The starting point needs to be that the maximum possible number of people should be able to have access to the results of science and scholarship. Ultimately, that will provide the greatest benefit for science and scholarship.
Contact-image
Robbert Dijkgraaf
Royal Netherlands Academy of Arts and Sciences (KNAW)
The starting point needs to be that the maximum possible number of people should be able to have access to the results of science and scholarship. Ultimately, that will provide the greatest benefit for science and scholarship.
Contact-image
President
Royal Netherlands Academy of Arts and Sciences (KNAW)

Prof. Robbert Dijkgraaf

Robbert Dijkgraaf is tenured professor at the University of Amsterdam, and director and Leon Levy professor at the Institute for Advanced Study in Princeton. July 1, 2012 he stepped down from his position as president of the Royal Netherlands Academy of Arts and Sciences (KNAW). 
In 2003 Robbert Dijkgraaf was awarded the NWO Spinoza Prize, the highest scientific award in the Netherlands. He has been a guest professor at universities including Harvard, MIT, Berkeley, and Kyoto, and is on the editorial boards of numerous scientific periodicals, and is also the scientific adviser to institutes in Cambridge, Bonn, Stanford, Dublin, and Paris.


In July 2012 Robbert Dijkgraaf met Vice-President of the European Commission and Commissioner for the Digital Agenda Neelie Kroes and talked about Open Access.

 

 


President Robbert Dijkgraaf van de Koninklijke Academie van Wetenschappen schreef een aanstekelijke column in de NRC van 28 april 2012 over de voordelen van Open Access in de wetenschap.

Onderzoeksresultaten verdienen het met zoveel mogelijk mensen gedeeld te worden.

"Londen was op zijn mooist. De lente hing in de lucht. Op weg naar de collegezaal in de City passeerde ik de restanten van het Occupy-tentenkamp bij de trappen van de St. Paul's Cathedral. Hoe de strijd tegen het mondiale grootkapitaal zal aflopen weet ik niet, maar er waart op dit moment ook een bevrijdingsgolf door universiteiten en laboratoria".

Lees meer in de column.


In 2009 Robbert Dijkgraaf was interviewed about open access.

Are physicists concerned about the topic of Open Access?
Yes, very much so, particularly in the field of high-energy physics. In fact, this discipline is the avant-garde as regards Open Access. That’s because of the long-standing preprint culture that we have, with research results being distributed to colleagues immediately after the article has been written. That goes back as far as the 1960s, in the age of paper. Back then, the preprints came by post; in this country, it sometimes took two or three months, which is a significant amount of time in my field. But the announcements came sooner, via printouts from a database of preprints maintained by colleagues at SLAC, the accelerator lab at Stanford. You could then use special printed order cards to get an offprint from the author. That meant that you could be up to date on the latest results within about two weeks.

In about 1990, we started sending out preprints by e-mail. That had disadvantages too, however, for example because your mailbox quickly filled up, certainly if you didn’t have a chance to check it. We discussed it at a conference in Aspen, and Paul Ginsparg casually mentioned that the problem could easily be solved via “the Web”. “The what?”, we asked. Ginsparg then went ahead and put together the first Web archive within an afternoon. The first web page I saw was one that used a Gopher to give access to the new ArXiv database. You could collect the articles themselves by means of FTP. A year later, there was already a Netscape interface, which was followed by the WWW. Since then, that approach has spread unchecked to cover other disciplines: mathematics, mathematical biology, astrophysics (although the astrophysicists generally don’t upload their publications to ArXiv until they’ve already appeared in a journal).

That development has had enormous consequences for my field. It has created a worldwide level playing field. The fact that you are immediately brought up to date on the literature worldwide meant, for example, that it was much easier for me to start up a research group here in Amsterdam. You no longer lag behind groups at Princeton or Harvard that are close to the actual research.

Citations of preprints are not registered in Web of Science or Scopus. Is that a problem?
I have to admit that I’m sometimes guilty of that too. Sometimes I simply forget to publish because I’ve already moved on to the following article. What was perhaps my best-ever article was never actually sent to a journal. I wrote it jointly with a senior colleague and we simply knew that it was a good article. And that was that! I find it a lot more exciting, for example, that my formula is named after me.

By the way, you can in fact look up the citations on special web sites that count preprints as well as publications. To be honest, I have a bit of a mixed role here because in my other job (President of the Royal Netherlands Academy of Arts and Sciences, KNAW, LW) I often say how good Dutch science is measured by the impact of our publications. I realise perfectly well that preprints don’t count for that impact.

Robbert Dijkgraaf (photo: Annemiek van der Kuil) Robbert Dijkgraaf (photo: Annemiek van der Kuil) Robbert Dijkgraaf (photo: Annemiek van der Kuil)

Managers and financiers often base things on citation and impact scores. What influence do citation statistics and impact factors have in your own discipline?
You will never get a job or an assessment of 'excellent' merely on the basis of citations. It is far more important whether people quickly build on your work and disseminate your insights. That appears from publications but also at conferences or in research proposals. And citations of preprints are then sometimes a more effective indicator than those of the actual official articles, which appear a lot later. But the significance of those citations is relative, certainly in the field of particle physics. In big experimental groups, the number of authors of an article may be as many as two thousand. The names of the authors appear on any publication that derives from their collaboration. Impact factors and prestigious citation indices don’t play a dominant role in my discipline.

So why do they in other disciplines?
In my field, people are rather blasé about the applicability of our model; they think the rest of the world needs to follow what the physicists do. But we are a relatively small discipline, with a very specific past history. My role with the KNAW makes me even more aware of the fact that physicists are not the norm. Sometimes, disciplines have become so big and therefore so anonymous that formal publications and citation indices may well need to play a bigger role. The unfortunate thing is that the more distanced you are from research, the more you have to rely on figures and the closer you are the less you dare to do so. In the KNAW, we are really concerned about this. A purely quantitative approach does not do justice to the great diversity between the disciplines in the culture of publication or the way in which quality is measured and assessed. The typical number of publications differs enormously between disciplines. Take medical sciences, for example, in which it is not unusual for a middle-aged researcher to have a publication list running to more than a thousand publications, whereas in mathematics your age is more or less asymptotic for the number of your publications. If you don’t know that, you will take the best mathematician to be worse than a mediocre medic. The role of technology is also very different in the various disciplines.

To be honest, I see only advantages for Open Access.

I have my own theory about the success of Open Access in my discipline. We have never aimed to develop a comprehensive and perfect model all at once from the very beginning. We have always worked towards maximum distribution of knowledge on a bottom-up basis: from paper preprints to e-mail and to ArXiv, which does in fact have a 'light' form of quality control via a small number of moderators. It’s naturally relevant, of course, that for us reproduction is the ultimate quality control. The risk of this whole development, however, is that you never move beyond the initial phase.

But where high-energy physics is concerned, an obligation to publish via Open Access doesn’t really seem to be necessary.
Where communication between scientists is concerned, that is true. The preprints system works just fine in that respect. It has even allowed me to read articles for Nature before they were published. I never – absolutely never – go to the website of a journal except for articles from before 1970.

Why don’t other disciplines start using this system?
If you were to start on about it nowadays, you would immediately get a discussion about copyright and publishers. We had that at the time too. But it never occurred to us that copyright would play a role or that you would need to ask publishers for anything. We just did it. It also has to do with the extent to which the scientific community feels responsible for spreading its knowledge. You live according to the rules that you set yourself and there’s no higher instance that orders you to do anything. Disciplines differ in that regard.

Robbert Dijkgraaf (photo: Annemiek van der Kuil) Robbert Dijkgraaf (photo: Annemiek van der Kuil) Robbert Dijkgraaf (photo: Annemiek van der Kuil)

Should the bodies that finance research play a role here?
Definitely – it’s already happening. More and more research councils are starting to require that the results – the articles – of research that they have financed should be made publicly accessible worldwide. But that naturally also applies to universities, which are increasingly wondering to what extent they are visible in a world that is becoming more and more knowledge-dependent. MIT, for example, took a major step by putting lecture notes on the Internet. Individual scientists and scholars are also uploading their articles to their own website. Google Scholar then makes them available.

To be honest, I see only advantages. My own experience concerns a research group in India that is performing really well. In the past, members of a group like that moved on to places where they were closer to the information. That’s no longer necessary: we are now literally linked to one another. I check the Internet every morning and so do they, which means that we have a level playing field as regards knowledge. If I publish an article today, I already get questions from Japan or Iran the very next day. That means that the dynamics of the scientific process are becoming totally different. If you do not want these open dynamics, you need to investigate carefully what interests are then involved. At the moment, those who are in favour of openness have to constantly argue for it. In fact, the burden of proof should be the other way round.

Ultimately, the principle is that the government invests large amounts in research because it is seen as a public interest matter. The basic principle should therefore be Open Access to the results.

That’s the position adopted by the European Commission. But what about the Dutch government?
The Dutch government could adopt a firmer position. The Minister of Education, Culture and Science, Ronald Plasterk, was emphatically in favour of Open Access when he was a scientist and I believe he wants to see more action in this area, and to enter into discussion with parties like the Netherlands Organisation for Scientific Research (NWO). At the KNAW, we have made Open Access a firm component of our Strategic Plan. We are also taking measures to put it into practice in various places within the organisation. For example, we have an institute with an Open Access fund to provide funding for the publication of articles, with publishing in Open Access journals, and an institute that is taking part in the Springer experiment. We are consulting with the NWO and the Association of Universities in the Netherlands (VSNU) about this matter because we also believe that we need to act jointly. Each organisation just acting by itself is pointless.

The Dutch government could adopt a firmer position. The Minister of Education, Culture and Science, Ronald Plasterk, was emphatically in favour of Open Access when he was a scientist and I believe he wants to see more action in this area, and to enter into discussion with parties like the Netherlands Organisation for Scientific Research (NWO).

Isn’t there then a risk that it will be the slowest that determines the pace?
I don’t think so. There are naturally all kinds of initiatives. Besides what we are doing, there are the universities with their repositories. Individual researchers are also taking action. There was recently an article in the NRC Handelsblad newspaper. Open Access is definitely on the way. But we don’t intend getting involved in the breathless approach of the Open Access movement, with their constant question “Why hasn’t it been arranged yet?” As scientists, we were there too when the accent came to be on the publishers. You can’t just reverse that.

How do you see the future role of the publishers?
Basically, the publishers do valuable work. Their business model is not sustainable, but organising the review process and the journals are important tasks. We also mustn’t lose sight of the long-term effects: a journal needs to still be in existence in twenty years time.

The publishers were traditionally very close to the authors, something that you can still see in the world of literature. In the world of science, that model has got lost in the huge volume of research. The interests of shareholders have taken precedence. In fact, it’s very reminiscent of what happened with the banks.

What has to happen is that the publishers need to link up once more with the values of the scientific community and for us to develop a model that reflects those values. The Open Access model could be such a model. Open Access would appear to be viable, and more and more Open Access journals are being awarded a good impact factor, sometimes remarkably quickly. That’s encouraged by the fact that some top scientists have deliberately started publishing in Open Access journals. There are also some journals that have switched with their whole editorial board to a learned society and are continuing to publish their journal, but now as an Open Access publication. It’s the kind of individual activism that ultimately makes possible a large-scale movement. So far, all those changes are all going in a single direction. Open Access really is on its way.

There are also some journals that have switched with their whole editorial board to a learned society and are continuing to publish their journal, but now as an Open Access publication.

How?
Universities are not very happy about paying double as at the moment – for licences for the journals on the one hand and for publishing articles in the same journals on the other. An experiment by Springer in the Netherlands may perhaps provide a solution. As long as the licence with the Dutch university libraries still runs, Springer is prepared to place articles by Dutch authors in the Open Access component of its journals free of charge. Another very recent initiative comes from five universities in the United States, including MIT, Harvard, and Berkeley. They are calling on universities to form funds, if necessary only modest ones, to pay for the publication of articles, with the idea being that if we do that jointly we can speed up the transition to the new situation and shorten the current agony.

The essential thing is to have the will and the ability to cooperate, including internationally and across disciplines. Part of the problem is due to the 'walls' that there have been between the various scientific disciplines. By getting rid of those walls, we can inspire one another reciprocally. That doesn’t mean “one size fits all” but that we can learn from one another on the basis of a common desire for greater visibility and accessibility for research. Physicists can definitely make a contribution to that debate.

The discussion will also need to be about books. That doesn’t just bring us into the domain of the humanities: books play an important role in mathematics too.

The discussion will also need to be about books. That doesn’t just bring us into the domain of the humanities: books play an important role in mathematics too. A mathematician is almost by definition someone who walks around with a yellow book under his arm (i.e. a book published by Springer, L.W.). The publication culture and the role of the publishers is then very different. Mathematical articles are in that sense rather like mini-books. Besides being a high-energy physicist, I’m also a mathematician. The differences are significant. Take peer reviews, for example. In physics, these often just involve a few sentences: “Good article, but take another look at this or that.” In mathematics, you get a report several pages long with a solid, detailed commentary. Really impressive. Mathematics journals are also very precious documents, with the publishers playing a genuinely enriching role. They provide support for conferences, and editorial boards are groups of friends. My predecessor at the KNAW, Frits van Oostrom, may perhaps look puzzled when he hears the word 'preprints', but in the Open Access debate he will be well able to understand the mathematicians.

The extent of digital experience within a discipline also plays a role in the debate, as well as the size of the community, the role of reports, etc. That doesn’t mean that we all have to wait for the slowest; rather, it involves throwing the right switches at the right time in the various different situations. Open Access is the basic principle, and you will gradually see it finding its way into the requirements for providing subsidies. That won’t just be the case with the NWO: the universities also finance research, as do private funds and industry. They are also facing the question of how open they want access to be to 'their' knowledge. The starting point needs to be that the maximum possible number of people should be able to have access to the results of science and scholarship. Ultimately, that will provide the greatest benefit for science and scholarship.

Interview by Leo Waaijers and Annemiek van der Kuil on 23 September 2009.
Photography by Annemiek van der Kuil.