Polish Contributions to Computing

Resources:

Postwar Developments in Polish computing began at the end of the 1940s. There are too many people to mention here, so we focus only on two of them who were among the first inventors and published the account of their accomplishments in historical overview articles in the IEEE Annals of the History of Computing [1,2]: Romuald Marczynski and Leon Lukaszewicz (no relation to Jan Lukasiewicz, the inventor of the Polish notation).
Romuald Marczynski was born on January 7, 1922 in Skarzysko-Kamienna, Poland, and died on January 1, 2000, in Washington, DC. His interest in computers was sparked in 1946, when he read an article about the ENIAC machine in a Polish popular science magazine, Problemy. He graduated in 1948 from the Electrical Engineering Department at Warsaw University of Technology and began work there as an assistant. At the end of 1940s he joined the Group of the Mathematical Apparatus (GAM) at the Mathematical Institute of the Polish Academy of Sciences (PAS), and continued working at PAS for the rest of his life. He was the primary author of the EMAL machine designs [1,3].
Leon Lukaszewicz was born on November 20, 1923 and lives in Warsaw. He graduated in Radio Engineering in 1948 from the University of Technology in Gdansk, and started work at the Institute of Telecommunications in Warsaw, simultaneously studying mathematics at the Warsaw University. He joined the GAM at the end of 1940s and later founded the Institute of Mathematical Machines, where he was a director between 1956-1966. He was a chief designer of the first fully operational Polish computer XYZ [2,3] and the main inventor of the programming language and system SAKO [4,5]. He was the Vice President of the International Federation of Information Processing (IFIP) between 1964-1968 and has been the member of the Polish Academy of Sciences since 1976.

Major Contributions: In hardware, development of the EMAL design (Marczynski et al., 1953-1955) and subsequently the XYZ machine (Lukaszewicz et al., 1956-1959); in software – the invention of the SAKO programming language and system (Lukaszewicz et al., 1960).

Basic Questions
What were the computers that originated in Poland? The first EMAL (Electronic Machine Automatically Computing), called EMAL-1, was a one-address computer based on a vacuum-tube logic and mercury memory, with 512 40-bit words and fixed-point, sign-plus, absolute-value arithmetic [1]. It was never fully operational and was dismantled before making any computations. XYZ was one-address machine implemented in diode logic and dynamic vacuum-tube flip-flops, with 36-bit words and sign-plus absolute-value arithmetic. It evolved into ZAM-2 and later into ZAM-41 computer, whose design has been fully described in English by its principal designer [7]. It is also worth mentioning that beyond developments of groups led by Marczynski and Lukaszewicz, Z. Pawlak designed a computer named BINEG, built between 1957-1959 in electron tubes technology, which worked in a negative binary notation, with 512 36-bit words, and was used mostly for teaching purposes at the Warsaw University of Technology.

Have these computers been ever produced on an industrial scale? This part of computing developments in Poland is less known. According to Madey and Syslo [6], the first computer maker in Poland, ELWRO, was created in 1959, in Wroclaw. Their first product was Odra 1001 machine, released in 1961 and based on the prototype of first Polish transistor machine, S1, developed at the Institute of Mathematical Machines in Warsaw (a successor of the GAM group, and later ZAM – the establishment for mathematical apparatus). Between 1962-64, ELWRO produced twenty five UMC-1 machines, which were based on the BINEG machine [1]. The year 1967 seems to be crucial for the computer industry in Poland, since during that year the production of new Odra 1204 model has started and an agreement has been signed with the British computer maker ICL, to start production of Polish machines compatible with the ICL 1904 model [6]. A more detailed account of industrial developments in Polish computing is given in [6].

Significance
It is not easy to judge how important were the post-war developments in Polish computing, but we can quote from the developers themselves. "EMAL-1 was based on the British EDSAC machine, as far as the general machine organization is concerned" [3], but its construction was genuinely Polish. "The architecture of the XYZ and the list of instructions were based on the IBM 701 computer" [2], but the implementation was originally Polish and relied on a serial organization rather than parallel. One truly original Polish project was the development of a BINEG machine, based on a conceptual design of Zdzislaw Pawlak and implemented in a negative binary notation [1, 10]. Regarding software developments, the programming system "SAKO has eliminated almost entirely the need to use machine-code when programming for XYZ and ZAM-2" – (XYZ’s successor).

Findings
The detailed history of the post-war Polish inventions has been described in English in two articles published by the Annals [1,2], and in multiple articles and interviews published in Polish [6,8,9], but they were all published in the 1980s and 1990s. Our recently found source of general information predates all these sources by at least twenty years [3]. It is a small book in Polish written by a popularizer of computing, Adam Empacher, and is as comprehensive as the other sources, but has the advantage that it is contemporary to the events being described, since it was published in 1960, when all the developments were actually taking place. Interestingly, Empacher mentions one machine, which predates all the other Polish developments. It was called GAM-1 and was built by Z. Pawlak. It was a little funny, because it used only 2-bit numbers, but could do all four arithmetic operations and was fully functional, although could be used only for demonstration and teaching purposes.

References [show]

[1] R.W. Marczynski, The First Seven Year of Polish Digital Computers, IEEE Annals of the Histroy of Computing, Vol. 2, No. 1, pp. 37-48, January 1980
[2] L. Lukaszewicz, On the Beginnings of Computer Development in Poland, IEEE Annals of the History of Computing, Vol. 12, No. 2, pp. 103-107, 1990
[3] A. Empacher, Maszyny licza same [in Polish: Machines Compute by Themselves], Wiedza Powszechna, Warsaw, 1960
[4] L. Lukaszewicz, SAKO - An Automatic Coding System, Annual Review in Automatic Programming, Vol. 2, pp. 161-176, 1961
[5] A.W. Mazurkiewicz, Arithmetic Formulae and the Use of Subroutines in SAKO, Annual Review in Automatic Programming, Vol. 2, pp. 177-195, 1961
[6] J. Madey, M. Syslo, Poczatki informatyki w Polsce (in Polish: The Beginnings of Computing in Poland), Informatyka, nr 9-10, 2000.URL: http://www.gazeta-it.pl/rozmaitosci/poczatki_inf.html and http://www.ii.uni.wroc.pl/~rekrut1/rekrutacja_ii/misc/HistInf.pdf
[7] L. Lukaszewicz, Outline of the Logical Design of the ZAM-41 Machine, IEEE Transactions on [Electronic] Computers, Vol. 12, pp. 609-612, 1963
[8] M. Czarkowski, Rozmowa z prof. Romualdem W. Marczynskim, twórca pierwszych, polskich komputerów (in Polish: An Interview with Prof. Romulad W. Marczynski, Inventor of the First Polish Computers), Bajtek, nr 3(39), p. 3, 1989. URL: http://www.aresluna.org/attached/computerhistory/articles/odemalado
[9] H. Krawczyk, Poczatki informatyki w Polsce - rozmowa z prof. Leonem Lukaszewiczem, konstruktorem komputera XYZ (in Polish: The Beginnings of Computing in Poland – An Interview with Prof. Leon Lukaszewicz, a Designer of Computer XYZ), Pismo PG [Newsletter of the Polytechnic of Gdansk], February 1999. URL: http://www.pg.gda.pl/pismo/99_02/s28.shtml
[10] Z. Pawlak, The Organization of a Digital Computer Based on the "-2" System, Bulletin de l’Academie Polonaise de Sciences, Ser. Sci. Tech., Vol. VIII, pp. 252-258, 1960