6 Principles of Open Science
As the guiding principles of open science take hold, new publishing formats and services are emerging. These align with a philosophy of sharing articles, code and data from the research process in order to empower researchers.
In this way, making research results freely available becomes a natural part of the scientific process and not an afterthought once the project is completed.
A core principle of open science is that research output should be freely available to the scientific community and the public. Ideally, this would include all data, materials and analysis scripts alongside published articles.
Openness requires a cultural change away from solitary and secretive science practices that are often driven by competition for scant resources or the desire to make a discovery before others. Nevertheless, there are barriers to full engagement with open science, such as the stigma around publishing in open access journals or the pressure to maintain high impact factors when job hunting.
Barriers must be addressed and incentives for engagement should be replaced by those that encourage researchers to embrace the principles of openness. cOAlition S is a leading initiative to address these barriers and champion best practice.
As part of its practice, open science encourages collaboration among scientists and with citizens. This requires changing the ways that scientific research is evaluated and incentivized, moving away from esoteric metrics such as research papers to broader measures of academic impact, fostering a community of scholars focused on collaboration rather than competition.
It also means easing the pressure that many scientists feel to produce headline-worthy research as quickly as possible, so they can meet the demands of funding bodies and meet societal expectations to deliver solutions. This also requires ensuring that all researchers have access to shared data and resources. This would include agreeing upon common data structures, vocabularies and metadata standards that would allow scientists to easily integrate open datasets.
Research that can be reproduced allows other researchers to test and verify findings. It also helps to improve scientific methods by increasing collaboration and sharing of data sets.
Reproducibility is essential for ensuring the validity and reliability of science . A lack of reproducibility can have negative impacts, including on health, slower scientific progress, wasted time and money, and may erode public trust in science.
Reproducibility can be promoted by providing incentives for scientists to publish their work. It can be encouraged by making all scientific outputs available through open access and by facilitating collaboration among researchers and between researchers from different disciplines (e.g. in astronomy and astrophysics).
This pillar seeks to lower barriers that prevent researchers from sharing research products and/or accessing existing knowledge resources. It also champions efforts that enhance the quality of research through the sharing and re-use of existing information.
The democratic school advocates that everyone has a right to access the results of scientific research. It opposes the idea that publishing journals should claim copyright over experimental data, which prevents it from being reused.
It also argues that scientists should not be under pressure to produce headline-worthy results quickly, as this will lead to wasteful work. Instead, the school proposes that the research process be made more efficient by collaborating with external people and institutions.
The new digital technologies are triggering novel open science principles and practices in universities. This is expanding the research ethos of these institutions to encompass not only university researchers, but also companies/industries, municipalities, citizens and international organisations in a mission-led approach for societal improvement.
It increases the accessibility, transparency, and reliability of scholarly outputs. It also aligns with the principle of equity, diversity and inclusion by opening the research process to marginalised scholars and societal actors beyond the traditional scientific community. It also counters the Matthew effect (Merton, 1968/1988) where established scientists get credit for work that can equally be attributed to non-established researchers.
The open science ethos holds that scientific outputs should be freely available to all. Yet a commercial collection of vast databases called Big Data can potentially generate barriers to research that the open science ethos seeks to erode – including imposing fees for access and blocking secondary research that would be more useful than the original work itself.
Legal obligations around, for example, personal data protection, create legitimate boundaries to opening up research, but much of research is also regulated by soft law, guidelines and shared norms. Reflections on these questions could help researchers to operationalize open science recommendations in ways that are responsive to and compliant with relevant research ethics and integrity norms.