(PDF) Before the Collapse: The Soviet Programming Cultures

7"#'$#'3 List of Abbreviations vii Acknowledgments ix Introduction: Russian Economies of Codes ! Mario Biagioli and Vincent Antonin Lépinay I. Coding Collectives "#$. Before the Collapse: Programming Cultures in the Soviet Union %& Ksenia Tatarchenko '(". From Lurker to Ninja: Creating an )' Community at Yandex *& Marina Fedorova '+,$$. For Code and Country: Civic Hackers in Contemporary Russia -. Ksenia Ermoshina II. Outward-Looking Enclaves /"0,. At the Periphery of the Empire: Recycling Japanese Cars into Vladivostok’s )' Community !!% Aleksandra Masalskaya and Zinaida Vasilyeva /)1$. Kazan Connected: “)'-ing Up” a Province !2* Alina Kontareva 3)4. Hackerspaces and Technoparks in Moscow !5. Aleksandra Simonova 3$1$#. Siberian So6ware Developers Andrey Indukaev !&* *+,-.%/ 01% BEFORE THE COLLAPSE Programming Cultures in the Soviet Union Ksenia Tatarchenko “We desperately need a programmer.”—“I’ll talk to the guys—I promise—I know Z few who are unhappy.”—“We do not need any programmer—said the hook-nosed.—Programmers— are the sought-after people, and are spoiled, but we need the unspoiled one.” —STRUGATSKY BROTHERS, Monday Begins on Saturday (1964) In the early !""#s, in a typical middle school located in an industrial neighborhood of the city of Novosibirsk, we had an “informatics” class where we learned about the principles of hardware and programming and could play computer games. $e machines were not called “computers” but “%&'” (electronic calculating machines); they had a gray and green interface and were all connected to the main computer controlled by the teacher. My high school was an experimental school endowed with additional funds, and was where I (rst experienced a modern computer class with “real” personal computers. $e classroom itself o)en stayed closed behind iron doors and barred windows—during the “wild” !""#s the robbery of school computer classrooms supplied with expensive foreign machines was common. $is classroom was closed because the informatics teacher I had met during the admission tests le) for Israel; she was greatly missed by older students, who said she was very competent. Eventually, the administration found a replacement and we began to learn how to use a text-editing application. By this time it was generally understood that Word and Excel were what informatics classes should be about. In those days, I was busy discovering French existentialism and Russian semiotics and remember cheating on the Excel assignment. Among other things, the collapse of the Soviet Union wrecked the national education system and opened the country’s frontiers: a calamity turned into an opportunity when I got a chance to study abroad. Moving from Russia to France and then to the US, my own personal trajectory impacted my research subject: the history of Soviet computing from a transnational perspective. Working on my PhD thesis and book manuscript (Tatarchenko >8!%), I uncovered in the history of computing itself explanations and connections that shed light on what I experienced in my computer classes as compared to those skills taught to my American friends. In this connection, it is worth pointing out that claims stating the Soviets had missed the “Computer Revolution” were at best misleading and that the relative rarity of personal computers in Soviet homes did not represent the absence of a computer industry or professional programmers. I learned that Western and Eastern )' histories were entangled on many levels and that the Iron Curtain simultaneously isolated and connected these two worlds. To discern the depth of the transnational connections, we need to consider multiple facets of Soviet )', including: hardware and so6ware as complex technological artifacts; the emergence of a new mathematical discipline called “computer science” in English and informatika in Russian; a set of localized practices; and machines as emblems of political legitimacy. ?e Cold War military and space race was the chief driving force behind the miniaturization of electronic components encapsulated by what is termed “Moore’s law.” ?e military origins of American networking systems and the parallel Soviet e@orts to computerize their economy into a single “big data” network are other well-known cases of contemporary )' systems that had Cold War origins. If electronics and computer network technology were the material embodiments of competition between the East and West, the capitalist and socialist versions of modernity were equally rooted in a techno-utopian imaginary that led to di@erent visions of the “Information Age.” Accordingly, the curricula for school computer education reAected two versions of an “information society”: the American one was predicated on a proBcient instrumental use of the personal computer as a basic commodity and a data-processing device; the Soviet one aimed at inculcating thought habits and programming skills in an e@ort to enable self-control and self-expression for a new kind of responsible individual. ?e collapse of the party state and the ensuing political transformations put an end to the project of creating a “socialist information society.” ?e invasion of global )' products following the opening of the Russian markets during the !&&8s dramatically altered the material landscape of computing 28 < 9 3 $ # ): ;:': ,< + $ # =" > in the New Russia. Yet a half century’s worth of Soviet experience with computing did not just disappear; instead, important continuities exist across the !&&! fault line. In this chapter, I take a synthetic approach to the history of Soviet programming in order to provide context and genealogy explaining the distinctly national dimensions of the contemporary )' landscape. First, I provide an overview of the pioneering stage of Soviet programming e@orts, as shaped by early Soviet hardware and cybernetics. Next, I focus on the commodiBcation of programming work and analyze the professionalization e@orts led by Soviet programming experts who came to claim that programming was a form of human and machine brotherhood. I then conclude with reAections on the philosophy behind the !&-* educational reform, which introduced compulsory programming classes within a context where the cloning of Western hardware became the norm. EXCLUSIVE AND ILLUSIVE: EARLY PROGRAMMERS BETWEEN ENGINEERS AND CYBERNETICIANS ?e speciBcity of Soviet computing history is inextricably linked to key features of the socialist state: its planned economy and the party’s ideological guidance. Centralized power and the planned economy did not lead to an absence of inventiveness or competition. On the contrary, from the Brst days of Soviet digital computing e@orts in the late !&28s, the development of Soviet machines was marked by a rivalry between di@erent groups of specialists. Similar to the Anglo-American debates on the “Brsts” embroiling the epithets of “digital,” “programmable,” and “fully operational,” a controversy surrounds the status of the “Brst” Soviet computer. ?e chief protagonists in the battle for public memory are E-! and E$3E machines (the former built in Moscow, the latter in the suburbs of Kiev), both important less for their performance characteristics than for their inAuence on the design and training of the workforce involved in two larger, also competing, hardware projects: Strela and I$3E. Operational by the mid-!&*8s, these two computers engendered a new kind of occupation: professional programming. Western specialists studying Soviet technology during the Cold War era were little concerned by these priority disputes. Instead, they grappled with the issue of technology transfer, asking questions such as: Were the Brst Soviet computer developments independent of one another? Were the snippets of information in the form of publicity and published overviews really all the Russians had? As the evidence relating to these questions remains partial at best, it seems sensible to turn our attention to the well-known case < / C," D , : E E ) # D 7 0 F' 0 ,$ 3 ) # ' + $ G "1 ) $ ' H # ) " # > 2! of the Soviet nuclear bomb project. According to the latest analysis by historian of science and technology Michael Gordin (>88&), even such sustained information-collection e@orts as were organized by Soviet intelligence under the secret police chief Lavrentiy Beria could not solve the major problem of information fragmentation, management, and trust. In fact, it was the public knowledge about the technical feasibility of the project and various published reports that were most responsible for the astonishing speed of Soviet nuclear e@orts. ?ese observations help elucidate the issue of transfer in the case of Soviet digital computing: feasibility was no doubt the most crucial piece of Western knowledge for early Soviet projects. If the circulation of scientiBc overviews stimulated the e@orts of Soviet engineers and mathematicians interested in calculation problems, the relationship between early Soviet and early Western computer technology was not limited to a unilateral Aow of technical information and soon became shaped by an ideological confrontation over technology’s place in society. ?e speed, size, and cost of early computers attracted the attention of the media and fascinated the Western popular imaginary, famously associating computers with “giant brains.” However, in Russia what became known as an anticybernetic campaign was founded in a series of publications that appeared in the Soviet press from the spring of !&*8 to the winter of !&**. ?e Brst publication was a witty analysis of the militarism implied in the anthropomorphic representation of the Harvard Marc III machine on the pages of Time magazine, but later articles attacked a speciBc enemy: the new American science of cybernetics (Peters >8!>). In his book Cybernetics: Or Control and Communication in the Animal and the Machine (!&2-), the American mathematician Norbert Wiener introduced the term and gave it a vague deBnition as “scientiBc study.” ?e interdisciplinary and metascientiBc ambitions involved in drawing connections between biological and man-made systems, as well as the explicit analogies between machines and human institutions articulated in Wiener’s bestseller, !e Human Use of Human Beings: Cybernetics and Society (!&*8), made cybernetics an inviting ground for ideological contestation. ?e Soviet publications labeled cybernetics a pseudoscience, a triumph of semantic idealism, and the newest form of mechanical philosophy, all the while stressing its role as a tool of Western militarism. In the wider context of rising geopolitical tensions and Soviet domestic e@orts to exert a hold over the creative and scientiBc intelligentsia (Zhdanovshchina), the anticybernetic campaign was peculiar because unlike the infamous case of Lysenkoism and the antigenetic campaign, it was not directed against any Soviet individuals 2> < 9 3 $ # ): ;:': ,< + $ # =" > or institutions. ?e most direct result of the anticybernetic publications— the withdrawal of Wiener’s books from Soviet libraries—did not mean that Soviet experts ignored this new Western development. ?e Soviet ideologues were not the only Soviet specialists to observe the military roots and implications of the new calculating technologies and cybernetic notions. Wiener’s works were held in closed libraries and several unoJcial translations circulated among experts. One reader was a young graduate of the Dzerzhinsky Artillery Academy, Anatoliy Kitov. Employed as a military representative at the special construction bureau—the 3=I>2*—and responsible for the design of the Strela computer, Kitov became an early Soviet programming expert and a proselyte of cybernetic ideas among the Soviet military and scientiBc elites. Reading and being interested in such ideologically suspicious works was not an act of rebellion but a logical step within a belief system that postulated employing Western technology in a battle against capitalism. Working on secret military projects, early Soviet experts fascinated by cybernetics and digital computing were eager to Bght against an imagined Western aggressor, but their immediate threats were much closer—the competitors for state funds o@ering alternative technical solutions. Created in !&2- under the auspices of the Soviet Academy of Sciences, the Institute of Precise Mechanics and Computational Technology (ITMiVT) had its mission to develop computational technology inscribed in its very name; tellingly, however, computation did not yet imply “electronic” or “digital” but rather “mechanical.” ?is situation changed in !&*8, when the mathematician and explosives expert Mikhail Lavrentyev took over the directorship of the new organization. In his memoirs, Lavrentyev (>888, *.–58) detailed his shrewd solutions for dealing with material scarcity by returning to patronage networks, revealing that no tactic was too dirty. Lavrentyev recounts how he accused leaders of the analogue technology projects of machinations with bureaucratic documents (a common Soviet practice at which Lavrentyev excelled) in order to force them out of the institute. Competitors out, old friends in. Returning to the capital from the Ukrainian Academy of Sciences, Lavrentyev also transferred his protégé and the designer of the E$3E computer, Sergey Lebedev, and his team of engineers from Kiev to Moscow. Lavrentyev’s reliance on his prewar Moscow networks also brought to the ITMiVT the mathematician Lazar’ Lusternik, an old companion from the famous Luzitanya, a group of mathematicians formed around Nikolay Luzin in the !&>8s, and a colleague at the Central AeroHydrodynamic Institute < / C," D , : E E ) # D 7 0 F' 0 ,$ 3 ) # ' + $ G "1 ) $ ' H # ) " # > 2% (TsAGI), the cradle of Soviet aviation during the !&%8s. A leading Bgure in early Soviet programming e@orts at the ITMiVT, Lusternik recreated the practices with which he was familiar from his time at the TsAGI: a close cooperation between mathematicians and engineers as well as scientiBc forms of interaction. In !&*8, he organized a seminar on programming where the available literature was read and discussed in a scientiBc fashion. Lusternik’s seminar led to the publication of an inAuential overview volume: !e Solution of Mathematical Tasks on the Automatic Numerical Machines, a collective work, with the subtitle Programming for High-Speed Electronic Calculating Machines (Lusternik etKal. !&*>). Over three hundred pages long, it covered all aspects of programming, from a brief introduction to digital computers and numerical systems, to detailed examples of programming techniques for a “hypothetical” three-address machine, which was in fact the I$3E computer, which was under development. Circulated under conditions of restricted access, this was the book that introduced most early Soviet programmers to their new cra6. While mathematicians in Lusternik’s group considered the problem of how to solve mathematical tasks using computers before any operational Soviet machines even existed, the Soviet science administrators began to work on the crucial question of who would operate them. In !&2-, the same year as the creation of the ITMiVT, Moscow State University (ED0) created a new chair in computational mathematics. Here, at the most prestigious school in the country, the mathematics department had few volunteers among its student body willing to abandon their aspirations in pure mathematics for the unknown perspectives of machine mathematics. Assigning students to the chair was the typical top-down solution to the problem of who but not the how of turning them into specialists of the new machine computation. ?e logic behind the curriculum—freed from many of the traditional subject areas in mathematics to make room for classes like “Algebra of Relays” and “?eory of Machines and Mechanisms”—implied that in order to program one needed to understand the mechanisms of machines. A6er struggling through the eclectic curriculum, the Brst graduates of the computational mathematics chair were to learn their jobs on the Ay. According to graduates’ memoirs, they spent their last year as interns at the ITMiVT learning to code on the I$3E computer. ?e actual experience of interaction with the new machine was immersive: its twinkling lights, sounds, and heat combined with the very size of the installation impressed its operators with a sense of almost mysterious power. But it is the human element of interaction, the shared learning and competition between peers 22 < 9 3 $ # ): ;:': ,< + $ # =" > on how best to control the machine, that had a deep impact on one’s sense of self: “Programmers were counted on Bngers, and joining this tribe Blled one with a feeling of exclusivity” (Podlovchenko >88%, %.>). ?is tribe, a small group of pioneers bound by the unique experience of working on the Brst Soviet machines, would inAuence Soviet programming for several decades to come. ?e members of the group would go on to lead so6ware projects, consult for new hardware development, and teach many generations of programmers. Unlike the designers of the Brst machines, they are not in the spotlight of public memory but act as a less visible network transmitting disciplinary mythologies. ?e important events that shaped the professional representation of this core group were closely associated with cybernetics, which had radically changed its status from a tool of imperialism to a mathematical metascience in the ser vice of communism. By the fall of !&**—when the existence of Soviet computers was Brst oJcially announced in conjunction with an international conference in Darmstadt, West Germany—the scientiBc reputation of cybernetics had already been publicly redeemed through the appearance of a seminal publication (Sobolev, Kitov, and Lyapunov !&**) titled “Osnovnyye cherty Kibernetiki” (?e main features of cybernetics) appearing in the key Soviet ideological journal, Voprosy "loso"i (Questions of philosophy). ?e text of the article was dra6ed by the young colonel Kitov and cosigned by his former teacher from the Dzerzhinsky Artillery Academy, mathematician Aleksey Lyapunov, and Sergei Sobolev, an academician and mathematical prodigy who contributed to the Soviet nuclear bomb project. ?e article focused on explaining the subject and methods of cybernetics and stressed the legitimate scientiBc nature of the discipline. According to Slava Gerovitch (>88>), the almost decade-long delay in the introduction of cybernetics to the Soviet public had an impact on its very content: unlike the servomechanisms that inspired Norbert Wiener, computers became the machines of reference in the Soviet version of cybernetics, which began to gain popularity in the late !&*8s. I argue that these crucial insights entail yet another observation: by !&**, Kitov, Lyapunov, and Sobolev not only presented computers as the machines of reference for cybernetics but also believed that programming itself was mathematical and a part of a machine’s self-regulative process and amenable to automatization. In addition to their highly visible and celebrated role in changing the status of Soviet cybernetics, the three authors played key roles in spreading this vision of programming as pioneer practitioner, mentor, and patron, respectively. Aligning programming with < / C," D , : E E ) # D 7 0 F' 0 ,$ 3 ) # ' + $ G "1 ) $ ' H # ) " # > 2* cybernetics by highlighting its mathematical foundations, their publication became the most visible national instantiation of the general international trend toward the development of so-called high-level computer languages and automatic programming systems. ?e relationship between programming and cybernetics in the !&58s was complicated by the mathematics-based cybernetic orientation on eliminating programming labor by developing system so6ware and the parallel growth of programming as a mass profession. ?is tension would eventually be resolved with the establishment of a separate disciplinary and professional identity for programmers. However, the visions stipulating the automation of programming labor still animate policy discourses such as that of Dmitry Marinichev (Biagioli and Lépinay, this volume). IN SEARCH OF IDENTITY: INTEGRATING INTERNATIONAL COMMUNITY, PRODUCING A NEW SOCIALIST COMMODITY Up to the mid-!&*8s, programming practice, coding, and notations were highly localized: the cra6 of an individual programmer was dependent on the intimate knowledge of speciBc features of a machine and on devising clever tricks to use them eJciently. Proliferation of computers and their transformation from military and scientiBc calculators into business dataprocessing machines brought forward the problem of coordinating human e@orts and introduced the diJculties of transmitting programs between machines. Across the Atlantic, “hardware” became the colloquial term for computer equipment during the !&*8s. ?e term “so6ware,” on the other hand, was initially invented in !&*& and came to denote everything that was not hardware: notation, consulting, and the new programming tools such as assembly systems, compilers, and operating systems. Running parallel to this process of the commodiBcation of programming—encapsulated in the “ware” part of so6ware—the professionalization of programming involved the creation of the Brst computer science departments in American universities and ongoing debates about their pedagogical mission. Confusion over the meaning of “so6ware” and “program” reAected the ill-deBned ontological status of a programmer’s work, astutely observed by the historian of computing ?omas Haigh (>88>, 5): “Not all ‘so6ware’ was programs, and not all programs were so6ware.” ?ese Western developments are important to Soviet programming for two reasons—comparison and contextualization. For instance, the Auidity of the better-known American case helps to clarify the paradoxical nature 25 < 9 3 $ # ): ;:': ,< + $ # =" > of the Soviet term similar to “so6ware,” which came into use in the early !&58s: “mathematical supply” (obespecheniye). Starting in the late !&*8s, the serial production of computers—such as the lamp-based E->8 and its transistor-based modiBcations—led to the same problems of coordination from human to human and from machine to machine familiar to the West. In !&58, the Soviet military and scientiBc organizations that used the E->8 machines met at a conference that resulted in the Brst user association created to facilitate the exploitation of the machines, and a crucial part of that process became the exchange of programs. Such bottom-up initiatives were soon institutionalized at the state level with a commission on mathematical supply attached to the State Committee for Science and Technology (D=#'). Similar to the vagueness of the English-language “so6ware,” the Russianlanguage “mathematical supply” did not provide for a clear distinction between product and service. However, the epithet of “mathematical” helps trace the direct relationship between Western and Soviet developments in programming—a relationship based on a shared belief in the mathematical nature of programs. In the political chronology of Cold War interactions, !&** (the year of the Geneva Summit) appears as an important turning point and a moment when the theory of peaceful coexistence was articulated and enacted. ?e theory provided a functional framework for the rise of Cold War scientiBc internationalism, best known for the Atoms for Peace meetings and the spectacular launch of Sputnik during the International Geophysical Year. ?e early Western-Soviet contacts in computing Bt the same larger scheme: spanning activity from participation at international professional conferences and workshops held by the new International Federation for Information Processing ()/)L) to bilateral exchanges. While similar to other strategic technologies, information collection was one of the driving forces behind the exchanges in computing; the dynamic of the Brst Soviet-American exchanges of computer specialists demonstrates that intellectual coordination and preoccupations with the nature of human-machine interactions were the key elements. Soviet integration into the international Algol project—a result of visitsK by American scientists in the late !&*8s—provides the best example ofKmutual e@orts triggered by shared beliefs in the power of mathematics for transcending all barriers. By !&58, the project acquired a large set of European participants and an anti-)IE ideology. Unlike the already popular Fortran devised to Bt the characteristics of )IE machines, Algol was thought to be a truly universal, machine-independent, and mathematically sound < / C," D , : E E ) # D 7 0 F' 0 ,$ 3 ) # ' + $ G "1 ) $ ' H # ) " # > 2. language, empowered not by corporate capital but by scientiBc internationalism embodied in collective work on its deBnition and standardization. Considered a practical failure in the US—on the basis of the number of compilers and not its larger inAuence—and a moderate success in Europe, Algol became the most widely used computer language in the Soviet Union and Eastern Bloc countries during the !&58s. ?e puzzle behind Algol’s popularity in the East still involves many unknown elements, but the core element on the level of ideas was the conceptualization of the program as a mathematical object, an approach familiar in Soviet circles thanks to the Soviet version of mathematical cybernetics. By the late !&58s, there were at least half a dozen Algol compilers for the E->8 computers in the Soviet Union, along with others for the newer and more powerful Soviet machines, such as the I$3E-5. Competing research groups in Moscow, Leningrad, and Novosibirsk made important e@orts at distributing and publicizing their work. ?e particularly rich published accounts and documentary sources produced by the Novosibirsk group—which beneBted from the showcase status of the scientiBc center Akademgorodok, located in Novosibirsk—demonstrate the changing conditions of programmers’ work and the emergence of new organizational challenges. ?ese challenges surmounted by Akademgorodok computer pioneers would Bnd an echo in the post-Soviet Siberian initiatives to capitalize upon the region’s reputation as an )' hub dating from the !&58s (Indukaev, this volume). To produce large-scale so6ware systems such as compilers it was not enough to add together individual skills and a labor force. Published by the project leader Andrey Ershov in the local newspaper in JanuaryK!&5*, “?e Alpha-Birth” recounted the challenges of producing an automatic programming system competitive in quality to manual programming. ?e unexpected technical troubles, the missed deadlines, and the doubling of the code volume from the expected twenty thousand to forty thousand lines, were all typical problems that demanded solutions bridging the technical and the social. ?e coordination of e@ort was paramount for the ultimate success of the Novosibirsk group and is clearly still a major issue for today’s companies, such as in the case of Yandex’s emphasis on a shared set of code-writing skills (Fedorova on Yandex, this volume). In Akademgorodok, collective coding became a personally fulBlling experience. “We will keep the gained experience, deep satisfaction with the completed work and the priceless camaraderie,” wrote Ershov, “that was born and matured during the years of work on Alpha-system.” In other words, the “birth” of a compiler was predicated on the creation of a collective with a family-like cohesion. 2- < 9 3 $ # ): ;:': ,< + $ # =" > ?e work of Ershov’s group is of particular importance not only because of the technical features of the system, but because the creation of a compiling system and the formation of a programming collective came together with the coming of age of a new leader in the Beld of programming. Riding on the success of his Alpha system, Ershov claimed a professional and disciplinary identity separate from Soviet cybernetics. A talented and ambitious student who participated in Lyapunov’s famous class on the principles of programming at ED0, and in !&*. a PhD student and group manager at the new Akademgorodok Computer Center, by the late !&58s Ershov had grown into a pundit and spokesperson for programming on both the national and international level (Kraineva and Cheremnykh >8!!). Ershov was named head of the state commission formed by the D=#' to monitor the development of mathematical supply in the Soviet Union and became the main author of its report submitted in summer !&5- (Ershov !&5-). A snapshot of the Soviet programming landscape, the report estimated the number of Soviet system programmers at about one thousand, almost equally distributed between the academic computing centers, the key hardware production facilities, and a series of military-industrial organizations. But the report’s most interesting aspect was its language and the policy recommendations that squarely placed the Soviet programming community within the international milieu. In it, Ershov argued for rapid growth and the professionalization of system programming in order to achieve the Soviet computerization goals: the creation of “system programmer” as an established engineering profession; the separation of service and research functions in academic computer centers; and an orientation on borrowing so6ware libraries for the new Soviet family of computers, later known as the United System ($3). ?e date of the report, JulyK !&5-, is crucial for understanding the full meaning of its content. On the one hand, it captures a moment in time when Soviet experts were still debating the costs and beneBts of orientating the new Soviet series on the )IE architecture and the best mechanism for doing so. On the other, the language of and arguments within the report reAected and preBgured the largest concern of the international community at the time, which became encapsulated in the notion of “so6ware crisis.” ?e #:'"-sponsored conference on so6ware engineering held in Garmish (West Germany) on OctoberK.–!!, !&5-, became the epicenter for reAection on so6ware risk and reliability as well as a forum for a very pragmatic preoccupation with the costs of so6ware production, the status of programmers’ labor, and the solution to the “so6ware crisis” associated with the < / C," D , : E E ) # D 7 0 F' 0 ,$ 3 ) # ' + $ G "1 ) $ ' H # ) " # > 2& creation of “so6ware engineering.” To sum up, Ershov’s deep integration into the international community, traceable to the Algol group’s formal and informal networks, shaped his solutions to peculiar Soviet problems. At the same time, as a Soviet professional, Ershov took on as his responsibility the state’s interests in international prestige, the computerization of research and production, and the education of new professionals. Recognizing that to become an accepted profession programming needed its own mythology, he used the available cultural resources to articulate his vision of an ideal professional for both domestic and foreign audiences. Invited to deliver a prestigious keynote speech at the main American professional conference for computer experts in !&.>, Ershov described a universal ideal programmer by creatively combining Soviet rhetorical structure with Western references: “In his work, the programmer is challenged to combine, with the ability of a Brst class mathematician to deal in logical abstraction, a more practical, a more Edisonian talent, enabling him to build useful engines out of zeros and ones, alone [sic]” (Ershov !&.>, *8>). To emphasize the transcendent quality of the new profession, Ershov did not shy away from borrowing biblical language and imagery, where a programmer “feels himself to be the father-creator of the program, the son-brother of the machine on which it runs, and the carrier of the spirit which infuses life into the program/ machine combination” (*82). ?e highest aspiration of such an ideal practitioner according to Ershov was to spread the gospel of programming to all humanity in a recognizable logic of both Christian and Marxist worldviews. “Is it not however the highest aesthetic ideal of our profession,” concluded the Soviet expert, “to make the art of programming public property, and thereby to submerge our exclusiveness within a mature mankind?” (*8*). ?e reception of the speech, which immediately met with enthusiasm from its Western audience, and its present-day relevance visible in recent citation patterns point to the ongoing elaboration of professional identity in the Beld of programming. All grandiloquence aside, Ershov nonetheless reAected upon the mundane and concrete aspects of a programmer’s labor as shared across the Atlantic and behind the Iron Curtain: their interaction with the machine, between themselves, and with society at large as complicated by the pace of hardware evolution, the status of scientiBc knowledge, and institutional struggles. At the same time, such shared concerns did not preclude Ershov’s and his Western colleagues’ awareness of the di@erent political and economic structures that were in place. ?ey considered it their duty to serve the needs of their respective countries. For Ershov, this duty *8 < 9 3 $ # ): ;:': ,< + $ # =" > found its utmost expression in his involvement in the !&-* reform introducing computer education into the Soviet school system. CONTESTED VISIONS OF LATE SOVIET COMPUTERIZATION While it is beyond the scope of this overview to provide a full account of Soviet programming developments during the !&.8s, !&-8s, and early !&&8s, the following question is nevertheless unavoidable: Is failure the only way to describe late Soviet computer developments? By the early !&.8s, there was not only a Soviet computer industry but also a new Soviet profession: the programmer. Although not nearly as numerous as their American colleagues, programmers had specialized journals, professional meetings, and even a public relations coup thanks to the national and international victories of the chess program, Kaissa. Yet when the Cold War ended with the Soviet collapse, Western specialists were quick to observe that the Soviet Union entered the !&&8s without computers. Although such observations chieAy referred to the absence of personal computers in Soviet households, they implied a more general Soviet failure to experience what is generally coined the “Computer Revolution” and to enter the “Information Age.” ?e !&5- state decision to make the new Soviet computer series compatible with )IE %58 architecture was considered a major turning point away from original Soviet research and development to systematic illicit borrowing. In fact, the practice of reverse engineering shaped the materiality of late Soviet computing from $3 large-frame computers to minicomputers to Apple and L<s and even to pocket programmable calculators. While there is no single systematic study of the late Soviet computer industry, the common Western perception holds that the planned economy could not handle the sophistication of microelectronics production, and that the party’s monopoly on information could not allow for the di@usion of personal computers, a technology that allegedly enabled freedom of expression. To reconstruct the major changes in Soviet programming in this material environment deBned by imitation we need to break with the circularity of such explanations and account for the ongoing contestation of Soviet computerization schemas from within. Several observations are relevant for understanding the professional challenges and aspirations of Soviet programmers during the late Soviet period. On the one hand, changes due to the importation of Western technologies were mitigated by important institutional and social continuities. < / C," D , : E E ) # D 7 0 F' 0 ,$ 3 ) # ' + $ G "1 ) $ ' H # ) " # > *! For instance, the pioneer computer organization focusing on military computing, the 3=I->2*, was integrated into the ScientiBc Research Center for Electronic Computational Technology (NITsEVT), the lead organization supervising the developments of $3 computers in the Soviet Union. By the same logic, the NITsEVT so6ware department relied on established academic experts in programming to oversee its projects. On the other hand, major changes appeared due to large-scale di@usion of hardware: by the late !&.8s about .8K percent of all computers in the country were $3 machines. Although plagued by many delays and reliability issues, the mass production of $3 machines (estimated at about sixteen thousand units for the entire period of production) led to the spread of computers across the country, entailing the demand for many more exploitation engineers, system programmers, and operators. New economies of scale led to major changes in the Soviet computing landscape. Yet social continuities also translated these changes into particular hybrid practices on the ground; hybridization not only explains the operation of post-Soviet )' but in a more uncanny away appears to foreshadow the unpredictable outcomes stemming from the Putinera initiatives to transplant Western forms of innovation onto the Russian soil (Simonova, this volume). A sketch of the activities of Ershov’s department at the Akademgorodok Computer Center elucidates the issue. ?e work on the original computer research and development never completely stopped. In the early !&.8s, Ershov was able to create a production spin-o@ of his department, a so6ware construction bureau institutionalized as the Novosibirsk branch of the ITMiVT and charged with developing so6ware for the supercomputer that continued the I$3E-5 line, called Elbrus. At the same time, Ershov’s group was able to obtain funds under the $3 umbrella to work on an extremely ambitious experimental project that was never delivered to its customer—a multilingual translating system called I$': for Algol 5-, Simula, and LF/!. Understanding how various actors adapted to the practice of borrowing on the local level helps us to appreciate the paradoxical characteristics of more general patterns that appear on a national scale: for example, the widespread custom of modifying hardware and programming in machine codes at the institutional level was disrupted, but tinkering with electronics became a common hobbyist practice with the availability of discarded pieces and the mass production of a@ordable programmable calculators; experts trained in the tradition of Soviet mathematical cybernetics turned into traveling lecturers on the $3 operating system, while in the !&.8s LF/! took the place of Algol 58 as the most popular language in civilian computer centers; *> < 9 3 $ # ): ;:': ,< + $ # =" > Western home computers were sold on the black market and at the closed foreign currency shops, and the network of a6erschool education centers carrying programming classes for children led to a wide popularity of Basic by the mid-!&-8s. ?is usage of Western programming languages and so6ware packages and the adoption of corresponding programming practices had its advantages, and its costs. As became apparent to the early critics of the Soviet policy of giving priority to technology transfer over investment in the indigenous projects, the o6-cited argument about the beneBts of copying hardware architecture to obtain access to millions of lines of so6ware did not account for two crucial factors: First, there was the inherent diJculty of understanding foreign, o6en illegally obtained, programs. Without appropriate documentation and human interaction, the economy of labor was questionable. Second, there was the concern for nurturing a critical mass of qualiBed specialists working on cutting-edge developments in systems’ so6ware. ?e expertise of system programmers was not solely evaluated by a degree or diploma but represented a personal proBle built over years of experience and corresponded to a particular mindset. Borrowing so6ware disrupted such dynamics of professional growth. To sum up, the orientation on copying hardware to save on so6ware development led to a radical growth in the number of programmers, but it simultaneously aggravated questions surrounding the status of their work and training. ?ese implications are best demonstrated following Ershov’s and his group’s trajectory. While Ershov ascended in the Soviet scientiBc hierarchies, and as his research on the mathematical nature of compilation became recognized on the national level with the prestigious Krylov Prize, his vision of the professionalization of programming soared. Masses of programmers were trained on the job by the computer industry and dispersed throughout the computer centers of the production ministries. While there were many specialized national and international conferences on the topics of computer science and theoretical programming held in the Soviet Union in the !&.8s and !&-8s, no meetings followed the two Union-wide professional conferences for programmers held in Kiev (!&5-) and Novosibirsk (!&.8). Despite the requests for more specialists, the universities, including the Novosibirsk State University where Ershov taught, resisted classes in computing. In the context where Soviet scientiBc hierarchies prioritized the authority of pure science, applications, including programming, became niches for what was considered a less reliable labor force—women—and safe havens for “suspicious” social elements: Jews. As a result, Soviet computer centers < / C," D , : E E ) # D 7 0 F' 0 ,$ 3 ) # ' + $ G "1 ) $ ' H # ) " # > *% were prone to high rates of turnover; there was widespread headhunting for system programmers and women o6en did not return or had to be retrained a6er taking maternity leave. In addition, programming was one of the areas directly impacted by the Soviet decision to allow Jewish emigration in the early !&.8s (Fedorova on Israel, this volume; West, this volume). ?e challenges of the Soviet professionalization of programmers’ labor described above provide crucial context for understanding the trajectory of the late Soviet project to universalize programming skills, crystallized in the !&-! slogan “Programming, the second literacy” (Ershov !&-!). During the !&.8s, Ershov reaJrmed his belief that reliable so6ware is a direct reAection of the personal virtues of the programmer: logic, patience, and discipline. Addressing the young readers of the specialized mathematics journal for youth, Kvant, Ershov claimed that “good programmers are people of a special kind” and invited schoolchildren to learn programming to become members of tomorrow’s computer centers (Ershov and Zvenigorodskiy !&.&). In the following years, Ershov articulated a fully developed pedagogical agenda of computer education whereby the human qualities shared by “good programmers” could be nurtured early on, from childhood, by developing a set of mental habits: “algorithmic thinking.” Such an agenda was backed by results of the collective work of his computer education group, which developed experimental pedagogical so6ware and ran successful summer camps and distance-learning experiments throughout the !&.8s and !&-8s. Although the educational activities of the Akademgorodok group were not unique per se, they led to very distinct results: Ershov’s status as a pundit contributed to the international di@usion of his philosophy of computer education, and his authority within the Soviet hierarchies helped him to integrate Moscow’s political networks. Ershov’s alliance with the ambitious physicist Yevgeniy Velikhov, known for his close connection with the rising party leader, Mikhail Gorbachev, underlie the transformation of his vision into state policy. Among the Brst state orders signed by Gorbachev when he entered oJce in winter !&-* was the computerization of education, including the introduction of compulsory classes in “?e Basics of Informatics.” Although grounded in the ideas and experiments of the !&.8s, within the changing political context of the mid-!&-8s the reform and its pedagogical agenda became rearticulated in coordination with perestroika and its ideals of democratization. At the same time, the implementation of the reform was anything but a simple a@air: teachers were to be trained, textbooks printed and delivered, computers and so6ware produced and distributed. Characterized by numer*2 < 9 3 $ # ): ;:': ,< + $ # =" > ous setbacks at every level, the reform became associated with the weaknesses of the Soviet state. ?e delays with the delivery of computers to schools led to descriptions of the reform as a late Soviet absurdity: learning to program without computers was akin to learning to swim without water. ?e relative ease of obtaining the Japanese computers in Vladivostok, as documented by Aleksandra Masalskaya and Zinaida Vasilyeva in this volume, was more of an exception to the general scarcity. But the main attack driving numerous discussions in the Soviet press concerned the meaning of “computer literacy” itself. According to many Soviet and international specialists, the proBcient use of applications such as word processing and spreadsheet calculations, not programming, should be the goal of computer literacy and should deBne computer education in school curricula. For the proponents of the reform’s philosophy of universal programming skills there was little contradiction between its goals and the instrumental approach: learning to program does not prevent mastering typing or editing. Agreeing that every child would not grow up to become a professional programmer, Ershov (!&--) insisted that interacting with the computer should be an empowering human experience: “?e discipline of action is equally necessary for a human, like the discipline of mind and the discipline of speech. Exercising in the control over a computer, a human being elaborates an ability to control himself.” Drawing on the cultural baggage of the revolutionary project of making a new man and the early Soviet campaign for eradicating illiteracy, the slogan of “Programming, the second literacy” allowed proponents of the reform to envisage a socialist “information society” where programming fostered personal virtues and social harmony. Because he died in !&--, Ershov did not witness the fall of the Berlin Wall and the dissolution of the Soviet Union. Even if many computers were eventually delivered to schools and the class on “?e Basics of Informatics” remained on the curriculum, the reform’s goal of developing special thinking habits—algorithmic thinking via programming—was discredited. While the limited and short-lived !&-* reform does not lend itself to causal claims about the numbers or quality of Russian )' specialists abroad, beyond the speciBc cases of several model schools in capitals and Novosibirsk, the intense debates it provoked demonstrate the high level of interest in )' technologies in late Soviet society and help situate the radical transformative visions among its post-Soviet heirs from Kazan to Estonia (Kontareva, this volume; Savchenko, this volume). Furthermore, the philosophy and pedagogical experiences that guided the reform did not disappear altogether; they can still be found at Akademgorodok—with its ongoing tradition of annual < / C," D , : E E ) # D 7 0 F' 0 ,$ 3 ) # ' + $ G "1 ) $ ' H # ) " # > ** summer camps for young programmers—and in the present-day work of prominent computer education specialists who started their careers within the reform framework of the !&-8s. CONCLUSIONS From their place in secret military installations to their use as tools of political legitimacy, computers were at the very core of post–World War II Soviet history. Soviet programming did not appear as an a6erthought but was part and parcel of Soviet computing e@orts. Its speciBc features were a product of national structures as well as interactions with the West. Following the radical reappraisal of cybernetics in the Soviet Union in !&**, the early public discourse and professional identiBcation of programmers took place under the umbrella of a new metascience. ?e technological challenges of so6ware portability and international coordination of expertise shaped the deBnition of a separate professional identity in the !&58s. ?e rationale behind the much-contested orientation of the late Soviet computer industry on reproducing Western hardware architecture ranged from scientists’ interest in integrating the Western community, to problems of coordination between the military and civil sectors, to the state’s goals of mass computerization. However, the numerical growth of Soviet programmers did not lead to a stronger corporate identity as the distance between the aspirations and realities of academic professionals and typical computer centers mounted. ?e Soviet expert in automated programming, Andrey Ershov, became the main spokesperson for the profession at home and abroad. According to his vision, the essence of programming was the realization of human intelligence predicated on an intimate connection: the brotherhood of man and machine. Fusing the experiments of American and Soviet educators in the !&.8s and Soviet revolutionary aspirations to create a “new man,” Ershov developed an original philosophy of computer education that guided the Soviet educational reforms of the !&-8s and the attempt to create the socialist “information society.” According to the slogan “Programming, the second literacy,” programming should become a universal skill and a guarantor of social and political cohesion. Although the reform’s goal of human engineering was contested and discredited, the end of the Soviet state did not entirely bring to an end the technocratic visions in post-Soviet societies. While accounting for the radical novelty and the rapid evolution of digital computer technology associated with the notion of computer generations, *5 < 9 3 $ # ): ;:': ,< + $ # =" > the concepts of “Computer Revolution” and more recently “big data” do not answer all our questions; it is equally important to trace the continuity of practices predicated on the much slower change of human generations. ?e Soviet programming cultures and visions of the socialist “Information Age” reveal overlapping social networks, values, and rhetoric stretching from the early Soviet to the post-Soviet periods. From Lenin’s famous declarations that without literacy there is no politics and that communism equals Soviet power and electriBcation, to Gorbachev’s perestroika and education computerization, the communist project was predicated on the re-engineering of nature, humans, and machines. However, unlike pilots and cosmonauts— the iconic Soviet heroes embodying the power of Aight gained by a manmachine interaction—Soviet programmers were and remain less visible and members of a potentially disposable occupation caught between mathematical designs to automatize programming and the realities of the laborious and error-prone practice of reading and writing code. REFERENCES Ershov, Andrei. !&5*. “ ‘Al’fa-rozhdenie’ ili kak sozdavalas’ sistema avtomaticheskogo programmirovaniia.” Za nauku v Sibiri, JanuaryK!-. Ershov, Andrei. !&5-. “Ob urovne matematicheskogo obespecheniia elektronnykh vychislitel’nykh mashin.” July. Ershov Archive, f. >&-/l. !–2*. Ershov, Andrei. !&.>. “Aesthetic and Human Factors in Programming.” Comm. #$% !*, no.K.: *8!–*. Ershov, Andrei. !&-!. “Programming, the Second Literacy.” Multiprocessors and Multiprogramming -, no.K!: !–&. Ershov, Andrei. !&--. “Kom’iuterizatsiia shkoly i matematicheskoe obrazovanie.” International Congress on Mathematical Education, July. Ershov Archive, f.K%%2/l. >*5. Ershov, Andrei, and Gennadii Zvenigorodskii. !&.&. “Zachem nado umet’ programmirovat’?” Kvant &: 2.–*!. Gerovitch, Slava. >88>. From Newspeak to Cyberspeak: A History of Soviet Cybernetics. Cambridge, MA: E)' Press. Gordin, Michael. >88&. Red Cloud at Dawn: Truman, Stalin, and the End of the Atomic Monopoly. New York: Farrar, Straus and Giroux. Haigh, ?omas. >88>. “So6ware in the !&58s as Concept, Service, and Product.” &''' Annals of the History of Computing >2, no.K!: *–!%. Kraineva, Irina, and Nataliia Cheremnykh. >8!!. Put’ programmista. Novosibirsk: Nonparel.’ Lavrentyev, Mikhail. >888. “Opyty zhizni. *8 let v nauke.” In Vek Lavrentieva, edited by Nataliia Pritvits, Valerii Ermikov, and Zamira Ibragimova, %8–.8. Novosibirsk: Izdatel’stvo 3",:#. < / C," D , : E E ) # D 7 0 F' 0 ,$ 3 ) # ' + $ G "1 ) $ ' H # ) " # > *. Lusternik, Lazar,’ Aleksandr Abramov, Viktor Shestakov, and Mikhail Shura-Bura. !&*>. Reshenie matematicheskikh zadach na avtomaticheskikh tsifrovykh mashinakh. Moscow: Izdatel’stvo Akademii Nauk 333,. Peters, Benjamin. >8!>. “Normalizing Soviet Cybernetics.” Information and Culture 2., no.K>: !2*–.*. Podlovchenko, Rimma. >88%. “Vospominaniia o pore uchenichestva u Alekseia Andreevicha Liapunova.” In Istoriia Informatiki v Rossii: uchenye i ikh shkoly, edited by Viktor Zakharov, Rimma Podlovchenko, and Iakov Fet, %.8–.*. Moscow: Nauka. Sobolev, Sergei, Anatolii Kitov, and Aleksei Liapunov.K!&**. “Osnovnye cherty Kibernetiki.” Voprosy "loso"i 2: !%5–2-. Tatarchenko, Ksenia. >8!%. “A House with the Window to the West: ?e Akademgorodok Computer Center (!&*-–!&&%).” PhD diss., Princeton University, NJ. Wiener, Norbert. !&2-. Cybernetics: Or Control and Communication in the Animal and the Machine. Cambridge, MA: E)' Press. Wiener, Norbert. !&*8. !e Human Use of Human Beings: Cybernetics and Society. Boston: Houghton MiMin. *- < 9 3 $ # ): ;:': ,< + $ # =" >