The Open Access Interviews: F1000 Founder Vitek Tracz

Vitek Tracz is a hero of the open access movement, and it is not hard to see why. Fifteen years ago he founded the world’s first for-profit OA publisher BioMed Central (BMC), and pioneered pay-to-publish gold OA. Instead of charging readers a downstream subscription fee, BMC levies an upfront article-processing charge, or APC. By doing so it is able to cover its costs at the time of publication, and so make the papers it publishes freely available on the Internet.[See the comment below the Q&A for clarification of this]. 

Many said Tracz’s approach would not work. But despite initial scepticism BMC eventually convinced other publishers that it had a sustainable business model, and so encouraged them to put their toes in the OA waters too. As such, OA advocates believe BMC was vital to the success of open access. As Peter Murray-Rust put it in 2010, “Without Vitek and BMC we would not have open access”.

Today Tracz has a new, more radical, mission, which he is pursuing with F1000.

Vitek Tracz

As always, I have written an introduction to the Q&A below with Vitek Tracz; as sometimes happens, the introduction turned out to be longer than readers might expect, or wish to read.

I have, therefore, put the introduction into a PDF file, which can be accessed by clicking on this link.

Those interested only in the Q&A need simply read on below. 

The Q&A begins ….

RP: As I understand it, F1000 now consists of three main services — F1000Research, F1000Prime, and F1000Workspace. In addition, I believe there is something called F1000 Specialists. Can you say something briefly about each of these services, and when they were launched?

VT: The newly launched F1000 (F1000.com) is an integrated site combining three services: F1000Prime, F1000Research and F1000Workspace.  These services are built and supported through the active collaboration and participation of the largest high-level group of experts (over 11,000 and growing) from across biology and medicine, the F1000 Faculty. This consists of experienced leaders (Faculty Members) and talented young researchers (Associate Faculty Members, appointed by Faculty Members), in about equal numbers.

We started what is now called F1000Prime 13 years ago, which has become the largest and most comprehensive article-level quality assessment of biomedical literature: the F1000 Faculty identify those articles they find interesting in their daily work, rate them at one of the three levels of quality (all positive, the goal is to find the best articles) and write a short text explaining why the chosen article is interesting to them.

F1000Research, launched over 2 years ago, is an open science publishing platform that offers a completely new way of publishing research in biology and medicine: it uses immediate publication followed by transparent peer review, requires the underlying data to be shared, and encourages the publication of all research findings. It also now offers a platform to freely share scientific posters and slides.

Recently, we launched F1000Workspace, a comprehensive set of tools to help researchers write articles and grants, discover literature, manage references and reference libraries, and collaborate and prepare for publication.

The F1000 Specialists are not an external service; they are a growing group of young active supporters of our services who work with us in key institutions to support new users of our services and bring feedback that then contributes to future development decisions.

RP: You say F1000Prime has become “the largest and most comprehensive article-level quality assessment of biomedical literature”. I believe PLOS has introduced “F1000 recommended badges” to PLOS articles that have been evaluated by F1000 reviewers. Have any other publishers or organisations adopted F1000 reviews for assessment purposes?

VT: A number of publishers, institutions and societies identify articles that have been recommended by F1000 by using an F1000 Recommended badge including PubMed, PLOS, BioMed Central, The Physiological Society and The Royal Society. Others take into account F1000 when assessing scientists and articles, including Scopus, CNRS’ Evalscience and a number of funding bodies. We are also in discussion with other publishers to include F1000 recommended badges on their tables of contents.

We are informally told by scientists that at many funding bodies (such as Wellcome, MRC, NIH etc.) F1000 recommendations are considered when making grant assessments. Anecdotally many tell us that F1000 recommendations are seen in grant applications and for tenure and job applications.

RP: Can you talk me through the pricing of the different F1000 services, and say who usually pays for the services (i.e. researchers, their institutions, research funders etc.)?

VT: The new F1000 is available on subscription. It is a continuation of the established F1000Prime service, already subscribed to by hundreds of institutions, including almost all of the top ones. The subscription price is based on the number of relevant potential users in the institution.

All existing and new subscribers now have access to F1000Workspace, and each user in a subscribing institution or an individual subscriber can publish one short free article per year (less than 1000 words, normally $150) in F1000Research.

For all other articles, F1000Research charges competitive Author Processing Charges, and special agreements are available to institutions to include publication in F1000Research as part of their institutional subscription. Sharing of posters and slides on F1000Research is free of charge.

RP: I see a link to F1000Research’s charges here. This shows that APCs depend on article length and go up to $1,000 for papers over 2,500 words. Papers over 8,000 words also attract a $1,000 surcharge, plus a $300 hosting fee for data is sometimes applied. I also see here that individual subscriptions to F1000Prime are $9.95 per month.

Can you give me a couple of examples of what different sized institutions can expect to pay for an annual subscription to F1000? (I think you are saying that the new F1000 subscription now covers both F1000Prime and F1000Workspace?) And can you say whether F1000 has any other revenues beyond the APCs and subscriptions we have discussed?

VT: Institutional subscriptions for F1000 start at £1,750. Pricing depends on the number of staff students and researchers in the life science faculties of the institution. The average price paid is approximately £6,000. APCs and subscriptions are currently F1000.com’s only revenue streams.

RP: Can you give me some stats on the different F1000 products — e.g. the number of reviewers you currently have signed up, the number of papers you have published, and the number of users and access figures for the different services etc.?

VT: F1000Prime has published over 160,000 recommendations of articles published across about 4,000 journals.

The Faculty of F1000 (currently around 11,000 in number) all have the right to recommend articles for F1000Prime, and are also allowed to act as referees for articles in F1000Research.

We are close to publishing 1,000 articles in F1000Research and submissions are growing well.

Despite having only launched F1000Workspace a few weeks ago, we already have a rapidly growing number of researchers registering on the service, and expect to reach over 20,000 registered users before the end of the year, and about 100,000 by spring 2016.

RP: Can you share with me any figures in terms of page views, or other usage stats, that demonstrate the use of the F1000 reviews, or the number of times articles published by F1000Research are read?

VT: In the last year we had over 2.6m page views in F1000Prime.

For F1000Research, we had 0.5m published article page views in the last year. Of course, as the articles are fully OA, they can also be viewed in PubMed Central and other places, for which we do not have the numbers.

In addition, F1000Faculty Reviews (specially commissioned reviews of current topics, published in F1000Research) had 0.25m views in the last year.

Six main areas

RP: You were for many years a traditional publisher. I think it fair to say, however, that since around 2000 you have been more focused on leveraging the Internet to “fix” what you see as the problems with scholarly communication, than simply making money. So, for instance, BioMed Central was intended to solve the access problem. If that is right, what problems are you trying to fix with each of the F1000 products?

VT: As you say, F1000 is trying to tackle the many other major problems, beyond the access issue, with the way new research is shared and evaluated. There are 6 main areas we are trying to address.

The first is around the extensive and incomprehensible delays in the sharing of science (typically 6 months to a year, often longer before a submitted paper can be read).

Associated with this are the biases and many other ills caused by the traditional anonymous pre-publication peer review process.

The third issue is around the lack of access to the data underpinning most new research findings and making it available in a format that can be reused, reanalysed and reproduced.

And the fourth issue is around the bias in what is typically published, which is skewing our understanding of science to only those findings that are ‘positive’ and deemed as ‘significant’.

A major contributor and driver of all these problems is the existence of journals where the decision to accept or reject an article rests with an editor(s) and the related Impact Factor.

The F1000Research model is trying to address each of these issues through immediate publication, open and transparent invited peer review, a requirement to share all the underlying data (obviously with due consideration for data protection issues) and active encouragement of the publication of all findings including negative/null studies, small studies, case reports, replication studies etc.

Of course one of the biggest challenges with halting the inappropriate use of the Impact Factor is its current integral use in how researchers are assessed for grants and promotion/tenure decisions. Many funders and institutional review bodies are now moving away from metrics towards greater use of peer review, and F1000Prime is increasingly being used by funders and institutions to support this change.

Finally, the sixth area is around addressing the amount of time researchers have to spend on tasks associated with writing up their findings; F1000Workspace was built to streamline these activities through better technology, reducing the barriers to writing up and sharing of all research findings, and increasing the time researchers can spend in the lab conducting new research.

I have given a talk recently at Collège de France in Paris, titled “Nobody Knows” which summarises the issues we are trying to tackle with our new service. You can see the talk here, or read a summary in this editorial here.

RP: As you note, even in today’s digital world — and even in an open access environment — research is still primarily published using the traditional journal format, and researchers are still judged and evaluated on the basis of which journals their papers are published in. For their part, journals are judged on the basis of the Impact Factor — a metric you have described as “both problematic and idiotic”.

While F1000 has developed alternatives to these, the continuing obsession with the IF, and with the traditional journal, inevitably limits what you are able to achieve. What is striking to me is that everyone understands what the problems of scholarly publishing are today, but no one seems able or willing to fix them. Rightly or wrongly, I have come to think that reforming the scholarly communication system is as difficult as reforming the UK House of Lords, which people have been trying to do for over a hundred years now. What do you think might finally make the difference in terms of reforming scholarly publishing? Who has responsibility to ensure it happens, and how should they go about it?

VT: I agree with your description of the IF producing seemingly insurmountable problems. We are now working on challenging this directly and I believe we have a chance to succeed, though it is clear this is both the hardest, and the most important issue to solve. I think the responsibility lies with the institutions that fund and carry out research in biology and medicine. I will describe in my next answer how we plan to do this.

The Seer of Science Publishing

RP: The journal Science has described you as the “Seer of Science Publishing”, and many believe that if anyone can change the sclerotic system we have today it is you. I understand you are currently presenting a set of proposals to major funders and institutions, which (if you succeed) you expect to significantly change life science research publishing. Can you say something about these proposals, and your expectations of them?

VT: We believe that the main problem is the very existence of journals, and the methods their editors use to accept or reject what to publish. Their reason for making these choices is significantly affected by their battle to maintain and raise their Impact Factor.

We believe science today does not need journals (no one reads journals, everyone reads articles). We have developed a super-efficient service — F1000.com — which combines literature discovery (including recommendations), authoring tools (to write articles, collaborate, manage article libraries), and quick and effective publication using a high quality transparent refereeing process.

We want to convince funders and other major institutions that they have the responsibility to address the significant issues with publishing as it is at present and they now have the means to do so by using services like F1000.com (I am sure there will be others as soon as we show that they are effective). They can also operate their own publishing platform, convincing (and later mandating) the researchers they fund to publish through these platforms.

There are many advantages (speed, transparency, cost and more) to the funders, to researchers, to science and so to society in general.

This is clearly quite a task, but we have the tools to achieve it, our arguments are strong, we have a lot of support from many leading scientists, and the response to our presentation is surprisingly strong and positive.

RP: You are surely right to say that this is quite a task. It seems to me that the key challenge lies in persuading research institutions (rather than funders) to wean themselves off the impact factor, especially university promotion and tenure committees. For instance, while HEFCE has repeatedly said that the IF should not be used to assess the quality of research outputs, its injunction seems to fall on deaf ears in universities, presumably because using the IF makes the task of evaluating researchers so much easier. And while asking universities to ignore what journals a researcher publishes in follows naturally from this, I wonder if it is currently possible to persuade universities (and researchers themselves) to change their habits and mind-set, when doing so would make the process of evaluating research that much harder. Beyond providing the tools to enable them to change their ways, and making the case for doing so, is there anything further that can be done do you think?

 VT: I think that as long as the main way to publish research articles is in editorially controlled journals, the IF will be next to impossible to dislodge from the consideration by the committees giving jobs and grants and by the authors who depend on these decisions. Our hope is to convince funders and institutions to operate their own “publishing platforms” and encourage (and in time perhaps mandate) the research they fund to be published on these platforms.

If this happens (and I believe it can happen) then a significant proportion of new research will be published without consideration for the IF of the platform it is being published in. At this point I do not see any other way to effectively move away from the range of problems caused by the influence of the IF on publishing research in editorially controlled journals.

RP: You have suggested a couple of times that institutions and funders should consider operating their own publishing platforms. I wonder if you could expand on that. Would these platforms be in competition with F1000Research, or different in some way? Do you envisage them being built on top of institutional repositories, as UCL has started to do (but using the traditional journal/book format), or something more radical like the physics preprint server arXiv?

VT: F1000 sees itself as a service provider. One of the services we can offer to funders and institutions is to operate a publishing platform using the system we have developed for F1000Research. These platforms can be operated by us (of which F1000Research is an example) or by others who I hope will develop competing systems.

Our main asset is the ability to operate efficiently and to the real benefit of authors, readers and the scientific community.

Of the examples you mention, arXiv (which had a real influence on the development of OA from the start) is the closest, except that our platform incorporates full, open and transparent peer review, as well as qualitative assessment at the article level, together with the tools to help authors and referees to write, collaborate, manage references, and more.  

While this is a new idea and it is early days, I imagine that in the future many such platforms will be provided by funders and other institutions, working in parallel. They will not be competitive (except in the way institutions are competitive) and the authors will be free to choose a platform that is the most natural and efficient for them.

The qualitative assessment of researchers’ work will not depend on the Impact Factor of journals but on the article-level assessments and other ways of assessing quality of work provided by many different groups. The articles and other forms of presenting findings will be open to a refereeing process both by invited referees and by other members of the research community, and will of course be freely available to all.

RP: Another issue that has arisen is that of “credit inflation”. With papers tending nowadays to have a much larger number of authors — sometimes in their hundreds, sometimes even thousands — many argue that there is now a need to evaluate researchers for things other than the papers they author/co-author. Do you have views on this and/or any plans to offer services to help in this?

VT: I do think that the evaluation of both articles and researchers is complex, and it depends on many things, and will change depending on who is doing the evaluation, as well as when and why.

Our task is try to make the publishing of new research as effective and clear as we can. In this regard, article level evaluation is superior to IF derived evaluation, but is not the only solution, and there is a significant task ahead (and I think many opportunities for publishers and researchers) to make that process better and fairer.

New challenges, and criticisms

RP: You pointed out that F1000Research practises post-publication peer review. Some have criticised F1000 on the grounds that its approach to review tends to generate short, superficial reviews aimed at “approving” papers rather than “improving” them (which is one of the stated purposes of peer review). They also complain that since F1000Resarch has no equivalent to a deciding editor able to ensure that papers are corrected and improved after the review process, there is an inherent quality problem.

Indeed, Tim Vines has gone so far as to suggest that since the editorial process makes “‘approval’ the path of least resistance”, and since authors are charged a publishing fee, F1000Research “flirts with predatory OA status”. Should we treat views like this as no more than the expression of an understandable anxiety amongst incumbents fearful of the consequences of change, or does F1000’s post-publication review process introduce new challenges? If the latter, how does one address these challenges?

VT: You can look at the referee reports on F1000Research and see for yourself that they are detailed, critical, specific, and generally well written, largely due to the fact that the refereeing is fully transparent.

The referees are aware that their contribution is available for all to read, can be cited and criticised, and is written for authors to help to make the article better, rather than for an editor to decide which articles to reject.

Inappropriate approval, just as inappropriate rejection, can be openly criticised. The refereeing system in F1000Research works better than in any of the many journals I have published in the past. The largest group of articles published in F1000Research comes from some of the top research institutions.

RP: Phil Davis has suggested that papers posted on F1000Research that are not reviewed “will eventually be deleted from the system”. Is that correct? What percentage of papers do not attract reviews? Also, what happens if the authors of a paper do not address any of the issues raised during the review process, by amending their paper?

VT: Nothing will be deleted from the F1000Research site, but transparency of our process ensures that the reader can judge exactly everything that has happened to an article.

There is a very small number of articles for which the authors and we did not manage to get referees. There may be many reasons for this. We do not remove these or any other articles published on F1000Research.

If the authors decide they do not agree with some referee comments or decide they do not wish to revise their article, that is their choice. We have taken great care to ensure the reader is always very clear at what stage an article is at, view the latest referee comments, and see whether the article has been revised so they can judge for themselves accordingly.

You could say that our process is “author driven”, and authors decide how and when to respond to referees comments, knowing that everything will be visible and transparent to the reader.

RP: When you founded BioMed Central you formed a friendship with Harold Varmus. At the time he was director of the National Institutes of Health (NIH) and you helped encourage him to create PubMed Central (PMC). I believe you are also a friend of David Lipman, who is director of the National Center for Biotechnology Information (NCBI), and so responsible for PMC. In 2013, Kent Anderson — now publisher of Science — complained that when it asked to be included in PMC, F1000 was afforded “special treatment”. More recently, PubMed agreed to a new form of citation — the “dynamic citation” — in order to accommodate F1000’s post-publication review process. After making a number of FoI requests Anderson concluded that he had found evidence of what he called “cronyism and favouritism”, and that this demonstrated that the open access movement has “a core group of insiders running the main stage”.

Would you agree that open access has been driven by a small group of people (rather than by the research community at large), and that this has encouraged a degree of cronyism and favouritism, or should we view such criticism as the inevitable slings and arrows that any attempt at revolutionary change attracts?

VT: I have nothing sensible or interesting to say in response to this. OA is a broad movement and I am not involved in its politics.

RP: It seems to me that any changes made to the scholarly publishing process are inevitably designed around the needs of researchers in the developed world. As such, researchers in the developing world often face new obstacles and disadvantages. To what extent does F1000 consider the needs of researchers in the developing world when introducing new products and services? And how does it try to avoid disadvantaging them?

VT: I am not sure that “any changes made to the scholarly publishing process are inevitably designed around the needs of researchers in the developed world”. We try to help in the most obvious ways, like some other OA publishers, but in the end these are issues that need to be solved by international organisations and governments.

RP: Let me give you an example of how changes to scholarly publishing can lead to new problems for researchers in the developing world, even when those changes are intended to improve the situation for them. While the pay-to-publish model being offered by OA publishers may remove the accessibility problem for those in the developing world, in doing so it ends up discriminating against those without the necessary funds to pay to publish, who are disproportionately located in the developing world.

This means that while they will be free to read other scientists’ research, researchers in the developing world will struggle to publish their own work. (By the way, this is impacting even on researchers in wealthy US universities like UCLA). I understand some publishers offer waivers, but waivers can disappear overnight, and researchers tend to find them demeaning in any case — as Raghavendra Gadagkar pointed out in 2008. Do you have any regrets about having pioneered the APC model (a model you still use with F1000Research)?

VT: I do not have regrets (the benefits of OA are clear to all), but the problems you mention are real. They will need to be solved by a) significantly reducing the cost of publishing an article, and b) creating special funding for researchers with these types of problems (e.g. HINARI).

RP: How would you describe the current state of open access? What has been achieved? What still has to be done?

VT: I am not actively involved in the politics and the groups concerned with open access, though of course we would not publish any original research articles other than in a full open access way.

My sense is that open access has now established itself as being integral to the publishing industry, and in time will become the default way to publish research; however, it still has some way to go to succeed in fully reaching this goal.

All this is wrong but natural

RP: I understand your reluctance to wade into the often heated politics of open access. But you do make the point in the video you mention above that you were there at the very beginning of the open access movement, when you went to NIH and “proposed the idea that developed into open access”. Importantly, it was you (when you founded BMC) who pioneered the pay-to-publish business model that is fast becoming the dominant model for open access. As noted, this involves publishers levying a fee for publishing an article (i.e. pay-to-publish gold open access) in order to make it freely available on the Internet.

In an interview with ATG in 2002 you said, “Currently science publishers do not provide services that can remotely justify their charges”. You were referring to the traditional subscription system. At that time BMC’s APC was just $500 per paper (£322), and you said to ATG, “it will become cheaper as the costs involved go down until finally, I believe, it will be free.” Today, as traditional publishers transition their businesses to open access they are pricing their APCs at a level intended to preserve their historic revenues which, as you indicated, are generally exorbitant. As a result, many are now charging several thousand dollars per paper to publish open access.

Would you agree that in doing so they are continuing to charge fees that cannot remotely be justified? If so, can anything be done to bring these costs down, given that scholarly publishing does not appear to be subject to normal market forces?

VT: I agree that publishers are trying to maximise the income from the new Open Access way, and meeting surprisingly little price sensitivity from funders. The addition of extra charges to publish individual articles OA in subscription journal [hybrid OA] adds (I feel) a difficult to justify cost. All this is wrong, but natural. Publishers (like everyone else) will try to get as much as they can from the market.

Not all my expectations came true. The costs of publishing OA as it is done now are more than I expected, but in the scheme of things, it is not the costs (which are a small part of life science expenditure), but the method that journals operate that does most damage. The service we offer, because it removes the editorial choice step, can be much more efficient and is already less expensive, and will get less expensive with time, if it becomes broadly adopted.

In the end it is all in the hands of the payers (funders and institutions) to ensure that the process of communicating findings in science is effective, efficient and cost effective. OA is essential for allowing access and in that it is broadly successful.

RP: Do you think there might have been ways in which you, and the wider open access movement, might have prevented traditional publishers from appropriating open access in the way they appear to be doing? Do you see any strategic errors, or missed opportunities?

VT: The growth of OA in traditional publishing is a positive development. I very much hope in time all research publishing in life sciences will be OA and available to all without restrictions.

RP: Since you launched BMC in 2000 the open access movement has broadened into, and become a component part of, a wider open science movement. Would you agree, and if so how would you define this wider movement? What are its core components, and what is driving its development?

VT: I am not sure what different people mean by open science. We use this term to mean complete transparency across the publishing process: no delay in publishing, open and transparent post publication peer-review, and provision of relevant data with the published article in a form that can be practically used.

I do believe that journals in biology and medicine are responsible for many of the problems with the current way that research is published. I do not think science needs journals to make editorial decisions of what should and should not be published. I would like to see the end of journals and their replacement with services to authors, readers and institutions to make the communication of research findings more effective. That is what we hope to achieve with our new integrated service, F1000.

RP: F1000’s Community Strategy Manager Eva Amsen has suggested that, in addition to OA, open science encompasses open notebook science, citizen science, crowdfunding, open source software and open drug discovery. I guess one might wonder what role, if any, publishers have to play in these areas. However, when I spoke to you in 2005 you suggested that one possible future for publishers would be to create value-added databases.

In the event, I think it fair to say that with F1000 you have adopted a model more like the one you created in 1990 with the Current Opinion series of journals in medicine (subsequently sold to Thomson), rather than taking the value-added database route? Have your views on the future role of scholarly publishers changed in any way since 2005? If so, in what way?

VT: You can see F1000Prime as a value added database of recommendations by experts of the most interesting articles with the reasons for their choices.

I see the future for scholarly publishers in the world of open science as being service providers to the broader research community of researchers, institutions and funders. It is likely that the forms of communicating new findings will go beyond the traditional narrative article, and we publishers will need to respond quickly and effectively to these changes.

RP: Some OA advocates have come to believe that commercial publishers have no useful role to play in scholarly publishing. It is time, they say, for the research community to “take back ownership” of the process by which they share their research. I assume you do not agree with that, so can you say what you feel for-profit organisations like F1000 offer that the research community cannot provide for itself?

VT: We must offer services that the science community needs and cannot do itself. Anything the scientists can do themselves is desirable. We are changing from “publishers” to “service providers”, and our services will only be used when they are needed.

I do not see at this point an appetite by scientists to take over these tasks, which are different from doing research, require investment and expertise, and so morph into for-profit businesses. The example of PLoS is instructive — it was started by scientists, and it is now a very profitable business and (even though called “not for profit”) is in fact working hard to maintain profitability, and I think this is natural.

RP: Let me turn my last question around: While librarians have been the most vociferous supporters of open access they have proved strangely reluctant to practice what they preach. I wonder if perhaps they failed to think through the implications of open access. Essentially, they advocated for OA because they believed it would provide a solution to the serials crisis. But as I think we have agreed, the affordability problem has not gone away with open access. Perhaps there are other surprises in store for librarians? They will tell you that their role in the future will be to act as guides to published research, and organisers of this knowledge in discovery platforms of some sort. Others, however, argue that it will be entrepreneurs or established firms (EBSCO, ProQuest, OCLC etc.) that will provide the guidance that librarians expect to offer, and that this will free universities from the need to pay the salaries of librarians at every institution. After all, it could be argued that a publisher-provided subscription service offering a guide and filter for students or researchers would be a very cost-effective and responsive approach. It’s obviously not clear that such a service would actually work, but it is a possible future.

What are your views on this? Where, for instance, do you see academic libraries in ten years? Do librarians need to start viewing publishers as competitors rather than partners — which I think is the view of Lorcan Dempsey?

VT: I have no opinion on this, except to say that I am an admirer of most of the librarians I know, and of the profession in general. They have usually a clear grasp of what is happening and what is needed.

Politics and money

RP: Earlier, you said that you don’t get actively involved in the politics of open access. Nevertheless, you have acquired cult status within the OA movement. Traditional publishers also have (sometimes grudging) respect for you and what you have achieved. I suspect your reputation has been enhanced by what might appear to be an unworldly attitude on your part, along with what your former employee Daryl Rainer described as your “unusual management style”. When she asked you about the latter in the 2002 ATG interview I cited earlier you replied “I never look at spreadsheets and wouldn’t know how to interpret a balance sheet”. And speaking to me three years later you said that you were not particularly interested in money. As you put it, “I am perfectly happy not to have money, although it is nice to have it. What I enjoy is working with this group of people, and inventing new things; things that were not there before; and things that are both difficult to do and complicated.” Yet you have clearly made a lot of money from publishing.

Why do you appear unwilling to talk about the financial details? For instance, when Springer bought BioMed Central from you in 2008 the purchase price was not disclosed. Why was that? And are you able to say today how much Springer paid you for BMC?

VT: Nothing has changed in this respect since our last conversations.

RP: In my final question I want to return to the issue of politics. A couple of times you have said that you don’t get involved with the politics of open access. By that I guess you mean that you don’t engage with the many debates, discussions and factional groups within the open access movement. Nevertheless, you have been involved in the politics of open access in the more direct sense have you not?

For instance, you presumably endorsed (if you were not actively involved in) the approach that Jan Velterop (then Publisher of BioMed Central, a company that at the time you owned) made to the then Chairman of the UK House of Commons Science & Technology Committee Ian Gibson. As I understand it, Velterop had lunch with Gibson in order to persuade him to consider launching an enquiry into scientific publishing. This led to a dinner with David King (then Chief Scientific Adviser to the UK Government) and other interested parties in the House of Commons, which in turn led to the 2004 Inquiry into Scientific Publishing (to which you and Harold Varmus gave evidence). The subsequent report — Scientific Publications — Free for All? — is viewed as having been key to the success of open access. Is that not politics? And where you not involved in it?

VT: I was deeply involved in promoting OA in the early days when it was an essential and difficult task. But OA has established itself, and for many years now, while making sure anything I do is fully OA, I am not involved in the politics of it. 

RP: Ok, thank you for taking the time to consider my questions. Good luck in your future endeavours.