- 1 The concept of open science
- 2 Open access to research publications
- 3 Open access to research data (open data)
This article is based on a memorandum compiled by the UBVA on the basis of discussions at a workshop on open science in relation to research data, including licensing models for software and research data, etc., which was held at Aarhus University (Emdrup Campus) on 30 March 2017. The memorandum has been published as part of a larger memorandum on the Danish e-Infrastructure Cooperation (DeiC) website.
The concept of open science
The main idea behind open science is that research, including research data, research publications and other research communications, should as far as possible be made freely available with a view to the better dissemination of knowledge. The benefits of open science have been fostered by, amongst other things, digital technology, “big data” and the globalisation of research.
Some important elements of open science are:
- Open access to research publications, which can be achieved via the publishers of the publications or through parallel publication in an open access repository.
- Open access to research data (open data), which provides others with better opportunities to assess the quality of the research and to utilise the value inherent in the research data already provided. In this connection, a general objective is that research data should be Findable, Accessible, Interoperable and Re-usable (FAIR).
- Open-source software, i.e. software for which the source code is publicly available.
- Active citizen participation in research (so-called “citizen science”).
Today, open science is a recognised research policy objective. It is based, amongst other things, on the view that the results of publicly funded research are a public benefit, i.e. that everyone should be able to make use of them at the same time, without the possibility of excluding others from it. In this context, the European Commission has established an open research data pilot under its Horizon 2020 programme. Many other grantors also impose requirements in this area. A guiding overview of this may be found in the SHERPA / JULIET database.
Open science provides an opportunity to consider how potentially patentable research results should be handled. There are examples of research collaborations in which it is agreed that as part of open science, the research results should not be patented. In such cases, it may often be appropriate to ensure patent prophylaxis, i.e. that inventions are published in such a way that no one can patent them.
However, the goal of open science cannot stand alone. Amongst other things, rules on intellectual property rights, personal data protection and trade secret protection may lead to legitimate restrictions being imposed in this area. The European Commission stated the objective in relation to the Horizon 2020 programme to be “As open as possible, as closed as necessary”. See “Guidelines on Open Access to Scientific Publications and Research Data in Horizon 2020” and ”Open Research Data Pilot”.
The EU Commission’s Digital Single Market (DSM) strategy contains several elements that are intended to support open data/open science, including through a European ”free flow of data” initiative. See, inter alia: Data Ownership and Access to Data – Position Statement of the Max Planck Institute for Innovation and Competition of 16 August 2016 on the Current European Debate.
Regarding the concept of open science, including open access, reference may in particular be made to the following sources (both with links to additional resources):
- Website of the Ministry of Higher Education and Science on open access
- Website of the EU Commission on open science:
Open access to research publications
The basic idea behind open access is that research publications should be digitally accessible to all, free of charge. Most research publications are published by commercial publishers, who require payment from those who wish to read them. Some publishers allow researchers to publish their articles in parallel in an open access repository – possibly after a certain waiting period (“green open access”). In some cases, however, this applies only to the non-peer-reviewed draft article that the researchers have submitted to the journal. Other publishers offer open access publishing in the journal for a fee (“golden open access”).
In 2014, Denmark’s National Strategy for Open Access was adopted, which contains the following vision:
“To create free access for all citizens, researchers and companies to all research articles from Danish research institutions financed by public authorities and/or private foundations.”
The strategy also contains a number of specific objectives for the spread of open access in Denmark.
Scientific monographs and anthologies, publications on patented inventions as well as PhD dissertations and doctoral dissertations are not covered by the strategy, cf. the website of the Ministry of Higher Education and Research.
On developments in the use of open access in Denmark, see Denmark’s Open Access Indicator.
Open access and copyright
Open access to research publications must occur with respect for the researchers’ copyright to their publications. This means, amongst other things, that open access publishing always requires the consent of the researchers involved. You can read more about this in the article “Open Access and Copyright Law”.
Researchers wishing to publish via open access should always examine the open access policy of the journal in question. Some guidance on this may be found in the SHERPA / RoMEO database, which is due to be replaced by a new version in 2020.
Open access to research data (open data)
Researchers and research institutions will usually have an interest in regulating the conditions under which research data may be made available to others. This applies in particular to the question of how others may dispose over research data. For use in this regulation, the following may, inter alia, be highlighted:
Creative commons (CC)
Creative commons licences may be used to regulate the right to use material such as images, texts, audio and video recordings. CC licences can now also be used to regulate the right of use of databases.
A prerequisite for the use of creative commons licences is that this is done with the consent of everyone who owns the copyright to the material in question. For copyright on research data, see “Memorandum on access to and disposal over research data”.
Creative Commons licences allow you to set four restrictions on the use of the licensed material:
- The author must be named (BY)
- Commercial use of the material is not allowed (NC)
- Processing of the material is not allowed (ND)
- Processing of the material is permitted on condition that the processing is made available on the same terms (SA)
These restrictions may be combined in six standard licences: BY(-NC), BY-ND(-NC) and BY-SA(-NC).
Creative Commons has also developed the special CC0 license (“No rights reserved”), in which you as the author/rights holder may completely waive all your rights.
The European Commission recommends that researchers use Creative Commons licences CC-BY or CC0 to ensure free access to research data under the Horizon 2020 programme’s Open Research Data Pilot, see page 20 of these guidelines.
You can read more at: www.creativecommons.org.
Open data commons
The Open Knowledge Foundation (OKF) has developed three standard licences that can be used to provide access to research data:
- Public Domain Dedication and Licence (PDDL): Use of the PDDL licence releases all rights to data/databases to the public domain
- Attribution Licence (ODC-By): The ODC-By licence is similar to the “CC-BY” licence and allows both commercial use and processing, but requires reference to the original database and licence.
- Open Database Licence (ODC-ODbL): The ODC-ODbL licence allows commercial use and processing, but also requires – in addition to reference to the original database and licence – that processing be distributed under the same conditions as the original, and that an open version without DRM access restrictions must be available.
Open access repositories
A number of digital platforms allow for the archiving and provision of access to research publications and research data, etc. (so-called “open access repositories”). Some of these are commercial (e.g. Elsevier, Amazon), while others are non-commercial (e.g. Figshare, Dryad and UK Data Archive).
The following are also of particular importance:
- The OpenAire (Open Access Infrastructure for Research in Europe) portal provides comprehensive access to Horizon 2020-funded research and other repositories.
- Zenodo is the recommended storage location for Open Research Data Pilot projects.
- CLARIN (Common Language Resources and Technology Infrastructure) may be used “to find linguistic data, process data and deposit data”.
Agreements on research data in research collaborations
See the article “Agreements on research collaborations”
Open source software – licences
As a researcher, you should be aware of section 59 of the Copyright Act, which is worded as follows:
Section 59. Where a computer program is created by an employee in the execution of his duties or following the instructions given by his employer the copyright in such a computer program shall pass to the employer.
It is the subject of debate whether researchers at Danish universities and similar research institutions are covered by this provision, at least when they conduct free research, as the university and the employed researcher do not usually have a superior-inferior relationship. You can read more about this in the article “Research and copyright”.
If section 59 of the Copyright Act applies, researchers cannot decide for themselves whether their software may be “open source”, as this requires the consent of the university.
Open source software
As early as 1999, agreement was reached on common guidelines in the form of an “Open Source Definition” (OSD). This has since been updated several times, but the basic principles remain unaltered: Licences that meet the ten OSD criteria are accepted on the open source list, which is certified by the non-profit organisation Open Source Initiatives (OSI). The certification has become the de facto standard and today numbers approximately 80 licences.
The ten OSD criteria: Free redistribution, Source code, Derived works, Integrity of the author’s source code, No discrimination against persons or groups, No discrimination against fields of endeavour, Distribution of licence, Licence must not be specific to a product, Licence must not restrict other software, Licence must be technology-neutral.
Around two-thirds of all open source projects are currently licensed under a few of the most commonly used licences, which are discussed in more detail below. The licences have various different purposes and provisions, but some common basic principles apply (cf. OSD), in particular the user’s freedom to (1) copy and share the product, (2) use the product for any purpose, and (3) process, redefine and create derived works from the product.
The licences are often written in dense language and relate to American copyright law, which can give rise to doubts about their interpretation in a European context, cf. Henrik Udsen, IT-kontraktsret (“IT contract law”) (2014).
Some of the most commonly used licences are briefly described below. In addition, the European Commission has drawn up a European open source licence, the European Union Public Licence (EUPL), which has not, however, yet become widespread.
One of the oldest licences, the GNU (General Public Licence) is estimated to regulate more than half of all open source software. The licence is maintained by the Free Software Foundation, and is now available in the GNU-GPL 3.0 version from 2007.
The GNU-GPL licence protects IT programs, which are defined as any copyrighted work under the licence. A popular example is the Linux operating system. GNU-GPL incorporates a so-called “copyleft” provision, which states that software that is further refined must be made available on the same terms. This means that the licence also regulates the rights to derived works.
The MIT licence was originally issued by the Massachusetts Institute of Technology in 1988. Unlike the GNU-PGL licence, the MIT licence does not contain a copyleft provision. This means that further refined software may be published as proprietary software.
One example that uses the MIT licence is the popular web framework Ruby on Rails developed by the Danish programmer Daniel Heinemeier Hansson.
BSD (Berkeley Software Distribution) licences consist of a number of different licences. They do not contain a copyleft provision and have only been OSI-approved in later versions.
The Apache licence was originally developed for the Apache web server in 1999. Like the BSD and MIT licences, the Apache licence does not contain a copyleft provision. The licence has been available since 2004 in a version 2.0.