By Mariano Fressoli [1] and Valeria Arza [2]

[1] Associate researcher at Scientific and Technical National Research Council (CONICET) and Research Center for the Transformation (CENIT). E-mail:
[2] Independent researcher at CONICET and Director of CENIT.


The general idea behind open science is simple: the free availability of publications, data and other scientific tools could democratize access to knowledge and increase the possibilities of collaboration between scientists and the rest of society. In addition, by collaborating widely across disciplines and cognitive backgrounds several learning economies could kick off.

Over time, scientists have begun to take up practices aiming at sharing data, publications and problems by using social networks and electronic mediums like This appears to have opened up the possibility of creating new forms of collaboration that promise to be transformative. Scientific institutions and development agencies such as the OECD, the World Bank and others have shown interest in these practices, and are developing policies and incentives for open science. Much of this interest seeks to increase the transparency, impact, and efficiency of scientific knowledge production. However, there are still questions regarding the democratization potential of open science. Is Open Science capable of generating a genuine culture of collaborative production of common goods to be fully appropriated by everyone?

This article explores three challenges faced by open science regarding promises for a more democratic science: data accessibility, knowledge appropriation, and barriers to collaboration in the context of diversity.

Access and accessibility

Publications play a key role in the scientific system. On the one hand, they drive the competition between scientists: the system of peer review privileges ‘novel’ results, and scientists must consequently worry about being the first ones to discover them. Through these publications, scientists gain prestige and can advance their careers. Without access to publications, scientists would neither be able to know the state of the field, nor establish common challenges and collaborate with each other. Thus, it is assumed that once scientists validate their knowledge, the results are published in the form of articles becoming part of a public heritage that anyone can access.

In practice, however, this does not happen. On the one hand, commercial publishers running most major scientific journals like Nature or Science can charge around US$40 for accessing an article. Forty U.S. dollars do not look like much, but when 20 or 30 references are used to produce an article, the bill increases rapidly. When this expense is considered at the national level, the sum is enormous. In Argentina, the Ministry of Science, Technology and Productive Innovation spends about US$ 25 million per year on subscriptions to scientific publishers, and these are only available for institutions with access authorization.

The other problem is that, unlike hackers and programmers, while scientists are habituated to publishing articles, except for a few disciplinary exceptions, we are not acquainted with the habit of sharing data, projects, laboratory notebooks, software, instrument design, evaluations of articles and projects, etc. Therefore, there is a lot of information that could be useful to other social actors, but which is not shared.

The open access movement seeks, through different practices, to increase the available knowledge and is mainly responsible for the construction of digital repositories. In Argentina, the recently regulated Law 26899 for the Creation of Digital Institutional Repositories of Open Access, Owned or Shared allows to create the infrastructure necessary for scientists to begin to share their publications and their data not only with other scientists but also with the rest of the population. The Open Repositories Act is a major step towards building an important infrastructure for open science. The free availability of data and publications increases the visibility of scientific research: a recent study on the use of open repositories in Latin America shows that 25.2% of articles are consulted by non-academic users, either to satisfy their personal interest (10.5%) or their professional interest that is unrelated to scientific practice (14.7%) (Alperin, 2015).

The creation of an infrastructure for the storing the production of scientists and its free access by others is only one aspect for ensuring accessibility. For example, as Leslie Chan, OCSDnet PI recently highlighted, the increased mainstreaming of Open Access could in fact be even further entrenching particular inequalities around who can produce knowledge. Furthermore, it is also important to understand that data availability is not equivalent to building knowledge. Data is information, but its use requires skills and learning that are not always available. Reading scientific publications requires a certain understanding of the problems and concepts and, above all, it requires tacit knowledge (Collins, 2010) if one wants to replicate the available knowledge.

A case of open science in Argentina serves to think of this dilemma. In 2009, the New Argentinean Virtual Observatory (NOVA) was created, which brings together the most important astronomical observatories in Argentina. NOVA’s goal is to collect and centralize already processed astronomical data – in the form of images, spectra, catalogs, lists or measurement tables – to enable its reuse by other scientists and the community. As of October 2015, the NOVA site had received around 85,000 visits and recorded over 1,000 data downloads. However, to access the data, it is necessary to have advanced knowledge of astronomy and, above all, some idea of ​​what is being sought. Fortunately, NOVA partnered with Cientópolis, a computer science group at the National University of La Plata, which develops citizen science tools. In cooperation, the two organizations developed a series of free video games like Galaxy Conqueror or Runaway Stars (in development) that allow any citizen to visualize the available images (photos of galaxies or stars) and to contribute to its classification. Other possible solutions may include the construction of data visualization mechanisms, infographics, documentaries, web animations, etc.

Translating data into formats accessible to other social actors is something that scientists do not always know how to do by themselves. Interaction with social communicators, designers, artists, hackers, etc. can help to develop translation mechanisms that act as a bridge between the information and the next generation of new knowledge.

Citizen participation and knowledge appropriation

Citizen participation in the creation of scientific knowledge is nothing new. For the last 170 years, workers, technicians, sailors and explorers, hunters and amateur naturalist, housewives and philanthropists have voluntarily helped with the collection and classification of data (see Cooper, 2012). Only recently, has this been denominated as “citizen science” (Catlin-Groves, 2012).

The use of new telecommunication tools, like the internet, mobile phones, sensors, etc. has revived the citizen science tradition by making it possible for thousands to contribute online from all around the world. Some of the most famous open science projects such as Galaxy Zoo, the Great Sunflower project or eBird have thousands of participants generating millions of data points. Such enormous amounts of data have also facilitated the development of new tools for visualization and analysis. For example, eBird makes use of artificial intelligence tools to create certain maps. These maps show changes in the population of certain species and their relation to environmental changes, including phenomena like climate change, environmental contamination, etc.

Although the motivations to participate can vary, one key factor for the success of these projects is the generation of interesting questions for the public and the creation of simple tools that enable the participation of  people with different origins and capacities. The standardization of data collection and upload protocols makes it possible to evaluate the data quality quickly and eliminate those points which do not comply with the established parameters. Following this approach, independent of the growth of the number of participants, scientists can assure the data quality and protect their credibility.

But the necessity to standardize tools and protocols, consequently, limits the possibilities for the public to provide knowledge and to participate in the process. One way to tackle this issue is to use social tools like forums, contests and meetings where participants can discuss their experiences, learn about the movement and create a sense of  belonging. The usage of social tools is important to loyalize users and promote participation and engagement (Wiggins and Crowston, 2015). Nonetheless, a majority of projects which  take advantage of public participation do not provide the right surroundings to learn more than just the capabilities relevant for the project. Usually, citizen science projects are designed by scientists and involve the creation of standards or kits to allow participation in specific task such as data collection, but there is much less involvement of citizens in the analysis of information, conceptualization or writing up.

The absence of a space where citizens are able to come up with their own inquiries, to progressively use the knowledge available and to collaborate on assigning priorities, limits and ultimately, undermines any attempt by society to appropriate science.

Some initiatives, like the Public Lab in the United States or the Center of Academic Technology at the Universidad Federal do Rio Grande do Sul, in Brazil, offer some clues on how the citizen science model could go beyond the simple collection of data. These spaces build open source (code) tools for environmental monitoring while benefitting from the collaboration between scientists and members of the community. The objective is to actively involve citizens and to facilitate the appropriation of these technologies.

Of course, these cases are neither easy to replicate nor do they present the only approach possible. Actually, it is likely that in order to introduce more participative mechanisms, new institutions and spaces are required, which facilitate more horizontal ways of interaction between scientists and citizens. This leads us to the third challenge of open science, the collaboration between actors with different capacities and interests.

Collaboration in the context of diversity

The belief that interdisciplinary collaboration can resolve complex global problems is not something that has just emerged recently. During World War II, the coordination of several disciplines, actors and institutions turned out to be fundamental for the development of technological inventions, such as the atomic bomb and the radar. Scientific organizations have promoted the establishment of interdisciplinary projects for decades and every time, it is more common that scientists of distinct laboratories collaborate in the use of infrastructure, technologies, and research resources that were generated by public funding.

The development of new instruments like open databases for data and publications, or open hardware and tools for online collaboration are reinforcing the interest in collaboration practices and creating new strategies. For example, the Polymath Initiative shows that voluntary online collaboration of dozens of mathematicians enables the solution of complex problems in record time. The usage of big data techniques allows scientists, who come from different disciplines, to use the data for diverse purposes. The availability of large publication databases opens up the possibility for software to conduct text and data mining, and to, consequently, discover unexpected correlations in health issues. All these possibilities, in reality, benefit from data that can be viewed from different interdisciplinary angles. When scientists are able to interact fluently (between themselves and other social actors), new forms of collective intelligence can be generated. They then can share, validate, or discard thoughts like ideas, speculations, or hypotheses in a much more dynamic way. This, at least, is what open science promises.

Nonetheless, despite this progress, two structural barriers make collaboration more difficult. First, the scientific incentive system neither rewards collaboration with other actors nor the sharing of data, methodologies or knowledge. Scientific groups that venture into interdisciplinary production can improve their comprehension of problems and how they face the generation of solutions. Unfortunately, their research also takes longer and they possibly encounter obstacles for publishing their results (Leahey et al. 2017). 

Second, not many mechanisms or models exist that allow for the establishment of flexible collaboration between actors that have different conceptual and cognitive backgrounds. A considerable amount of literature on interdisciplinary collaboration already exists in the scientific setting. However more practical guides or manuals that contain more than just some simple recommendations (e.g. create trust, establish clear compensation mechanisms, etc.) are not available. This is likely because each case of interdisciplinary collaboration is unique and requires the development of new methods to manage disciplinary differences. Perhaps then, the development of flexible instruments and institutional strategies to enable new spaces for interdisciplinary learning will help to promote the identification of common problems.

A much more complex challenge is how to attract other social actors to collaborate and participate in knowledge creation. As we saw in the previous section, the participation of other actors in citizen science projects is often limited and standardized. This ensures that contributions from the public are reliable, but it comes at the price of limiting not only more flexible collaboration but also the promotion of creativity. This is a tension that we are just starting to understand: namely the need to balance reliable mechanism of validation of knowledge with the need to include a wider diversity of actors in the production of scientific knowledge. If open science is to become a more democratic force for development, we need to find new ways to tackle this issue.

So, how to tackle these limitations? One possible solution is to attempt learning from these spaces of open and collaborative, production such as makerspaces, digital fabrication laboratories (fab labs) and citizen laboratories that are developing methods and tools to foster collaboration between actors of different backgrounds, with differing interests and knowledge. Events like hackathons (programming or technology prototyping marathons), project training, and open days, where new actors are invited to propose new ideas, can be examples to further explore.

Again, these spaces are not easy to replicate in universities and research institutes. The few existing examples seem to have troubles expanding participation beyond their own students and scientists. Despite these challenges, these examples serve as proof that it is possible to design new forms of collaboration and knowledge creation.

Final remarks

While the idea of open science gains new supporters among scientists and funding institutions, the expectations coming from the transition towards a new form of producing knowledge are rising. However, some tensions still remain that create drawbacks for fully achieving its democratization potential. These include: the tension between the increasing availability of data and scarcity of mechanisms to translate information into knowledge that could be appropriated by anyone; the tensions between inviting contributions from anyone and the traditions of competence and authority that underpin scientific institutions, and finally the tension between the unlimited theoretical possibilities of collaboration and the incentives and capabilities to do so in practice in the context of diversity. Perhaps it is the right moment to understand that scientists and policy makers need to begin experimenting with new collaborative formats and practices present in other collaborative communities, such as hacker spaces, maker spaces, fab labs to learn from each other.



Alperin, J. P. (2015). The Public Impact of Latin America ’ s Approach to Open Access. Stanford University. Retrieved from

Catlin-Groves, C. L. (2012). The citizen science landscape: From volunteers to citizen sensors and beyond. International Journal of Zoology, 2012, 12.

Collins, H. (2010). Tacit and explicit knowledge. Chicago: University of Chicago Press.

Cooper, C. (2012). Victorian-Era Citizen Science: Reports of Its Death Have Been Greatly Exaggerated. Scientific American Blogs. Retrieved from

Leahey, E., Beckman, C. M., & Stanko, T. L. (2017). Prominent but Less Productive : The Impact of Interdisciplinarity on Scientists ’. Administrative Science Quarterly, 62(1), 1–35.

Wiggins, A., & Crowston, K. (2015). Surveying the citizen science landscape. Firts Monday, 20(1), 1–15. Retrieved from


Featured image source: STEPS America Latina