Computer Science

With roots in mathematics and electrical engineering, the still young academic discipline of computer science has always adopted an open approach to the re­search results achieved. Success stories include open source operating systems, such as Linux, and the typesetting system LaTeX, which is well established in the field of scholarly publishing, in particular.

Because of the many links to other disciplines (bioinformatics, business infor­matics, …), framework conditions for scholarly publishing differ across the indi­vidual subfields. The main focus of the present article is on the core of compu­ter science.

Since the mid-1990s, conferences, symposiums, and even workshops, have established themselves as publication venues for the dissemination of research results, and they hold a dominant position vis-à-vis the classical scholarly jour­nals. This is also reflected in the programme areas of scholarly publishers and of professional societies that act as publishers. However, the strong focus on conference publications regularly leads to perception problems in the scientific landscape, for example, when this form of publication is ignored in evaluation processes (Jagadish, 2008; Vardi, 2009; Vardi, 2010; Konstan & Davidson, 2015). This tension between conference publications and journal articles also manifests itself in the current open access offerings in the field of computer science. In individual subfields (e.g., theoretical computer science and crypto­logy), there is a well-established preprint culture. Various publication venues for workshops and conferences have become established that differ in terms of their level of requirements or their thematic focus. Among journals, progress in the direction of open access is somewhat subdued but nonetheless tangible. The commercial publishers have recognised that open access is an additional source of revenue. However, many open access journals in computer science operate without the backing of a traditional publisher, and are organised from within the community (which sometimes means that the degree of professiona­lisation is very different, for example, with regard to getting indexed in the Directory of Open Access Journals (DOAJ).

The national and international scientific societies in the field of computer science have taken a more conservative approach to date in relation to open access, although there has been some movement on the topic in recent years:

• The German Informatics Society (GI): Although the GI signed the Berlin Declaration of Open Access to Knowledge in the Sciences and Humanities in 2007, and although it makes its series Lecture Notes in Informatics (LNI) freely available online, the scholarly journals of the thematic fields are still strongly tied to the classical subscription model. Via the Digital Library, subdivisions of the GI can make their own publications available in open access.
• Association for Computing Machinery (ACM): The US-based ACM has been participating in the Plan S initiative since January 2021. Moreover, via ACM Author-izer, ACM authors can generate and post links on their homepage and or in their institutional repository for visitors to download their articles from the ACM Digital Library free of charge. However, from a licensing point of view, it often remains unclear whether reuse is permissible. Hence, the goals of open access are only partially achieved. Authors also have the possibility of making articles published by ACM available in open access by availing of the author-pays option. 
• Institute of Electrical and Electronics Engineers (IEEE): The IEEE Computer Society is the section of the professional society IEEE that is responsible for computer science and engineering. It acts as a publisher for scholarly journals and conference proceedings. Open access is available in hybrid journals, and especially in the open access megajournal IEEE Access. However, there are no explicit possibilities of publishing articles in open access in IEEE conference proceedings.

Some commercial publishers offer authors the option of paying a fee (article processing charge, APC) to make individual articles available in open access (ACM Author Gateway; Lecture Notes in Computer Science). This hybrid publishing model, which is practised in the area of journals by many publishers, has also become established to a certain extent in the area of conference proceedings.

The provision of open access to data and software is an important step towards transparency of scientific works. These open data and software allow results and experiments to be reproduced to a certain degree. In the field of computer science, so-called artefact evaluations have become established both in the area of conferences and of journals, and related data are very often made freely accessible.

As of March 2024, the Directory of Open Access Books (DOAB) listed 3,220 titles under the keyword Computer Science.

Because of the conference culture in computer science, the majority of open access books are conference proceedings. Purely monographic works are rarely published in open access. The following list contains the most important conference proceedings series in computer science: 

• CEUR Workshop Proceedings
• Electronic Proceedings in Theoretical Computer Science (EPTCS)
• JMLR Proceedings
• Lecture Notes in Computer Science (LNCS) (Springer; hybrid via Open Choice)
• Lecture Notes in Informatics (LNI)
• Leibniz International Proceedings in Informatics (LIPIcs)
• Open Access Series in Informatics (OASIcs)
• Procedia Computer Science (Elsevier)
• USENIX

Although DOAB and OAPEN list a number of open access books in computer science and its subfields, they are currently very incomplete in relation to open access conference proceedings.

Conference management systems play an important role in the scientific self-or­ganisation of conferences in the field of computer sciences. That such systems are available in open access or free of charge is crucial in the planning and orga­nising of conferences and workshops. Although there is no concrete link as such between open access and the availability of conference management systems, the latter aspect must be kept in mind when developing future open access of­ferings, because conference management systems are where the metadata and the draft versions of contributions are collected. The following is an exemplary list of conference management systems that are frequently used in computer science:

• EasyChair (free of charge but not open access)
• HotCRP (free of charge but not open access)
• SoftConf START V2 (fee based)

Searches in Computer Science

Because of the focus on conference publications in its publication culture, the field of computer science is inadequately covered in trans-disciplinary indexing and search systems that are based on journals. The dblp computer science bib­liography, which was created in the early 1990s, established itself as an open access – and by now broad-scope – bibliographic service. It enables searches for individual authors and especially for conferences and journals in computer science.

The following is a list of relevant search services for computer science:

• ACM Digital Library
• citeseerx
• Computer.org
• dblp
• IEEE Xplore
• Semantic Scholar
• The Collection of Computer Science Bibliographies