Open science and the SDGs: what the UNESCO Recommendation asks of countries
Science can only help reach the Sustainable Development Goals if the people who need its results can read them, check them and build on them. Open science is the umbrella term for the policies and practices that try to make that possible. Since 2021 it has had a global reference text: the UNESCO Recommendation on Open Science, the first international standard-setting instrument on the subject.
Why open science is an SDG 17 question
Goal 17 is about the means of implementation, and two of its targets deal directly with knowledge. Target 17.6 calls for more North-South, South-South and triangular cooperation "on and access to science, technology and innovation" and for better knowledge sharing on mutually agreed terms, including through a global technology facilitation mechanism. Target 17.8 asked for the full operation of the technology bank and the science, technology and innovation capacity-building mechanism for least developed countries. The UN's Goal 17 page lists both targets and reports that the Technology Bank for the Least Developed Countries has been established.
Behind these targets is a practical problem. Much publicly funded research ends up behind subscription paywalls; datasets are kept on private drives or in formats no one else can use; and researchers in lower-income countries often pay the highest relative price for access, while contributing data and samples that others publish. Open science is meant to correct part of that imbalance.
The UNESCO Recommendation in brief
The Recommendation was adopted by UNESCO's General Conference in November 2021. According to UNESCO, 193 countries agreed through it to abide by common standards for open science. Like other UNESCO recommendations, it is not a treaty: it sets out principles and asks member states to act on them, and to report on what they do.
A broad definition
UNESCO describes open science as a set of principles and practices that aim to make scientific research from all fields accessible to everyone, and to make the production of knowledge more inclusive, equitable and sustainable. The text deliberately goes beyond open access to articles. It covers:
- open scientific knowledge: publications, research data, software and source code, hardware, and educational resources;
- open science infrastructures: the repositories, platforms, archives, laboratories and computing facilities that make sharing possible;
- open engagement of societal actors, from citizen science to collaboration with communities affected by research;
- open dialogue with other knowledge systems, including indigenous knowledge.
Values and principles
Four core values frame the text: quality and integrity; collective benefit; equity and fairness; diversity and inclusiveness. Six guiding principles translate them into practice: transparency, scrutiny, critique and reproducibility; equality of opportunities; responsibility, respect and accountability; collaboration, participation and inclusion; flexibility; and sustainability.
Two of them deserve emphasis. "Flexibility" acknowledges that there is no single route to open science, and that a country with little research infrastructure should not copy the model of one with a dense network of repositories. "Sustainability" is a warning about projects that publish data on a platform funded for three years and then disappear.
The seven areas of action
The Recommendation asks member states to:
- promote a shared understanding of open science, its benefits and challenges, and the diverse paths to it;
- develop an enabling policy environment;
- invest in infrastructures and services that contribute to open science;
- invest in training, education, digital literacy and capacity building;
- foster a culture of open science and align incentives with it;
- promote innovative approaches at every stage of the scientific process;
- encourage international and multi-stakeholder cooperation, in order to reduce digital, technological and knowledge gaps.
The fourth and seventh points connect open science to the skills agenda of the SDGs: a published dataset is of little use to someone who cannot download, clean or analyse it. We look at how those skills are defined and measured in digital and high-tech skills for the SDGs.
Data: FAIR, and CARE
Open data does not simply mean posting files online. The FAIR principles, published in 2016 in the journal Scientific Data by a group of researchers, funders and publishers, ask that data be Findable, Accessible, Interoperable and Reusable: described with rich metadata and persistent identifiers, retrievable through standard protocols, expressed in shared vocabularies and formats, and released with a clear licence and provenance. FAIR is not the same as open: data can be FAIR and still restricted, which matters for health records or the location of endangered species.
The CARE principles for Indigenous data governance, published by the Global Indigenous Data Alliance in 2019, add a complementary perspective: Collective benefit, Authority to control, Responsibility and Ethics. They respond to a long history of data being collected from communities without their say in how it would be used. The UNESCO text echoes this concern in its call for dialogue with other knowledge systems and respect for the rights of those who hold them.
Research infrastructures are where these principles meet daily practice. Biodiversity e-infrastructures, for example, depend on shared standards to combine observations collected by thousands of contributors; see our explainer on LifeWatch ERIC. In health, where openness must be balanced with privacy, controlled-access repositories and federated analysis are common compromises, as discussed in our guide to genomics in public health.
Tensions worth knowing about
- Who pays for open access. When journals move from subscriptions to article processing charges, the cost of reading falls but the cost of publishing rises. Authors without grant funding, often in lower-income countries, can be shut out of publishing instead of reading. Diamond open access, where neither readers nor authors pay, is one response, but it depends on stable public or institutional funding.
- Infrastructure that lasts. Repositories, identifiers and archives need long-term funding and governance. The Recommendation favours community-owned, non-commercial infrastructures; many researchers still rely on commercial platforms whose terms can change.
- Incentives. Careers are still largely judged on publications in prestigious journals. Sharing data, code or negative results takes time and is rarely rewarded. Aligning evaluation with open practices is one of the hardest parts of the agenda.
- Openness and protection. Personal data, dual-use research, and traditional knowledge all limit what can responsibly be opened. The principle often summarised as "as open as possible, as closed as necessary" is easy to state and hard to apply.
- Measuring progress. Member states report on implementation, but open science is diverse and much of it hard to count. Indicators based on the share of open access articles capture one dimension well and the others poorly.
What to look for in a national open science policy
For anyone assessing whether a country, a funder or a university is acting on the Recommendation, a few concrete questions are more revealing than declarations:
- Does the policy cover data, software and educational resources, or only articles?
- Is there funding for repositories and their long-term maintenance, not just for new platforms?
- Are researchers evaluated, at least in part, on what they share?
- Are there rules for sensitive data and for research involving indigenous peoples and local communities?
- Does the policy include training, and does it reach institutions outside the main research centres?
Open science will not by itself deliver Goal 17. But the Recommendation gives member states, funders and researchers a common vocabulary, and a yardstick against which the gap between principle and practice can be measured.
Cooperation and skillsDigital and high-tech skills for the SDGs
What indicator 4.4.1 measures, the connectivity gap, frameworks such as DigComp, and what tends to work in building skills.
Research infrastructuresLifeWatch ERIC and biodiversity data
How a European research consortium gives ecologists shared data, models and virtual labs, and what that means for SDGs 14 and 15.
Health and dataGenomics in public health
Sequencing for pathogen surveillance, rare disease diagnosis and newborn screening, and the equity questions WHO has raised.