation of the noble role pure mathematicians assign themselves She spoke of the 50% high school drop out rate in Pr0vidence of the remaining students that cannot really read or write and want less technical schooling DiPrete opposes this and wants more engineers biomedical workers scientists The members of AMS are the role model for them DiPrete wants to forge a partnership to guide our children and society President Mostow reacted by calling this a message we rarely get but appreciate i e don't call us about such a partnership we'll call you Next Joseph Almagno spoke He said we need you to help public education The public is losing confidence in our product We need to retool reshape and find better methods He CRYPTOLOG • page 21 FOR OFFICIAL USE ONLY I DOCID 4019712 apparently thought he was addressing an audience interested in education He did not understand the role of the AMS as inventors of the truth of pure mathematics Instead he hoped they could help him with the problem of non-English speaking students largely Hispanic and Southeast Asian How can we teach them Apparently the differential geometers and algebraic topologists should show the way Phillip Stiles from Brown spoke of the long record of collaboration between AMS and Brown Brown reaffirms its commitment to math and its outstanding math library by which he meant they had recently bought a large collection of ancient mathematics manuscripts Brown University Mathematics Dept was conspicuous by its lack of active participation as such at the meeting For example there was no effort to encourage mathematicians to visit the department A search on foot allowed me to locate the Brown Mathematics Department building in an outwardly unprepossessing old Victorian house This was rather in contrast to the Applied Math and Engineering buildings or for that matter the History of Mathematics Dept quarters choice of talks precisely the importance of nonmathematicians to the future of mathematics Finally President Mostow spoke He said that math today is so diverse that no one person can describe it all Few people know the breadth and depth of the influence of math The computer revolution shows if anything the invisibility of math in the world at large The educated public does not share our enthusiasm for the subject We must make efforts throughout education to find opportunities to expose people to real mathematics AMS must get involved in this Math will be permeating the world So we are feeling our way towards a solution Future textbooks on the history of ideas should give mathematics its proper place This will encourage youth to go into math The role these textbooks give math is totally opposed to the role most academic mathematicians would like to have and are willing to fill How do you keep this from turning off those students you have educated on the role of mathematics in the real world Edward David Renewing U S mathematics An agenda to begin the second century After the Presidents of AWM and the London Math Society had congratulated AMS and given it some gifts Bill Gear of SIAM spoke Math is the infrastructure of our future progress SIAM and AMS must work together We are all developing countries and mathematics is at the heart of the process We must have a literate population applying math to the problems of the real world The distinctions between engineers scientists and mathematicians should be blurred and we should all work together unlike the traditional disdain of mathematicians for all these others The highlight of the political activity at the meeting was the address by former Presidential Science Advisor Edward David He is remembered as the leader of a committee that wrote the David Report see my Trip Report on the AMS Meeting in Louisville 5 years ago David was helpful in selling the government on increasing funding for mathematics drastically Now he has come back to AMS to lecture the mathematicians on their failures as citizens and professionals In particular funding is now in the hands of democratic institutions which means that sciences must work to inform the public if they want money Show that math is significant to the public Integrate math into the world All these points were repeatedly made by the applied mathematicians at the conference but are distinctly distasteful to the true mathematicians They are at variance with the sentiment expressed in the overwhelming support for the anti-DoD resolutions of the AMS last year The organizers of this conference however did seem to stress by their He started by expressing his concern about how mathematics deals with its customer activities After five years of pushing how have we done Substantial progress has been made but there is a long way to go and the hour is late Frank talk is needed In the last five years math support has increased considerably You have been winning in the fixed sum game over your competitors such as housing for the poor and aid for the starving But the major issue was renewing mathematics and enabling it to sustain itself That issue is more pressing than ever In 1983 there were 800 new Ph Ds Now CRYPTOLOG page 22 FOR OFFIGIAL USB ONLY 4th Issue 1988 DOCID 4019712 there are 700 a year and half of those are foreign The increase in funding has been almost equally from NSF and DoD NSF gave high priority to support for grad students and post docs Unfortunately they did not help senior investigators But the problem is bottoming out salaries are skyrocketing and this may change the common perception that the world does not want any more mathematicians The David Report asked for actions from universities in support of mathematics but has gotten little such support from them To the Universities mathematics is not as worthy as other subjects if it does not bring in as much money Mathematicians must unite to educate the world on the value of math and on its needs The world thinks math exists there is no need for any more mathematics just teach the math already known Ken Hoffman's lobby has been quite good at fighting this attitude Of course Ken Hoffman's lobby has also taken a lot of heat from the Thurston branch of AMS precisely for selling math as useful to the real world in ways that many mathematicians oppose research There is a need to know what problems both the math and the user world think are important What do your results mean for these other people This may be done post hoc rather than ad hoc Do the research then look for how it may be sold to the outside world as useful AMS must work on its relation to the public if only out of concern for keeping its share of the discretionary budget We need political sophistication and cohesiveness and must consider our responsibilities to society and how we look to society David said he is not calling for overcoming political activism Mathematicians may offer advice to government in statesman-like ways But mixing science with politics is poison In Washington if you try to use the prestige of mathematics to decide nonmathematical political issues you will be treated as enemy politicians and all your views will be rejected as political Mathematicians must find better ways to attract young people to our field - NOT as currently take it or leave it The young have been leaving it in droves for adjacent fields Even those in adjacent fields are appallingly ignorant of what mathematicians really do Next David addressed reform of education Two Most students think we aim at winnowing out boards of the NRC are currently working on the weak and grindjng down the ungifted school curriculum reform and informing They are right Most outstanding mathematicians of opportunities for crossmathematicians view the world differently from fertilization This is the MS2000 project But David not a mathematician There is only a there are very tough political realities There comparative handful of people in any are budgetary restraints Federal R E is generation capable of making important already 25% of the nondiscretionary budget breakthroughs in theoretical pure mathematics Math is competing with social programs like They are of interest to us The rest are a housing and welfare Science has incredibly waste of our time Academic education should been winning in this competition But math be a pump not a filter cannot expect another 84% increase in the next David says we must improve mathematics five years The highest priority is of course sustaining the quality of math research There education in our schools We are a part of the educational crisis The country is failing to should be an update of the David Report every educate youth in mathematics We should help five years to document progress evaluate university instruction and overview research - and if we do it will be appreciated Twenty opportunities in the view of mathematics users five million kids take mathematics in school It should discuss how math should be supportthat gives mathematicians a big responsibility ed A mechanism for such an update is needed Elsewhere at the conference the New Math efforts of mathematicians to reform curriculum David said AMS must overcome last year's were discussed These lead to doubts on this combative attitudes and recommit itself to its point If mathematicians seek to make pure goals in Washington Micromanaged mathematicians out of the masses they may programmed government-sponsored research is end up messing up math education even worse inevitable As at Bell Labs perhaps What is needed is an effort to talk with the government can choose the research field but masses in ways they will understand and find let scientists determine the direction of the 4th Issue 1988 • CRYPTOLOG • page 23 FOR OFFICIAL USB O iL 'I DocrD 4019712 convincing It is rather doubtful that there are many members of AMS wanting to do this to see the interesting architecture My wife spent much of her time in nearby Newport In conclusion David said our success in renewing mathematics has been mixed at best We need new doctorates and we must give them reason to believe that going into mathematics will give them a reasonable future This will require the members of AMS to interact with schools other disciplines etc Are they willing There was an AMS short course on chaos and fractals that I did not attend AMS BUSINESS MEETING and STEELE PRIZES There was a daily newsletter throughout the conference featuring such information as that Rhode Island was the first state to attempt to tax sexual intercourse They also advised when driving in Rhode Island never ever beep your horn at black Cadillacs with blackout windows INVITED TALKS There was an amendment to the bylaws voted Michael Aschbacher Cal nst Tech without controversy to change the way that of finite groups as permutation Representations dues are computed for institutional members groups introduced by Daniel Gorenstein The effect is to break the connection between dues and the number of published articles from The classification of the finite simple groups in the institution to the probable detriment of 1981 changed the landscape of finite group non-academic institutional members like the theory and led to an increased effort to describe Agency Several long-time employees of AMS the structure and representations of the simple were thanked Then the Steele Prizes were groups Together with the classification this awarded Sigurdur Helgason won the prize for effort has made possible unexpected applications exposition Gian-Carlo Rota won the prize for a of finite group theory in other branches of fundamental research paper for his work on mathematics In 1860 the French the use of the Mobius function in combinatorial mathematics society offered a prize for the best theory and Deane Montgomery of IAS won the research on the following problem what are prize for cumulative influence on the profession the possible indices m of subgroups of the Paul Nevai with whom I have had some symmetric group of degree n and for such an interactions on the subject of orthogonal m what are the subgroups of index m polynomials got a job on an AMS committee Several great men published work on the Mel Hochster whom I knew in college is one problem but understandably made little of the candidates for Vice President William progress and no prize was actually awarded Browder will become the new President of AMS The purpose of Aschbacher's talk was to show in December that finally we are beginning to be able to do At the MAA Business Meeting Steven Galovich this problem and to illustrate how to use the classification theorem and Bart Braden won Carl Allendoerfer Awards James Epperson and Stan Wagon Luis Caffarelli lnst Advanced Study The received Lester Ford Awards Dennis Luciano geometry of solutions to nonlinear problems Gordon Pritchett and V F Rickey won Georg introduced by Louis Nirenberg Polya Awards Seven Certificates of Meritorious Service were presented Persi Diaconis Harvard Sufficiency as MISCELLANEOUS AMS was organizing tours of the AMS Headquarters building in Providence While I did not go on these tours I did manage to go to the headquarters building to take a look around on my own Not nearly as interesting as the wonderful little Rhode Island School of Design Museum of Art I also recommend walking tours of College Hill and Federal Hill statistical symmetry introduced by Gian Carlo Rota Charles Fefferman Princeton Problems from mathematical physics introduced by Felix Browder This session covered two problems in mathematical physics The first is from quantum mechanics and concerns how close our current state-of-the-art will take us to explaining why a roomful of hydrogen gas can 4th Issue 1988 • CRYPTOLOG • page 24 FOR OFFICIAL USB OP LY - - - - ------------ DOCID 4019712 happen The answer is quite close The second is from general relativity and concerns a proof that some small enough initial disturbance will not become a black hole As such it deals with a quite difficult analysis of the possible nature of solutions for a PDE Michael Freedman UC-San Diego Working and playing with the two-dimensional disk introduced by William Browder Harvey Friedman Ohio State The incompleteness phenomena introduced by Saunders MacLane Benedict Gross Harvard Modular forms and elliptic curves introduced by John Tate This session surveyed some major developments in the theory of elliptic curves the theory of heights and rational points Gross is married to Jill Mesirov whom some of you may remember According to Tate they had a child two weeks before this conference Joseph Harris Harvard Developments in algebraic geometry introduced by Phillip Griffiths Roger Howe Yale A century of Lie theory introduced by George Mackey Vaughan F R Jones UC-Berkeley A von Neumann algebra excursion from quantum theory to knot theory and back introduced by Joan Birman Victor Kac MIT Modular invariance in mathematics and physics introduced by Nathan Jacobson Upon hearing this talk I remarked that some deep mysterious truth was going on here Witten's talk later explained to me what it is Kac began by asking us all to sign the petition on behalf of Soviet emigres available at the registration desk a partial reminder of the role of immigrants in this celebration of math in America Andrew Majda Princeton Mathematical fluid dynamics the interaction of nonlinear analysis and modern applied mathematics introduced by Peter Lax Majda advertised the importance of an interdisciplinary approach to applied math problems These problems are too difficult to be handled by anyone method but can be reasonably well dealt with sometimes by combining attacks via new computational methods to guide research experimentation with physical models for example suggestive rigorous theoretical analyses of simplified mathematical prototype models etc Thus fluid dynamics is governed by a complex system of nonlinear PDEs with features of all kinds of PDEs The key new ingredient allowing the interdisciplinary approach is the use of computer simulation to guide research and test models and conclusions when experimentation is unreasonable This situation is a paradigm for even worse applied problems such as oil recovery combustion elasticity and magneto-fluid dynamics This use of gas dynamics dates back to ideas of Lax Charles Peskin Courant Institut Mathematics and computing in physiology and medicine Biomathematics from Hearts to Molecules introduced by Cathleen Morawetz Peskin gave yet another talk on the advantages of mathematicians working on applications rather than pure math Here the problem is the mathematical modeling of functioning human organs and in particular the designing and testing of heart valve replacements Several articles of this sort are in the latest engineering journals Similar problems arise in constructing oil rigs that have parts subject to both slow and fast vibration Also molecules undergo slow and fast vibrations Peskin described the problems of modeling heart valves and illustrated with films done by computer simulation showing how hearts with various kinds of replacement valve would in fact function Several possible valve designs can be studied this way and dismissed as defective or improved One molecular chemistry problem being modeled is how large molecules like DNA control the production of RNA which creates protein enzymes which in turn control the reproduction and action of the DNA The Genes in Action model is by no means done but is already being used to study the mechanics of molecules Again the so-called stiffness problem caused by the differing time scales within which nerve impulses and DNA operate is the difficulty in doing the modeling 4th Issue 1988 • CRYPTOLOG • page 25 VQR QVn I L TJili O LY I DOCID 4019712 Dennis Sullivan CUNY Graduate School Progress on the renormalization conjectures in dynamical systems introduced by Stephen Smale Robert Tarjan Prof Computer Science Princeton athematics in computer science introduced by Ronald Graham After a general introduction to theoretical computer science and the development of effective algorithms Tllljan spent most of his time talking about his work on trees and operations on trees He introduced rotation as a mechanism for balancing trees and keeping searches cheap He argued then that balancing is not in practice a good idea A possible answer is splaying Sleator-Tllljan 1983 William Thurston Princeton Threedimensional geometry and topology introduced by Lipman Bers Bers introduced Thurston with glowing praise The talk he gave merited the praise Karen Uhlenbeck Texas Instantons and their relatives introduced by S S Chern Edward Witten Prof Physics -mst Adv Studies Quantum field theory and Donaldson polynomials introduced by Clifford Taubes To me at least Witten's talk was the high point of the entire conference giving a glimmering of where the grand synthesis will come from that Uhlenbeck Kac Jones Tarjan Thurston etc are looking for in their work The point is that much of traditional mathematics can be looked at as related to 1dimensional quantum field theory The recent work of Jones and Donaldson are the first inklings of 3- and 4-dimensional quantum field theories The Kac and Bott talks as well as work by Gromov in Thurston's talk are inklings of a 2-dimensional theory So there is at least the possibility of inspiration for vast new areas to develop in mathematics out of this connection to quantum field theory Witten's talk was a sketch of this connection Raoul Batt Harvard The topological constraints on analysis introduced by Andrew Gleason Peter Lax NYU Mathematics Applied and Pure introduced by George Mostow Alternative titles from Lax The Flowering of Applied Math in America or The Deflowering of Pure Math in America Mostow summarized Lax's talk as 'blunt' It was He said a lot of impolitic things that ought to have been said This talk was part of the continuing counterattack by the applied people against the disdain for applied math by the academics Lax started by mentioning the Bourbaki attitude that once dominated pure math calling for math isolated from the rest of the universe and internally guiding its own future growth Lax said the great mathematicians of the past would have been horrified Math and Applications are equal partners feeding each other controversy in the audience Before 1945 applied math was treated shabbily in math departments despite some great examples Lax launched into a survey of recent work in applied math Equal to the gains from computing caused by improved hardware has been the gain from clever software algorithms code discretizations etc Computing forms an additional bond between pure and applied math Integer programming dynamic programming optimization problems and simulated annealing have become major fields Lax described simulated annealing at length especially attacking the traveling salesman problem via the Metropolis algorithm He mentioned FFTs the simplex method fast matrix multiplication Karmarkar's algorithm and Knuth's book as major subjects Lax then described Sarnak's work on network design which is done using the theory of automorphic functions and the Ramanujan conjecture Then he talked of Smale's work on average case complexity although he acknowledged that there are not many answers yet in computational complexity Lax then lambasted AI and catastrophe theory He said AI set its goals preposterously high and exaggerated its achievements He was particularly hard on Herb Simon's talk on BACON at the AMS meeting at Louisville I worked over that talk myself in a trip report Lax cited several other skeptics on AI He had similar harsh words for the advocates of catastrophe theory Lax is a believer of sorts in the theory as such But the hostility of many is caused by stupid attempts to oversell what catastrophe theory is CRYPTOLOG page 26 FOR OFFICIAL USE ONLY 4th Issue 1988 I DOCID 4019712 Zeeman was an honored guest at this conference representing the London Math Society Lax quoted Zeeman at length on the virtues of catastrophe theory as a breakthrough and then tore the quote apart For example he pointed out that the study of shock waves for PDEs dates back to Riemann and is older than AMS or even the London Math Society of which Zeeman is president He expressed skepticism that Zeeman did not know this The methods of applied math are sometimes rigorous when this is practical which is far from always Otherwise it must rely on asymptotics experiments simplifications to inspire intuition for further research In the days before computers these simplifications may have helped cause the bad reputation of applied math and the unwillingness of pure mathematicians to work in this area World War II demonstrated the importance of math to the real world which led to support for both applied math and math in general The US Government provided this support at first through ONR then AFOSR and ARO then through DOE the successor to AEC NSF took over much of this funding later on The AMS proposal to reduce DoD funding is offensive to many of us It implies we are doing something wrong when we accept government money to fund our work We object Immigrants to the United States were responsible for pushing applied math in the old days To them it was inexcusable not to look to the applied world for advice on what problems math should be confronting For example von Neumann and Weyl were dominant figures in American math that felt this way Finally Lax spoke of the need to change our educational system As for calculus he said he was one of the doubting Thomases groans at the pun Lax thinks we should replace calculus by a new style calculus course oriented towards applications UMAP can serve as a source for such Saunders MacLane Chicago Some major research departments of mathematics introduced by Leonard Gillman MINICOURSES l attended two of the seven ContributiOl S of algebraic coding theory to finite geometry E F Assmus Lehigh and J D Key Bryn Mawr Handout surveying parts of the course is available This minicourse suffered from the ravages of the MAA support personnel They changed the room for the course without telling half the participants As a result half of us missed the first half of the course The second half of the course had all the students but was held in a ludicrously hot and unfavorable classroom environment After an hour Key wilted to the exteht of not being able to go on After another hour Assmus was soaked to the skin Why the originally scheduled airconditioned room was not used I do not understand The course served as an introduction to linear codes and their connection to designs and finite geometries Groups graphs and computing Eugene M Luks Oregon Handout rather incoherently surveying parts of the course is available In addition to the handouts that were supplied at the course Luks mailed us a better version one representing properly the various overlays he used with his overheads as well as a reprint of an article '1somorphism of Graphs of Bounded Valence Can be Tested in Polynomial Time that led to some of this work and that originally appeared in August 1982 in Journal ofComputer and System Sciences Though it was given in the same impossible environment as the second half of the Assmus course this course went splendidly The classroom was not quite as unpleasant But mostly this was because of the presence in Luks of a true expert lecturing about the stateof-the-art of an important subject that he clearly loves The course dealt with a paradigm for solving computational group theory problems together with many examples Ultimately the goal is to come up with ways to use the classification of finite simple groups and divide-and-conquer to improve computer algorithms The role of the connection between group theory and graph theory is central to how Luks suggests doing this 0 page 27 OFFICIAL USE Or LY 4th Issue 1988 • CRYPTOLOG I OR DOCID 4019712 P L I 86-36 QUOTE WITHOUT COMMENT Extract from BANANA The Story ofa Fruitful Diagnostic Enterprise G42 TSR0588 March 1988 byl IG424 Cryptologic diagnosis is not an activity for the insecure It may be thought of as a mental wrestling match against an opponent who is assumed not known to have some weakness The diagnostician must be prepared for a long possibly futile struggle He must have enough confidence in his abilities to mount a relentless attack even though the referee may be the opponent's mother Rather than give up in a particularly tiring match he merely takes time out to face other opponents only to return with some new moves To the Editor U In response to your editorial in the 3rd Issue 1988 I wanted you to know that at least one Key Component T has a Yellow Pages Here's a copy for you Others can be had for the asking Jim Devine AIDDT The Editor replies U Thanks And dear readers some of the pages are yellow To the Editor lOUO laiHci on the language programs 2nd Issue 1988 was very good My experience with the language reserve program leads me to believe that is should be more widely advertized among the managers not just the retiring linguists I also senlte that most managers and linguists interpret the wording the director can call upon in emergency situations to mean crisis or almost warime situations In my case in G Group it was a critical shortage in the field and the need to fill the spot for a longer than average TDY period e g several months Unlike most other members of the cryptanalytic community the diagnostician is not armed with the hard knowledge of his opponent's game plan-no sacred diagrams He must possess an even more thorough knowledge of principles and techniques than his nondiagnostic colleague in order to seize the advantage on his adversary To add insult to injury his successes seem trivial to those who have never been in his shoes Diagnostic success is often shrouded in secrecy making it even more difficult for the diagnostician to enjoy the cryptanalytic limelight With his chances of fame and even respect somewhat limited the diagnostician presses on secure in the knowledge that he is working in the most difficult and creative area of cryptanalysis FOUO P L FOUO Thanks to the willingness of the linguist the preserverance of the managers and assistance from P16 we were able to overcome the initial hesitancy from some quarters to defining our language problem as fitting the criteria of emergency and of requiring directorate approval It was a successful endeavor I erceptions as SUSLO-2 U were very interesting also 1-- • 4th Issue 1988 • CRYPTOLOG • page 28 FOR OFFI6IAL USE ONLY P L 86-36 8 6- 3 6 DOC 'DL · 84-019712 LETTERS 86-36 To the Editor I recently received a copy of the article A Reform in Need of Reform bylL - _ published in CRYPTOLOG 2nd Issue 1988 Some of the things I read truly amazed me and I would appreciate an opportunity to offer a moderately vigorous rebuttal The points with which I wish to contend are synopsized below developer happens to think of something else the software should do the developer will just add it in Conferring continually with the customer the teams will carry on adding functions as knowledge is gained correcting course as necessary and documenting as little as possible until the team and the customer agree that the job is done The article suggests that if we follow this formula the software developers will be productive and happy the customers will be undisappointed and the malaise that has so cruelly affiicted our Office and Group Chiefs will be forgotten as they busily reallocate something like seventy-five to eighty-nine percent of their software acquisition budgets The author believes that the current NSA software acquisition methodology is hindering software acquisition that our software customers are well aware that the process is not working that the finance people know we are paying more for less and that our Office and Group Chiefs feel weak and uncomfortable The method that the author proposes to replace because all they can do is throw their hands up current software acquisition policy seems and go for a quick-response solution In a laissez-faire to an extreme Nothing could be curious contradiction the author acknowledges planned more than a few weeks in advance that the software being delivered today is because the customer might change his mind or cleaner and easier to maintain than in the there could be a dramatic technological past but in terms of costs the trade-off is breakthough that will redire t the whole unacceptable project Without a plan there is nothing to schedule besides nothing is ever going to get The article states '1ncontrovertibly contractor- done on schedule anyway It would be wasteful developed software comes at a much higher cost to document before the software is finished than in-house developed software software because there are no firm requirements and no developed out-of-house is costing the Agency one really knows what the finished product is perhaps four to nine times what it might supposed to do Never mind that the software otherwise cost developers unstifled by the need to document and uncontrolled by firm requirements may The author proposes that we produce software strike out on some wild tangent that comes to using a method loosely patterned after a level a deadend and may need to reestablish what of effort contract As I understand from the was done two weeks earlier but cannot article software developers should be assembled remember how they did that last tricky little into small groups and assigned small portions that got altered during the wild algorithm of a large system The software developers will tangent Just continue this method and one get a general idea of the customer's problem fine day the project will be finished produce some software that they think is approximately what the customer wants and demonstrate it to the customer If the result is not exactly what the customer wants the developer will simply rework the software until the customer is happy If the customer or the I In his article limpeached for reasons of greed or self-interest any witness who supported NSA software acquisition methodology Before I try to counter some of 4th Issue 1988 CRYPTOLOG page 29 FOR OFFIOIAL USE ONLY P L 86-36 I DOCID 4019712 the positions taken in the article let me defend myself against the charge of greed I am never enriched by prolonging a project I write no textbooks I teach no courses I am a software acquisition manager and my rewards such as they are depend upon my ability to acquire software within cost and schedule I confess to the charge of self-interest in this matter but because of that self-interest I am strongly motivated to transfer allegiance to any doctrine that promises to help me develop software within cost and schedule limits If there is a known method by which we can produce software on schedule at one quarter the cost of acquiring it under contract I am willing to adopt that method instantly Maybe the author is correct and we have been wasting three-quarters of our software budget I certainly do not have figures to refute this estimate does anyone know an accurate overhead rate or an authentic productivity figure for NSA software developers Maybe the author can help me out here Could I possibly see the research data that was developed to support this estimate Perhaps the author will be surprised to learn that many of the contractors I have dealt with recently stand four-square on his side If the contractors are being enriched by our methodology and documentation requirements they are almighty ungrateful the cry against NSA documentation requirements is almost universal among contractors Most contractors have their own way of doing business and are eager to demonstrate that their way meets the intent of NSAM 81-2 and NSAM 81-3 They want us to accept their standards and practices precisely because they are familiar with them and we probably are not They object to the amount of documentation we require Usually they cite prohibitive cost but I suspect they balk for the same reasons government developers do they do not want those nosey parker software managers telling them what to do According to the article the NSA acquisition policy reduces the productivity of our newlyhired software producers These people have had the benefit of the formalized experiences of their predecessors and are far better prepared than their predecessors to analyze design implement test and document yet they chafe under the yoke They are stifled the documentation requirements are stultifying and the methodology is demotivating sic They yearn to be free from the oppression of the wearisome NSAMs which xpect from them either too much or too little depending on how you read the author's remarks On one point the author is clear Agency J methodology is J in flagrant disregard of J Maislow Herzberg and Argyris I fail to see how our methodology violates the teachings of all the Management Hall of Famers Using it software managers can provide software developers with clear direction unambiguous measures of progress and frequent indications of accomplishment and success All of these are Good Things according to the Management Meisters It seems to me that drifting through a project with no clear indication of immediate or longterm goals continually trying to patch new requirements into the software with no real sense of accomplishment and no end in sight would demoralize the new-hires more than having to document their work I agree that today's developers are trained to analyze design implement test and document NSA software acquisition policy provides a framework of discipline that guides them while they do what they are trained to do If so much is wrong with our policy and if so many people including the newly hired the contractors the Data Processing customers the financial people and the author of this article object to it why does it remain in' place Who supports NSA software acquisition policy I think most software acquisition managers do I think the software support managers do I hope and trust that my upper management does Among us we may often hold spirited and enthusiastic discussions concerning the proper interpretation of finer points in 81-2 and 81-3 but we have in common certain beliefs software acquisition requires firm discipline Life Cycle Support requires documentation the NSAMs offer a way to achieve both Whether it is perfect or not our current methodology allows us to define and bound the software project divide it into discrete tasks estimate cost and schedule much more accurately measure progres throughout the project predict cost and schedule variances and react decisively when such variances occur As it happens much of the documentation produced as tools with which to manage software 4th Issue 1988 CRYPTOLOG page 30 VQR QV¥IGIAb Yli Q bY P L DOCID 4019712 production also serves another purpose it helps the poor devils in Life Cycle Support organizations who must patch and mend the result for the next ten years These people have one of the toughest jobs in the software world ask them if it costs too much to produce clean maintainable software and the documentation to go with it 86-36 the need for a disciplined approach to software development The author of this article presented several other viewpoints that I would like to address Some of his points are actually supported by the 81-2 and 81-3 philosophy but he states those points so casually and ofthandedly that one can easily overlook the implications of his Most of all I hope our Data Processing reference For example he states that the customers support methodology and discipline in critical ingredients of a successful project are simple design prototyping incremental software the aquisition As a result of front development small tightly organized teams loading software acquisition efforts the and minimal formality customer has reasonable expectations about what he is buying how much he is going to Aren't tightly organized and minimal pay for it and when it will be ready formality antithetical Still I think I get the Reverting to desultory software acquisition is point a small group of peoJ le working almost sure to result in a lot of disappointed together without having to deal with outsiders customers Without goals or requirements and without a lot of tiresome management without discipline and method we cannot types asking how things are going predict or plan events if we cannot predict or plan events we cannot estimate cost The No argument from me about incremental customer will never know what he is going to development Paragraph 2 1 2 2 on page 2-3 of get when he is going to get it or how much it NSAM 81-3 DOD-STD 1703 dated 15 April will cost until after the project is over 1987 is the definitive work on his subject as far as I am concerned The same goes for So I guess I am one of the author's axe prototyping which is actually a subset of grinders The axe I must grind remains the incremental development Sorry same whether contractors or govvies turn the but the projects I see these days ju-s ' t 'do-n-o-tstone accountability As a software yield to simple design Apparently all the acquisition manager I am responsible for simple software has already been written and producing useable maintainable software and the only projects we have left are devilishly the proper documentation on time and within complicated budget Therefore I want control or at least the illusion of control of the software The author offers a solution to the complex development process I want to know what software project If a project is large it should work is to be done how much money is going be broken into smaller projects and likens to be allotted to the task who is going to do the process to a diner eating steak Strangely the work how the work is going to be done he then leaves the subject and says nothing how soon the work is going to be finished how more about the tremendous sy tem engineering progress is going to be measured and how effort required to implement the Sirloin success is to be declared No additional work Approach The process of breaking a large will be done without a written requirement project into smaller projects is the design accompanied by a certified check more people and it is simpler said than done The process or schedule relief Furthermore I want current procedure is described in Paragraph 1 4 page 1information concerning how much work has 9 of DOD-STD-1703 which defines the design been done how much money has been spent during software development activities how much work is left to do when we are Traditionally these activities consume forty going to finish and how much it will cost in percent of the total acquisition effort the end In writing please NSAMs 81-2 and Developing software is not very much like 81-3 provide guidelines and methods for eating steak It is a process of defining the managing software projects and I need all the basic parts of a system manufacturing the help I can get Some software producers may parts according to specification and assembling find the process onerous but for the most part parts into a working system It requires the these people are professionals and understand thought planning organization cooperation discipline and a lot of hard work It requires I 4th Issue 1988 CRYPTOLOG page 31 FOR OFFICIAL USE3 ONLY I a lot of ormality 1 o0 Somebody has to document the requirements assigned to each segment of s flware the input each expects and the output it provides bec ause eventually all those segments must fit together to form a system that worke workers with the same diligence they display in applying it to contractor personnel I believe the statement is true If we allow our methodology to fall into desuetude internally we will always have great difficulty developing large software systems The author does recognize the importance and difficulty of system integration and masterfully reduces a task of imposing proportions to an elegant understatement 1 'Of course the interfaces amongst the teams must be clearly defined and the coordinator must do his job skillfully You are only too right sir The coordinator may even have to introduce a little more formality meaning anlnterface Control Oocument and a Software System Integration Plan And a Configuration Management Plan otherwise those hlghly motivated software developers in a high flush of esprit de corps will respond dynamically to new requirements and produce only mimimal adequate informal documentation while dev lopment is fluid and put the ICD and the Integration Plan down the sewer And of course the coordinator needs to know when each piece of the software is going to be ready which means a Schedule Here is Formality of the meanest sort with assigned tasks and deadlines and people's names and all like that so They knClW if you don't do your work But not to worry more than likely at least one other tightly organized team unburdened by formality iagoing to get tied up interacting with the customer and assimilating knowledge gained into the product and will blow the integration schedule to hell before you do Flexibility is built into NSA Acquisition Policy The documentation required for system development depends upon the needs and intentiOJ ls of the customer and the type and size of the project The policy itself so states the Manuals 81 2 and 81-3 contain many words about tailoring doewnentation requirements to the needs of the project If a manager takes exception to a particular ieq lirement of the policy there is usually relief aVl iilable If the leeway in our policy is inadequatel in T303 is always looking for a better method to develop software or ways improve the method weWle He has always been willing to help me with such problems In the meantime our method maybe all wrong but it is what we have to work with I suspect tha trikes close to the truth with one statement ' the Agency has suffered great difficulty in developing large software systems using today's software acquisition methodology If he refers only to software developed in-house I will agree But the great difficulty is not caused by our methodology we have great difficulty because we will not accept the discipline and the accountability that goes with any methodology I have been through some very large system acquisitions under contract and have seen methodology successfully applied Industry workers may not like it but they accept the discipline and methodology and it works Industry managers have remarked to me that the government managers cannot or will not apply 81-2 and 81-3 standards to their own I ---' 741 --- P s lreally need that study that compares the cost of developing software in-house with the cost of contracting for software development My managers are excited about the potential savings Could you send me a copy right away The author replies Iconclusion and A careful reading oft my own shows us to be in basic agreement Specifically my paper concluded by calling for a review and reevaluation of the Agency's current software acquisition methodology in his conclusion acknowledges that our method may be all wrong but it is what we have to work with which of course is my point I can only infer that he welcomes my proposal to review and revise appropriately and I heartily welcome his articulate support on this matter I I Admittedly we have a few point of disagreement but not on the need for discipline and control from my experience small teams usually possess and abundance of both commodities 4th Issue 1988 CRYPTOLOG page 32 FOR OFFI€lAL USH ONLY DOCID 4019712 I suspect our principal difference lies in what I would label the knowability assumption -- the basic premise upon which our prevailing acquisition methodology is based Today's methodology employs a waterfall model of steps identify and understand all requirements document design code test However my experiences at the Agency have led me to reject the notion that the desired system is knowable enough to document and design in cement at the outset a tremendous amount of knowledge is gained through an interactive development process With my approach a fast start can be undertaken with the knowledge gained rapidly assimilated into the process And note large teams get in the way of a fast start Yes today's methodology provides discipline and control but precious little in the way of timeliness j and I agree we would like to do better than we're doing Let the review begin I Him Then you'll have to phone us and we'll send a technician to your home He'll take the machine apart and reset the tension for that character Me But what if I want to switch back to English again Him Then it will punch holes through the papers Me Do you mean to say that every time I want to switch back and forth rve got to have a technician make a service call and reset the tension Him That's right Me That's impossible I want to switch back and forth from English to Russian sometimes even within the same sentence Him Say rve got an idea How often do you need that letter Me As often as it occurs Him Well then why don't you just use other words instead Or just leave out the letter completely I must confess dear reader that I was unable to explain to the expert over the phone that this would be comparable to having F Scott Fitgerald type elda and I went to the 00 today elda loves the ebra with his ig ag stripes To the Editor Art Salemme Ret A while ago I bought a Russian-alphabet element ftgolfball for my Selectric typewriter Every time I hit the letter iK it would print imperfectly only part of the letter would print Several phone calls to ffiM in the Washington area finally got me to the ftexpert He told me that on the Russian-alphabet element the iK is in the position that is occupied on the regular element by the colon and semicolon That position as well as the position occupied by the period and comma has reduced velocity because otherwise the English punctuation marks would pound holes through the papers The rest of the conversation went like this Me But I want to type Russian words containing the letter iK P L 86-36 Art tells us that he has since switched to a word processor for his free-lance translations Readers may recall that Art was the Editor of CRYPTOLOG in the late 70s who set the tone for this rag with his sprightly and irreverent sketches We've kept them in our portfolio for recycling and show one of them in the previous column Art was pleased to learn that CRYPTOLOG is still going strong and sends warm greetings to readers editors and contributors 4th Issue 1988 CRYPTOLOG page 33 FOR OFFICIAL USE 6NLT DOCID 4019712 1'611 SECRET UMBRA FROM THE PAST 4th Issue 1988 • CRYPTOLOG • page 34 'l'OP SKGKK'I' Y A'I8KA EO 1 4 c F L 86-36 DocrD 4019712 minier or What if Eisenhower Runs A' PP PART ART I'LL LIKE IKE The asterisk indicates a proper name If this puzzle proves of interest we'll publish more of them We also solicit contributions Here are three for this round KNIGHT AHAT TAGNKI TRANRTYNE AY EY ETRA SWOOSH KWYS PHooHY SHO PWK PHH PH SPECIAL OFFER Y I Lutwmia P1 Years ago the New York Herald Tribune published a short cryptogram along with its daily crossword puzzle There was a special ground rule -- every letter involved had to be used at least twice Over the years constructors began to torment solvers with very short cryptograms which while they conformed to the ground rules were not very easy to solve These held a strange sort of charm for some of us cryptanalysts and they may still have appeal The other rules were intelligibility of course and that the plain text consist of a sentence of three or more words There was no caption for the cryptogram nor were proper names so indicated To illustrate how brief the minicrypt may be consider A Father's day commercial I SIS IS A DAD AD Just to get to moving on this initial round a prize is offered to the irst person or team to solve the minicrypts Indicate whether the solution is an individual or a team effort Send your entries to CRYPTOWG PUZZLE PI HQS In caSe of a tie the winner s be selected by lot will Readers in the Field may sertd entries via PLATMAIL to cry p t I g @ bar 1 cO 5 Field entries will be competing only with other field entries Solution to LEFTOVERS 1 2 3 4 5 or What's the kazoo for A SEES AS I TOOT IT 6 7 CORN STAR ADAM EXIT APER REAL DIPS CRYPTOLOG page 35 ¥QK Q¥¥ICIAL YSK HfLY 4th Issue 1988 DEAN BAIL ISIS LENS FORD PIPE MEAN 3rdIssue 1988 SEND PITY IRON LIVE MERE TARS TALE RAIN RAMP PAIR RILL AIDE LINT PEAT I M P R I N T DOCID 4019712 P L NSA·CROSTIC plus No 67 86-36 1'---_---'1 G422 This puzzle has a user-friendly feature The squares are numbered so that all squares on the left side have an even middle tens digit while those on the right side have an odd middle tens digit Further all squares in the same column end with the same digit Fair warning There's a wild card in this puzzle I A Where Crockett County is 230 273 228 191 153· 255 235 lIol 102 40 257 239 272 190 212 226 200 113 137 23 65 112 51 215 115 75 53 18 27 11 111 133 113 152 67 127 208 195 170 192 99 117 10 44 87 125 106 169 189 46 '5 tC4 130 213 271 233 251 259 7 197 145 165 57 224 245 178 164 211 269 60 42 35 11 28 12 52 264 170 82 121 41 20 30 49 69 62 123 141 220 132 2522 232 t80 16' 205 79 10 151 244 '22 253 ' 157 25' US 211 199 159 156 135 175 210 201 240 221 260 160 140' 114 93 29 '4 38 59 119 128 150 161 146 tOO 6t 61 43 216 196 177 16 134 58 37 32 238 186 73 54 94 91 109 149 129 172 202 163 104 45 126 107 24 231 214 95 56 77 1 154 182 204 246 263 24 B Fixes 194 C A kind of cheer D Like tropical rain E Poisonous herb F Fletcher G Mock Roman naval battles H Respite from work I Kind of production J World War II program K Rank 61 L A canine M Anastasia for one N Said grazie or cnacu6a O Relating to a letter 25 P Religious fervor O Astonishment R • Hibiscus syriacus 36 S Inflammatory 31 T Every bride has one 247 218 90 86i 47 12 136 176 206 223 241 243 124 83 103 142 232 161 203 113 167 33 13 50 9S 89 110 148 185 221 265 274 225 250 268 266 21 101 to 76 91 138 U John Hancock used one v Necessities W Praise phrase x Specimens of xiphias gladius CRYPTOLOG page 36 FOR OFFICIAL y g Q bY 4th Issue 1988 17 DOCID 4019712 HOW TO SOLVE A DOUBLE-CROSTIC Using the Definitions fill in whatever Words you can Then copy each letter from the Words into the corresponding square of the grid below Scan the text in the grid from time to time from the recovered fragments you may be able to complete the word in context Copy the new entries from the grid into the Definitions where the fragments there might suggest a complete Word and so on working back and forth Also scan down the first positions of the Words as you recover them for additional clues IH 16p 17R 321' 36 37p 406 5211 56S 57 60 -1 72H 76 % 2°1 2£ 3' 71 278 MOVING CRYPTOLOG page 37 F6ft 6FFICIAL USI4J O lL¥ 4th Issue 1988 I DOCID Ef THIS D6etJ IENT e6NTAINS e6DEW6RB MA'f RIA' ---tOP--SECR J NOT RELEASABLE TO CONTRACTORS