Niklaus Wirth and Beyond: Safeguarding the Intellectual Heritage of Computing

On January 1^st 2024, Niklaus E. Wirth died at the age of 89. Wirth, an ACM A.M. Turing Award laureate, leaves a longstanding legacy in the field of programming languages. While other eulogy texts brought us precious insight into Wirth’s life and research, we take this moment to focus more generally on what can be done to preserve the memory and work documentation of the founders of computer science.

In the pantheon of scientific disciplines, computer science is still an infant. If we date the beginning of computer science to the work of Turing, the field has not yet reached a century of development. For instance, the second generation of scientists, the initial developers of programming languages, is still present and can bring important testimony for future generations. A precious effort can be found in the series of ACM SIGPLAN conferences on the History of Programming Languages (HOPL). These conferences occur infrequently, the most recent ones in 1993, 2007 and 2021, and often include testimonies from the language founders. Some examples are Bjarne Stroustrup’s account of the evolution of the C++ standard (along the 2011, 2014, 2017, and 2020 revisions) at HOPL 2021; a history of Erlang by Joe Armstrong (deceased in 2019), and the development of Modula-2 and Oberon by Wirth at HOPL 2007.

Preserving the memory of our field has many facets. Not only stories must be written, like those reported on HOPL, but they must also be told and recorded. Old papers must be digitized and made easily accessible, and physical records must be cataloged and preserved for the next generations. Next, we discuss some of these initiatives.

Audiovisual Memories

Several institutions have taken the initiative to interview our leading scientists and make those interviews available to the public. These first-person accounts bring us precious stories that would probably otherwise be forgotten. In the 2022 interview of Bob Metcalfe by Communications on the occasion of being granted the Turing Award, Bob tells us about the creation of the Ethernet, how the name was chosen on May 22^nd 1973—from the old physics concept of luminiferous ether, the fifth element permeating space—and the long delay, due to Xerox patenting application, before the paper was finally published in Communications in July 1976.

In addition to professional organizations like ACM and IEEE, other organizations also are playing an important role in preserving testimonies. In two videos, with the Franklin Institute and TEDx talks, 2008 Turing Award recipient Barbara Liskov talks about her path in what was the new field of computer programming, how the expansion of software in the 1970s led to increasing numbers of software failures and, finally, how the new approaches to modularity in software design helped solve the crisis.

ACM Digital Library

The ACM Digital Library is much more than a mechanism for the online dissemination of existing ACM journals, magazines, and conference proceedings. It plays a crucial role in the digital preservation of the printed media produced since the 1950s by the Association. It hosts the first edition of the Journal of ACM, published on January 1^st 1954. Its opening article talks about the genesis of the Association of Computing Machinery. There, it describes one of the initial meetings in 1947 and the definition of an initial policy “to keep the organization informal, to encourage meetings and discussions, to issue mimeographed information but no printed or more formal publications, and to maintain a mailing list of persons interested in the field …” (We leave it to the readers to investigate the meaning of mimeographed on their own.) Since then, many more formal publications have been established, and are now preserved and searchable in the Digital Library. Most computer science authors are present there. Those interested in the early works of Wirth in particular can readily find them under his ACM DL author page. 

Digitization makes printed and manuscript content readily accessible worldwide. In theory, it is also an important step towards content preservation, but only time can prove or disprove this claim. Humanity’s oldest texts are physically carved in stone and clay, later in parchments and paper. These documents are preserved in museums and archives, and specialists in document cataloging and preservation manage the process. It is wise to also preserve the documents and correspondence of our early researchers in a similar way. This is the story of the “Oxford boxes.”

The Oxford Boxes

In May 1960, Communications published the Report on the algorithmic language ALGOL 60. Two years later, the authors met in Rome to produce a revised report solving pending issues in the language, and also decided to create a more stable setting to support future revisions by creating the IFIP WG 2.1 (Working Group 2.1, on Algorithmic Languages and Calculi), which first met in August 1962. The group had many notable members and included N. Wirth, E. Dijkstra, R. Floyd, T. Hoare, and other pioneers of our field. When not in person, the discussions involved the exchange of physical paper correspondence, which former group chairs kept and passed on when stepping down. In particular, the quite rich personal archive of Willem van der Poel, chairman for the period 1962-69,  was donated to the group, amounting to several card boxes that were shipped to Jeremy Gibbons, chairman of the group and professor at the University of Oxford, around 2010.

In May 2011, the IFIP WG 2.1 met in Iceland to plan the celebration of its 50th anniversary in Rome the following year (2012), while the Grímsvötn volcano was erupting again. Interestingly, the meeting notes even show the members who could not attend due to the eruption. The matter of the Oxford Boxes was discussed and there was a proposal to archive the documentation in a state-funded historical archive in Portugal and initiate the process of digitization and cataloging. Eventually, the boxes would reach the archives in late 2017 and be subject to the archivists’ work, which was complete by June 2018, when the archival fund went live. All the minutes, letters, technical reports, and so on have been carefully preserved, living today next door to other documents dating as early as the 6^th century, when King Teodomiro restructured the Suevi kingdom in what would be Portugal one day. The group’s correspondence is particularly fascinating and can also be consulted online.

Concluding Remarks

The task of preservation extends beyond mere documentation; it encompasses the critical need to safeguard both software and hardware for the enlightenment of future generations. Institutions such as the Computer History Museum and initiatives like the Arctic Code Vault stand as beacons of these collective endeavors. The instances highlighted herein merely scratch the surface of the vast array of artifacts and avenues available for preservation. As custodians of this rich heritage, it is incumbent upon us to ensure that these invaluable memories are not relegated to the annals of obscurity.

Carlos Baquero is a professor in the Department of Informatics Engineering within the Faculty of Engineering at Portugal’s Porto University, and also is affiliated with INESC TEC. His research is focused on distributed systems and algorithms. José Nuno Oliveira is a professor in the Computer Science Department at Portugal’s University of Minho. He is a member of the IFIP WG 2.1 and also a staff member of INESC TEC. His research aims at improving scientific standards in software design through formal methods and calculational techniques.