Interview with Karel Luyben, first President of the EOSC Association

Karel Luyben, first president of the EOSC AssociationKarel Luyben, a representative of the European association of leading universities in science and technology, CESAER, was elected the first President of the EOSC Association at the first General Assembly of the EOSC Association in December 2020. Due to his outstanding contribution in the field of Open Science, his candidacy for this position was supported by the European University Association (EUA), the League of European Research Universities (LERU), the European Federation of Sciences and Humanities (ALLEA), and many others. Karel Luyben is the former Rector Magnificus of the Delft University of Technology, national coordinator for Open Science in the Netherlands, and the Chairman of the Board of the Dutch Techcentre for Life Sciences (DTL).


Mr. Luyben, what is your definition of the European Open Science Cloud?

Let me start by saying what EOSC is not. It is not going to be a big cloud, although the cloud is in the name and it is not an e-infrastructure.

In the SRIA, Strategic Research and Innovation Agenda, you can find a yin-yang model that on the one hand presents the e-infrastructures and on the other hand the data infrastructures. If you have a computing power, a network, or storage capacity and no data, it represents nothing. If you have data and you do not have an e-infrastructure, you have nowhere to store it. Data is only relevant with the e-infrastructures and the e-infrastructures are only relevant if have data in them, so they need each other and cannot exist without each other, like the yin-yang model.

In my dream, 20 years from now, EOSC has become a system (a commons) similar to what is now the World Wide Web where the Internet is the backbone. In this vision, something like 50% of the relevant research data would be available through these commons, this federated network to be created through the protocols and standards. And by relevant data I mean those data that are seen relevant by the researchers themselves because someone else cannot determine what data are relevant.

What EOSC is cannot be answered in a few sentences. For me, the EOSC is going to be ultimately a common system for relevant data that are made FAIR and thus can be found on the Internet, and thus are made interoperable and can be reused, in other words a Web of FAIR data.


Tell us something about your engagement with EOSC in the period that preceded the formation of the EOSC Association.

EOSC logoMy interest in the domain of data and things relevant for EOSC started a long time ago when I realized that my PhD students were producing data but they could not use each other's data because we did not have the necessary metadata. So, I tried to organize at least some exchange of the data in the department between the PhD students.

My engagement in this area began in 2013 when I became Chair of the board of the Dutch Techcentre for Lifesciences (DTL). DTL connects scientists, data experts, technical experts, and trainers that are specialised in a variety of high-end wet lab and data technologies, and working in life science domains ranging from health to nutrition, agro, biotech, and biodiversity. At one of DTL's workshops in 2014, the acronym FAIR emerged. Further work on this led to the Nature publication “FAIR Guiding Principles for scientific data management and stewardship” in 2016. Today, it has become a globally accepted set of principles that helps to make data Findable, Accessible, Interoperable, and Reusable.

For the last two years, as a President of the EOSC Executive Board, I have participated in discussions with the European Commission about the form of organization that will manage the EOSC. We realized that forming an association with membership, adopting the statutes of the new organization and defining a partnership with the European Commission was the way forward. This process led to the establishment of the EOSC Association in July 2020. Founding members were CESAER, CSIC, GÉANT, and GARR. We deliberately started with the minimum number because there were about 20 to 30 organizations that wanted to become a founding partner, which would have made things complicated.

What key points would you emphasize from that period?

ilustracijaThere were three ‘legally’ independent bodies created by European Commission: Executive Board, Governance Board, and Secretariat that worked together with the Commission as the fourth entity. We all realized that this was only going to work if we all did our best to work together. Otherwise, it would fall apart. We had people like Cathrin Stöver, Vice-Chair of the EOSC Executive Board, that spent quite a considerable amount of time working on it and six working groups Landscape, FAIR, Architecture, Rules of participation, Skills and Training and Sustainability with devoted chairs also spending much of their time and effort on this development.


Over the last summer, there were open consultations on SRIA. What community response have you received then, were they actively involved and who showed the greatest interest?

We had material coming from working groups and there were over 200 reactions to the consultations. The major response came from the research performing organizations, this is in line with the members of the Association. Some 62% of the members are research performing organizations. The first official version of SRIA will come out soon.


At the General Assembly of the EOSC Association, 142 new members were officially admitted to full membership and 49 as observers. Was this according to your expectations?

IlustracijaAround September 2020, I expected that we would have about 50 members by the end of the year so the number of current members is clearly higher than I expected. This is certainly not the end; we might expect that the number of members of the Association will reach several hundred. Membership is open at every moment; you can still apply for membership to become an observer or to become a member. The next step-up will be at the General Assembly in the middle of this year when we will formally accept the newly registered organizations to become members. We see a lot of interest because in Horizon Europe the requirements of Open Science will be strengthened and all the projects will need to comply to a certain extent with making the data as FAIR as possible, deposit the data, make their publications open access, etc.


What will be the key activities of the EOSC Executive Board in 2021?

We have nine Board members, officially called directors, including myself being president. The forthcoming activities of the Executive Board are the organization of the General Assembly in the middle of this year. We hope to elect the Secretary General of the EOSC Association soon. We held a meeting to discuss possible advisory groups to operate within the EOSC Association. Among other things, the Board is currently finalizing a proposal for a Memorandum of Understanding on the EOSC Partnership with the European Commission under the new EU Horizon Europe Research and Innovation Programme and the SRIA v1.0. One of the important things is of course to disseminate this information and your interview may help there.


What are you most proud of so far?

If you ask me personally, I am proud to be part of the development of the European Open Science Cloud for what I hope that after 20 years it will be something like the World Wide Web for data, the Web of FAIR data, and related services.


SRCE is currently the only member organization of the EOSC Association from Croatia and also a mandated organization. Do you think that the (high) membership fee is an obstacle for a larger number of members from Croatia and other countries in the region?

I expect that more organizations from Croatia will become members of the EOSC Association. We will probably need to consider the membership fee and start differentiating the membership fee based on the size of the organization or their involvement in this domain as the organization can be very large but the data part can be very small. So, we have to come up with criteria for differentiation.


Since you are the national coordinator for Open Science in the Netherlands, can you tell us what is the key challenge for SRCE and other mandated organizations?

I would expect a mandated organization to work in two directions, in the direction of the country and the direction of the Association. Mandated organizations should aim to connect with Open Science stakeholders at the European and national levels.

At the national level, mandated organizations should bring together and connect organizations that are actively involved in Open Science. I expect them to organize a coherence and interaction with all other organizations in the country. Through mandated organizations, we could have a better interaction with the countries.   

EOSC portalIn the Netherlands, in 2017 we started with our first national plan on Open Science. At the moment we have just started writing our next plan for the next 10 years. I was talking about 20 years for the EOSC to come to full maturity. So, we are talking about cultural changes in the domain of research and in other domains that will take time. We use a definition for Open Science in which there are three lines of development: first towards open access of which we have done quite a lot already. I think our country has the highest percentage of open access. Next, we are setting up the plan for the FAIR data program, the most expensive, most elaborate, and most difficult development with the largest effort in it. And third the development of Citizen Science. Recently we had a meeting with all the parties involved in FAIR data to discuss with them how will we organize our national FAIR data landscape. Knowing where we are, knowing where EOSC is, and knowing where Horizon Europe is, would we still create the same organizations if we would start now? Well, in my personal view I would create a data infrastructure network next to the e-infrastructures network, so that we create this ying-yang model with two types of organizations. Why two organizations? If you are a provider of high-performance computing, you want to have the fastest and biggest computer possible. If you are providing storage, you want to have the highest storage capacity possible and if you are servicing the connections, you want them to be as fast as possible. So, this is a competitive state of mind. However, on the data side, you need a collaborative state of mind. You need people that want to share their data that want to jointly make something happen that want interoperability. Now I am not saying that the same person cannot do both things, be competitive at the one end and be collaborative at the other end, but this not easy. And therefore, I am pleading in all the countries in Europe to separate this to a certain extent and, of course, you need to have an overlap. You need also data stewards in the e-infrastructure organizations and ICT experts in the data organizations. You need the two approaches, collaborative and competitive. Whatever you do in Croatia, discuss this, make it explicit, make both needs explicit and come to a constructive yin-yang approach.

Croatian version of this interview was published in the bulletin SRCE novosti nr, 84 (PDF, 4 MB).

Interview was conducted by Kristina Posavec, PhD and Ivana Veldić Špehar, SRCE

*image source