Defining the ecosystem: open science, open data and open source
Open science
The formal definition of open science provided by the UNESCO Recommendation on Open Science, is that open science is a set of principles and practices that aim to make scientific research from all fields accessible to everyone for the benefit of scientists and society as a whole. Open science is about making sure not only that scientific knowledge is accessible but also that the production of that knowledge itself is inclusive, equitable, and sustainable.
Open data
This refers specifically to the research outputs: the raw datasets, clinical results, and statistics. Open data practices ensure this information is shared according to the FAIR principles (Findable, Accessible, Interoperable, and Reusable) without restrictive copyright barriers, allowing others to verify or build upon the findings.
Open source
Open source refers to the software and code used to analyze the data. It requires much more than just access to the code. The software must be distributed under a license that guarantees free redistribution, permits derived works, and strictly prohibits discrimination against any person, group, or field of endeavor. This ensures that a global community can collaboratively peer-review, build, and improve the tools used for scientific discovery, without the constraints of commercial vendor lock-in.
The practical benefits for science and society
Solving global crises
Open science is a fundamental requirement for addressing complex and interconnected challenges. These issues need a more integrated approach to scientific knowledge to foster sustainable development. The global COVID-19 health crisis is a good example of why equitable access to scientific information is necessary. Fast sharing of data and research results was essential for responding to the emergency.
Accelerating scientific progress
Adopting more open, transparent, and collaborative practices makes the research enterprise more efficient, which in turn improves the overall quality, reproducibility, and impact of scientific work. Reproducibility and scrutiny are central to safeguarding the integrity of science. When scientific findings are subjected to critique through verifiable data and open peer review, they build a more reliable foundation for subsequent research. This level of transparency provides the robust evidence required for informed decision-making and is a key factor in maintaining public trust in the scientific community.The international movement towards openness
The international movement towards openness
International policy frameworks
The real turning point for the open science movement arrived in 2016 with the publication of the FAIR Guiding Principles. By demanding that research data be Findable, Accessible, Interoperable, and Reusable, FAIR fundamentally changed how the scientific community operates. These principles had a huge impact on open science, and then, in 2021, UNESCO stepped in to provide the world with a landmark global framework: the UNESCO Recommendation on Open Science. This framework establishes the first universal definition and a set of shared values and standards designed to govern how we create and share knowledge on a global scale.
Requirements from major funding bodies
While the high-level ideals of UNESCO set the stage, the real-world shift toward open science is being driven by the organizations that hold the checkbooks. The era of "closed by default" is ending because the world’s most influential funding bodies have decided that if the public pays for the research, the public should see the results. Consider these hard facts:
- The United States National Institutes of Health (NIH): The NIH leads the charge by requiring that all peer-reviewed articles arising from its funding be deposited in PubMed Central within one year of publication. Furthermore, for larger projects, researchers must submit a formal data management plan outlining exactly how they intend to share their findings.
- The United States National Science Foundation (NSF): Similarly, the NSF mandates that investigators provide plans for the management and availability of all data, software, and other research outputs.
- The Wellcome Trust: This international heavyweight has even stricter timelines, requiring open access within six months of publication. If the Trust provides the funding to cover publishing fees, they specifically mandate the use of a Creative Commons Attribution (CC BY) license, ensuring the work can be reused and redistributed freely.
Both the NIH and the Wellcome Trust have officially begun suspending or withholding grant funding from scientists who fail to meet these open access requirements.
The cost of closed science
Reducing the cost of duplicated research
Keeping science closed source creates a massive financial drain known as duplication costs. Researchers across the globe spend valuable time and money reinventing the wheel: collecting, cleaning, and transferring data that already exists elsewhere. This fragmentation turns the scientific enterprise into a series of expensive, isolated islands.
Archiving data openly serves as a significant boost to productivity. A study of over 7,000 NSF and NIH awards found that research projects with archived data produced a median of 10 publications. Projects that kept their data closed produced only 5. Sharing the raw materials of science effectively doubles the output of the same initial funding.
The financial efficiency of open source software
Instead of locking useful algorithms behind private walls, open-source models let us build platforms that actually speed up innovation.
Trying to build this kind of specialized software from scratch using a closed-source, corporate model is an absolute money pit. Just replicating what is already out there in a private environment could easily drain tens of millions of dollars. But when we build in the open, a whole network of different research and software communities across the globe can pitch in to create world-class tools. That is something no single organization could ever afford to build or maintain on its own.
We can look at the cBioPortal community as a perfect example. They have built a massive global ecosystem of clinical and tech contributors who collaborate to scale open-source tools for clinical genomics. With over 8 million visits and deployments across 93+ institutions globally, the platform has become an essential resource for cancer research and precision medicine.
Relying on these community-driven platforms, organizations can eliminate inefficiencies and duplicated work. It stretches R&D budgets further and keeps organizations from getting trapped by commercial vendor lock-in. At the end of the day, building on open software ecosystems means the people doing the research keep full control over their data, workflows, and tools.
The Netherlands, a blueprint for national open science
Coordinating a nationwide approach
The Netherlands operates as a primary case study for what happens when a country moves beyond theory to implement an integrated, systemic national strategy. This transition is anchored by a significant financial engine: the Dutch Ministry of Education, Culture and Science is dedicating 20 million euros annually until 2032 to fund the National Initiative on Open Science. Such a long-term commitment provides the stability needed for researchers and institutions to shift their fundamental workflows without the fear of a sudden resource vacuum.
The Dutch model relies on a 'quadruple helix' collaboration. This approach brings together government bodies, knowledge institutions, industry partners, and citizens to ensure that the creation and sharing of knowledge serve the entire society. By involving these four distinct groups, the program aligns scientific output with the real-world concerns and economic needs of the public.
To make sure these high-level ideals translate into daily practice, the strategy operates across three distinct levels. Locally, the program funds and strengthens bottom-up communities of academics and support staff within their own institutions. On a domain-specific level, thematic competence centers—covering fields like Life Sciences, Health, and the Humanities—standardize good practices and data formats unique to those disciplines. Nationally, coordination teams handle systemic steering and negotiations, ensuring that the entire country moves toward openness in a synchronized fashion.
Goals and reality for a FAIR digital infrastructure
Handling the vast output of Dutch scholarship requires a focus on the digital plumbing of science. The Netherlands has placed a high priority on ensuring that research outputs like data and software are FAIR: Findable, Accessible, Interoperable, and Reusable. This technical standard ensures that scientific work is machine-readable and ready for others to scrutinize, replicate, or build upon.
The Dutch strategy sets specific, measurable targets to track this transition, originally stating that by 2025, 40% of digital scientific objects in Research-Performing Organizations should be FAIR, and then scaling to an ambitious 75% by 2030. However, the lack of sufficient financial backing to support these high-priority goals directly impacted the results. Compliance data is usually affected by a time lag. For example, an evaluation report published by the Dutch Research Council (NWO) and ZonMw in February 2025 analyzed research outputs from 2023. This report revealed that while 44% of the 2023 publications contained a data availability statement, only 20% actually mentioned data sharing, and a mere 3% shared software.
These figures demonstrate that, just two years before the 40% deadline, the scientific community was quite far away. This is largely because the cost and effort of making data and software FAIR fall heavily on researchers without adequate financial support or a compulsory mandate. While the government promotes open-access publishing, article processing charges (APCs) are, on average, €2,000 per paper (depending on the publisher and journal), and dedicated national budgets often cover only a fraction of the country's total publishing output.
Given the reality of these results and the current funding structure, achieving the 75% target by 2030 is improbable. Reaching that level of compliance will likely require the government to either implement a strict compulsory open science policy, or finally match its high priorities with high funding by fully covering the costs of data archiving, software maintenance, and open-access publishing.
The Diamond Open Access solution
To fully realize the vision of open science, the path forward lies in enabling and accelerating the adoption of Diamond Open Access.
Diamond OA is an equitable scholarly publishing model, supported by shared infrastructures and open licensing, providing scholarly knowledge as a digital public good with no fees for readers or authors. Under this model, journals and platforms are funded directly by universities, research institutions, and collaborative grants, completely removing the €2,000 APC barrier. These conditions enable scholarly knowledge to circulate freely, allowing all researchers to publish and access science openly, regardless of institutional wealth or individual grant funding.
UNESCO advances Diamond Open Access in line with the UNESCO 2021 Recommendation on Open Science regarding open scientific knowledge and scientific publications. This work also contributes to the 2030 Agenda for Sustainable Development, the UN Pact for the Future, and the UN Global Digital Compact. These global frameworks emphasize equitable access to knowledge, inclusive digital transformation, and international cooperation.
The European Open Source Strategy
The European Union has explicitly linked open-source architecture to its strategic autonomy. Situated within the broader Tech Sovereignty Package, alongside frameworks like the Cloud and AI Development Act, the EU Open Source Strategy addresses the systemic vulnerability of over-reliance on dominant, non-EU technology vendors.
In order to prevent the economic value of European innovation from being concentrated externally, the EU has adopted a comprehensive lifecycle approach to open source, structured around four primary objectives:
- Technological Sovereignty and Deployment: Prioritizing open-source funding in critical sectors and scaling open source tools to provide secure, EU-aligned alternatives for public digital infrastructure.
- Ecosystem Viability: Establishing mechanisms like the Open Source Maintenance Instrument to map critical software dependencies, ensure codebase security, and provide structural and legal support for open-source startups.
- Public Sector Integration: Positioning public administrations as anchors for establishing Open Source Programme Offices (OSPOs) and embedding open standards into government procurement processes. In order to achieve this, an Open Source Solutions Catalogue has been created.
- International Standardization: Integration of EU open-source communities directly into global standardization processes to promote European digital tools internationally.
The commercial and technical friction working against adoption
With major funding bodies issuing mandates and 193 countries officially ratifying the UNESCO Recommendation on Open Science, it seems like openness should already be the global default. Yet, the transition remains incredibly slow. The reality is that ratifying a high-level diplomatic agreement is easy, but rewriting the foundational mechanics of global research is hard. There are several systemic factors that contribute to this:
- Ratification does not equal enforcement: When a country ratifies a UNESCO framework, it signals there is a shared value, but it does not automatically pass as a binding domestic law. Without strict, compulsory mandates at national levels, most institutions will default to tradition, closed methods of operating.
- The academic reward system and cultural inertia: The global academic system operates on a “publish or perish” model. Researchers are rewarded, promoted, and funded based on publishing in prestigious (and often closed) commercial journals. Furthermore, there is a fear of losing the first-mover advantage to rival labs.
- Commercial friction and intellectual property overusage: In fields with heavy private investment, the slow adoption of open science is fundamentally rooted in the tension between regulatory goals for data sharing and the private sector's drive for commercial exclusivity. Corporate entities frequently prioritize economic interests over public knowledge, utilizing aggressive intellectual property strategies to monopolize research outcomes. A major factor working against open science is the routine over-claiming of trade secret protections and database rights to lock away raw data, creating a "black-box" of information that evades external scientific scrutiny.
The necessity of open science infrastructures
Building sustainable digital services
Open science requires a long-term strategic investment in technical and digital infrastructures to work effectively. These investments provide the structure for a transparent research cycle, moving beyond the traditional closed model to one that prioritizes collective benefit. Such infrastructures consist of non-commercial computing facilities, repositories, and digital public services that must guarantee permanent and unrestricted access to research products such as data, source code, and hardware specifications.
Success in this field relies on the preservation of digital and academic sovereignty. To ensure knowledge remains a public good, digital infrastructures should be built on open-source software stacks and governed directly by the scientific community. This approach allows researchers to retain autonomy over the content and organization of their work rather than ceding control to external entities. Maintaining community-led governance prevents a dangerous over-dependency on commercial providers that often impose restrictive terms of use. By steering these systems publicly, the scientific community can safeguard its values and ensure that data stewardship remains focused on discovery rather than profit extraction.
The role of dedicated IT enablers
A professionalized technical workforce is essential for turning these infrastructure goals into a reality. This transition has led to the emergence of specialized roles, specifically data stewards and research software engineers, who bridge the gap between raw information and usable knowledge. These experts manage data according to rigorous principles, ensuring that the influx of digital information does not become unmanageable.
Their primary responsibility is to ensure that digital scientific objects remain FAIR—Findable, Accessible, Interoperable, and Reusable. Achieving this requires building robust research software ecosystems that allow for the seamless archiving and sharing of code across different platforms. Specialized IT companies and service providers, such as The Hyve, act as partners to help institutions build and automate these complex data stewardship networks. These enablers provide the technical depth needed to implement machine-actionable metadata and connect local repositories to international clouds.
The global movement toward open science represents a practical and economic necessity for the future of research. Opening the processes of knowledge creation multiplies the opportunities for local and global participation while reducing the redundant costs of data collection. This vision can only materialize through the maintenance of robust, transparent IT networks and the support of dedicated digital enablers.
Finding the balance: The boundary between open and closed science
While this document strongly advocates for open science, it is vital to acknowledge that not all science is, or should be, entirely open. The guiding principle for this boundary, explicitly established by frameworks like the EU's Horizon Europe programme, is that research data should be "as open as possible, as closed as necessary".
Closed science is permissible (and legally protected) when it is required to safeguard legitimate interests. This includes adhering to strict privacy and data protection standards (such as the GDPR) to protect vulnerable patients, as well as preserving legitimate trade secrets and intellectual property required for commercial exploitation. Without the ability to protect complex inferred data or highly customized algorithms, the private sector would lose the financial incentive to invest in costly research and development.
When is open science imperative?
Conversely, open science becomes a strict imperative when withholding information actively damages the public good or stalls scientific progress. Under the "essential facilities doctrine," if a dataset is an indispensable input for innovation and cannot be reasonably duplicated by other means, refusing access eliminates competition and stops scientific advancement. Furthermore, basic safety and efficacy data, such as the outcomes of clinical trials, must remain open. Hiding adverse event information behind the guise of commercial trade secrets does not protect a competitive methodology; it merely burdens the public's right to health.
How can we work together for the common good?
Open and closed science do not have to be mutually exclusive, they can function together through strategic public-private partnerships. The Open Targets consortium serves as a good example of this balance. In this model, pharmaceutical competitors and academic research institutes collaborate openly in the pre-competitive phase. They pool raw genomic data and share insights to identify biological targets for diseases, treating this foundational knowledge as a digital public good. Once these shared discoveries are established, companies transition to a closed competitive phase, utilizing their proprietary methodologies to develop and patent specific therapeutic drugs. This hybrid approach ensures that foundational data circulates freely to accelerate discovery, while preserving the commercial exclusivity required to fund final drug development.
Summary
Open science is now a global priority, recognized by 193 countries through the UNESCO Recommendation. By making the research process transparent and freely accessible, open science accelerates innovation and reduces the high costs of duplicated research.
However, the transition to open science remains slow, despite new policies and mandates from major funding bodies. Several structural issues stand in the way. Academic reward systems often discourage early data sharing, technical infrastructures are frequently underfunded, and commercial entities sometimes use trade secrets to withhold valuable clinical data. Furthermore, as seen in the Netherlands, setting ambitious national targets for FAIR (Findable, Accessible, Interoperable, and Reusable) data is not enough. Without adequate financial support and clear mandates, the scientific community struggles to meet these goals.
Moving forward requires practical changes in how science is governed and funded. It requires long-term investment in community-led digital platforms and the technical professionals needed to maintain them. The scientific community must also clearly define the boundary between open and closed science—ensuring data is "as open as possible, as closed as necessary" to protect legitimate intellectual property while supporting public health. Finally, to reach the level of equity envisioned by UNESCO, the research ecosystem should transition toward equitable publishing models like Diamond Open Access, ensuring that scholarly knowledge serves as a digital public good.