What actually happens to science when it opens up? From your perspective as a science researcher, can you describe why this perspective is important?

I believe that science takes place in a social environment and certain images of science emerge or have emerged, which are questioned in many respects in a digital and now increasingly open context or are subject to tensions. In this respect, it makes sense, especially in the context of crises and challenges to question how the image of science changes in such situations. Scientists are now interacting more intensively with the public and science is in even greater demand and directly consulted for decisions. I think this has something to do with the way science is negotiated in society and how science is perceived in different contexts.

When science is more negotiated, when it is asked, when it is called upon to open up, how does it change?

I believe that that is actually more of a questioning of one’s own role in different contexts. I believe that the way of communicating, the way of presenting oneself in different forms – public is a broad field, so to speak, but in different forms – in social debates is perceived as more relevant. One also wants to prevent, for example, distorted perceptions of scientific knowledge from reaching the public. In this respect, it is increasingly becoming part of being a scientist to think about how I communicate, what image I want to create in society and also to critically reflect on what knowledge can legitimately be brought into a decision-making process. 

There are also different scientific disciplines for which, of course, the communication may be quite different. Do you have an opinion on this?

I think that in this respect, I am completely with you that one cannot actually speak of science, but must use the whole thing in the plural and that there are very different ways of dealing with the public. Every discipline or every research field has its own ways of doing research, but also its own ways of communicating. In the social sciences, in sociology, for example, there are fields that intensively seek social dialogue, that engage in social debates, especially when it comes to questions of social inequality. But this is also increasingly true for other disciplines, for example, in the context of problem-oriented research, such as climate or health research. There are also subjects that have more narrowly defined publics, perhaps also a transfer or practical perspective, such as the engineering sciences, which at least say in their own accord that their results are less suitable for reaching broader publics. We know this, for example from the science survey, an instrument that we have here at the DZHW, that there are definitely differences in the way disciplines, in the way individual research fields communicate with their publics.

You have already touched on the question of opening up science and what happens to science as a result. I would also be interested to know whether trust in science is strengthened by open science or by opening science?

Basically, the idea is that by making scientific information and scientific knowledge accessible, by making data accessible and generally by creating transparency for the scientific production process, trust in scientific production and knowledge production will be strengthened. In other words, that something good is really happening, that something is happening that is relevant, that something is happening that is transparent. In practice, it’s at least ambivalent and it doesn’t always lead to an increase in trust. Yes, we increasingly have scientific publications that are also publicly accessible and that are accessible to various actors, journalists, NGOs, etc. and these publications also reach broader social circles than previously. That is an achievement. 

On the other hand, open science has also led to the uncertainty of knowledge, the provisional nature of knowledge, becoming clearer to a broader public. For example, when we think about how quickly knowledge circulates in preprints and how it then becomes clear through peer reviews that it is often provisional knowledge that can be contradicted. We see, for example through open peer reviews, that reviews in science, which contribute to stabilizing and certifying knowledge, are sometimes formulated very harshly. As scientists, we experience this frequently, and the case of Christian Drosten, who was confronted with scientific reports in the Bild newspaper that were quite similar to what we know, shows that this can of course also cause insecurity in society. Uncertainty, so to speak, about the fact that scientific knowledge has to be questioned again and again and that it is not as stable as sometimes assumed. And that doesn’t necessarily increase trust in science when this becomes visible. So I would say that there is friction between the claim of the opening and the actual result in communication.

I sometimes ask myself to what extent these projects of opening up, which we are now dealing with more frequently, to what extent do they actually go hand in hand with new commercialisation initiatives? To what extent does an opening up of science also mean at the same time a surrender of certain scientific autonomy? From your observation, are there forms of appropriation, do we have to reckon with? Can you perhaps address this specific idea of commercialization?

I think the idea of open science, so to speak, was that there would be a new ‘commons’ of scientific knowledge. So that also the scientific places of publication belong to science itself. And that has not turned out to be the case. In many cases, the major publishers have simply re-established open access as a business model with sometimes very high fees. The idea that there would be a decommercialization, for example, in the area of publishing, has unfortunately not been fulfilled. On the contrary, there are areas in scientific publishing and scientific production that have become even more commercialized as a result of this opening process. One example is the large amount of metadata that arises from the fact that we communicate with each other as scientists, for example via digital platforms, which now exist in much greater numbers, where knowledge and publication recommendations are shared. This knowledge is also increasingly being used by major publishers. Partly by offering these services themselves, which tap into these metadata or the process data about this interaction, but also by simply buying up, for example, new companies that offer services or organizations that create services, let’s think of Mendeley, for example. If you look at the entire scientific production process, you can see that there is more and more metadata about what scientists do and how they communicate with each other that is available to commercial publishers or commercial players.

So beyond the commercialization of scientific publication, also a use of the data, the researchers, the keyword data tracking. For what?  

It is certainly in the interest of large companies, such as Clarivate Analytics, to generate indicators of the speed of the review process by simply measuring the review processes in these digital platforms, measuring the speed of the review processes. That can become a new product in the future, a new indicator. This can be fed into the research information system. This will be transferred to a research information system. In this way, entire new data realms are created, in which knowledge about researchers is linked, as it were, to what is known from the publications, from the citations, with these new digitally mapped interactions.

Could you express one or two thoughts on how we can actually create good conditions for opening science? Which infrastructures would be sustainable and which reward structures and metrics, for example, would open science in a sustainable way?

I think the first key for opening up the sciences is actually to reflect on the diversity of science. Often this open science discourse gives the impression that there is one universal solution that fits all sciences. I think that’s a fallacy. I think science studies can also help here by highlighting the diversity, the variety, the epistemic practices, the forms of communication that I previously mentioned. Open science is not a universal project; it takes place in very different forms. It must be our concern to simply reflect on the diversity of the forms of knowledge and thus also on the diversity of the opening processes. In other words, diversity is a very important key. 

The second aspect, I believe, is the participation of the disciplines in these opening projects. This should not be a technocratic top-down approach; on the contrary, it should be an invitation to the various disciplines to participate in the development of, for example, open infrastructures. I think a very positive example in Germany, is the development of the national research data infrastructures, in the NFDI, where basic services are now also being promoted. I think the approach is good to intensively involve disciplines and specialist societies in the development and  networking of research infrastructures. Also it is important to develop these services bottom-up, which I believe is an essential element and could be a recipe for success. In any case, I am very excited and believe that this discipline-specific or field-specific form of participation with simultaneous communication about the differences, comparisons about the differences, can be  key in successfully shaping opening processes. I am quite curious, because we have the luck of doing accompanying research on the basic services of the national research data infrastructure. How does this process of developing open research infrastructures actually take place and how are field-specific differences mapped? 

The question about metrics is of course not an easy question. There was one idea that if you develop metrics, that is, if you develop reward systems that map new forms of open science that would make it easier for open science to prevail in various fields. There was also this initiative to develop new, so-called alternative metrics, and from my point of view, this has not worked well in many cases, in that this has also been left to a large extent to the platform operators. Let’s just think about the Research Gate Score, which was very platform-specific and also had a bit of the idea of incentivizing Open Science as a metric, where it was not at all clear what was actually being measured? For example, the significance of interaction with certain platforms was not clear. We definitely need more research in order to develop metrics that are not completely devoid of meaning. Another example, which I’m sure many people are familiar with, is the altmetric donut, which contains different calculations: the counting of Twitter, Facebook, etc., where it’s not at all clear how this is actually connected, according to which mechanisms this indicator is constructed and, above all, why is it done? In any case, we need to develop more time in order to find more suitable sources where substantial research or work is being done. If I think of GitHub, for example, actual systems where people work collaboratively on data and code, where I think information is actually created via scientific collaboration processes. We can perhaps take a closer look at such infrastructures and develop targeted metrics for them, in order to reward this infrastructure work in the long term. But perhaps first of all we must find out how one can actually depict such forms of scientific activities that are not publications. How can you actually map them?  

Take away messages: Meaningful metrics, rewarding infrastructure work, participation, recognition, and taking diversity seriously. Clemens, Thank you so much for taking the time and I hope we continue this conversation.