Open Access in Computer science

Computer science, a still young discipline with roots in mathematics and electrical engineering, has always been very open to sharing its research results. Success stories include open protocol standards (e.g. for the internet and for e-mails); free and open source operating systems (e.g. Linux); the free and open source typesetting system LaTeX; open encryption tools (e.g. PGP); and free and open source software in general.

Since the mid-1990s, conferences, symposia, and even workshops, have established themselves as venues for publishing and disseminating research results. They now hold a dominant position vis-à-vis the classical scholarly journals. This is also reflected in the portfolios of scholarly publishers - for example in Springer's Lecture Notes in Computer Science (LNCS) series - and of the professional associations, such as ACM and IEEE, who also act as publishers. However, this strong focus on conference publications continuously gives rise to perception problems in the scientific landscape, for example when contributions to conference proceedings are ignored in academic evaluation processes because they are not covered by the journal citation indices. Although a debate on the relationship between conference papers and journals publications has been conducted within the discipline for some years now (Vardi, 2009), (Vardi, 2010), (Mehlhorn, Vardi, & Herbstritt, 2013), and there are some innovative concepts for merging the two media - for example that practised by the VLDB Endowment (Jagadish, 2008)) - it must be assumed that the focus will remain on conference proceedings in the coming years.

This balancing act between conference papers on the one hand, and journal articles on the other, is also reflected in the current offerings for making computer science research results available in open access (OA). In some subfields (e.g. theoretical computer science and cryptology), there is a distinctive preprint culture. For workshops, various publication venues have established themselves, which differ in terms of their thematic focus and/or the exactingness of their standards. For conferences, a gold-OA series - Leibniz International Proceedings in Informatics (LIPIcs) - has been in existence for some years now. It focuses on the major conferences in the subfields of computer science, and very selectively takes into account the reputation and quality of the conferences whose proceedings it publishes. In the case of journals, progress in the direction of OA, although somewhat slower, is perceptible nonetheless. However, many OA journals in computer science operate without a traditional publisher in the background and are organised from within the community.

To date, the national professional societies in Germany have taken quite a conservative stance on OA. Although the German Informatics Society (GI) signed the Berlin Declaration on Open Access to Knowledge in the Sciences and Humanities in 2007, and makes its Lecture Notes in Informatics (LNI) series freely accessible online, the scholarly journals of the thematic subfields are still strongly attached to the classical subscription model. Via its Author-Izer service, the American Association for Computing Machinery (ACM) allows its authors to post links on either their homepage or their institutional repository to enable visitors to download the definitive version of their articles free of charge from the ACM Digital Library.(ACM Author-izer Service, 2011). However, from a licensing point of view, the conditions for reuse are not always clear, with the result that the goals of OA are achieved only in part. ACM authors may also select an author-pays option to provide OA to their articles. The IEEE Computer Society (IEEE Computer Society, 2015) is the section of the professional association IEEE that is responsible for computer science. It acts as a publisher for scholarly journals and conference proceedings. At present, there are no explicit possibilities of providing OA to articles published in IEEE conference proceedings. However, in the case of subscription-based journals, it is possible to provide OA to a work by paying an article processing charge (APC). Moreover, in 2013 IEEE launched an inter-disciplinary OA megajournal, IEEE Access (IEEE Access, 2015).

Authors who publish articles in Springer's LNCS series or in the conference proceedings of ACM, both of which are subscription-based, have the option of paying a "ransom" to make individual articles openly accessible (ACM Author Rights, 2015), (LNCS @ SHERPA/RoMEO, 2014). Hence, the hybrid publishing model, which is now offered by many journal publishers, has started to establish itself in the area of conference proceedings, too.

Open Access offerings

The following list of OA offerings in computer science is not exhaustive, but serves only to provide a rough overview. For journal searching, the Directory of Open Access Journals can be recommended. Some offerings (e.g. CoRR and LIPIcs) are open to all subfields of computer science, while others (journals, in particular) are limited to one subfield.

Open Access pre-print repositories

Open Access conference proceedings

Open Access journals

Conference management systems

Conference management systems play an important role in the organisation of scientific conferences in computer science. The availability of free or open source conference management applications is an important factor in the planning and organisation of conferences and workshops. Even though there is no concrete connection between OA and the availability of free and open source conference management systems, the latter must be borne in mind when developing future OA offerings. The following is an exemplary list of conference management systems that are frequently used in computer science:

Literature and data searches in computer science

Because of the publication culture in computer science, with its focus on conference proceedings, this field is not well covered by trans-disciplinary, journal-based reference and search systems. Since the early 1990s, the computer science bibliography dblp has evolved into a comprehensive open data service that enables users to search individual authors, conferences, and journals in computer science. The following is a list of search services:

Key players

Key players in the area of OA in computer science include

  • individual professional societies and associations (EATCS, IACR, USENIX, IACR, SoCG, ...),
  • Schloss Dagstuhl - Leibniz Center for Informatics, and
  • individual scientists who operate OA services.

Open science

Computer science acts as a technological pioneer for the current debate on open science. It was a computer scientist, Ben Shneiderman, who was instrumental in coining the term "Science 2.0" (Shneiderman, 2008) and who initiated the debate on the technological renewal of scientific processes. There are also open services such as GitHub (GitHub, 2015) that support the provision of OA to software-oriented research activities. The Semantic Web and linked open data are at the core of research in applied computer science. Knowledge generated in these areas directly promotes open science both in and beyond the field of computer science.



Content editor of this web page: Dr Marc Herbstritt, Schloss Dagstuhl - Leibniz Center for Informatics