Suber: Leader of a Leaderless Revolution
by Richard Poynder
[A shorter version of this interview is available in the July/August print issue of Information Today.]
What is remarkable about the open access (OA) movement is that despite having no formal structure, no official organization, and no appointed leader, it has (in the teeth of opposition from incumbent publishers) triggered a radical transformation in a publishing system that had changed little in 350 years. Most notably, it has demonstrated that it is no longer rational, or even necessary, for subscription paywalls to be built between researchers and research.
While many have played important roles in the movement, no one has been as influential, or as effective, as philosopher, jurist, and one-time stand-up comic Peter Suber, a man now viewed as the de facto leader of this leaderless revolution.
Among Suber’s unique contributions is that he played midwife at the birth of the movement. Specifically, he was present at the seminal 2001 Budapest Open Access Initiative meeting (convened by the Soros Foundation) in Hungary, where OA and its agenda were first defined. Suber also drafted the associated ( Budapest) Manifesto, doing so in a way that successfully fused the different agendas articulated at Budapest into a coherent and convincing whole, while serving to inspire and stir OA advocates.
Some argue that without Suber’s presence, the Budapest meeting could have splintered into warring factions. Likewise, without his strategic vision and frequent arbitration of disputes, an inherently fissiparous movement might have torn itself asunder many times since.
“Peter’s gentle, fair, ecumenical approach has proved far more persuasive than some of the more impatient approaches (including mine!), though, as he himself has pointed out, he can be quite firm and forceful too, if need be,” says fellow OA advocate Stevan Harnad.
Suber actually became the glue that holds the movement together and its chief strategist. His main platforms for doing this have been his blog, Open Access News (OAN), and the SPARC [the Scholarly Publishing and Academic Resources Coalition] Open Access Newsletter, where he has repeatedly and convincingly made the case for OA, deconstructed and rebutted the many arguments deployed against it, and consistently inspired the movement.
“Peter’s contributions in practical terms—OA News and the [SPARC Open Access] Newsletter—have been absolutely key in centering efforts and debate but I would say that even more important have been his unique powers of logic and argument,” says OA advocate and publishing consultant Alma Swan. “He puts the case in ways that are persuasive and difficult to refute: he is inimitable in that respect.”
Even publishers respect Suber. “I have never met Peter, but I have admired his clarity of expression and analysis, and I respect his stamina and dedication to the open access movement,” says Graham Taylor, director of educational, academic, and professional publishing at The Publishers Association. “It seems to me that his SPARC newsletter has for some time been the flagship on which the rest of the OA fleet relies for its sense of direction and rigour.”
Second, Suber has proved to be a highly effective behind-the-scenes advocate. “Peter has been a key contributor to the development of open access policies on both the institutional and national level,” says Heather Joseph, executive director of SPARC. “He has been widely respected for his thoughtful analysis, careful word-smithing and good counsel, so it will come as no surprise that his advice is highly sought after by policymakers, not only here in the U.S., but around the world as well.”
“It is not that OA would not have happened without Peter Suber: It is optimal and inevitable. But his leadership made it far more likely to happen sooner rather than later,” says Harnad.
Certainly no other OA advocate has made a greater sacrifice than Suber. In 2003, he put a successful academic career on hold to advocate for OA, a decision that meant giving up a tenured position for a series of uncertain, short-term grants.
But the recognition and rewards are just beginning: In 2009, Suber received a joint fellowship from the Berkman Center for Internet & Society with the Harvard Office for Scholarly Communication and Harvard Law School Library. And earlier this year, the American Library Association (ALA) announced that it had selected Suber as the 2011 winner of the L. Ray Patterson Copyright Award for his work on OA. Suber is also a senior researcher at SPARC and a member of the Board of Enabling Open Scholarship. He also serves as OA project director at Public Knowledge. Here then are some of his insights into OA and his role in this game-changing movement (the following interview has been edited for style):
Q: How would you describe your current role within the OA movement and the various hats you wear?
A: I try to stay on top of what’s happening with OA and what has happened. I try to digest or analyze it, make policy and strategy recommendations based on my analysis, and work with individual OA initiatives to move things forward.
I spend a good deal of time on direct assistance, or pro bono consulting, talking to universities and funders developing OA policies, publishers thinking about OA for journals or books, organizations thinking about launching OA repositories, conference organizers with sessions on OA, and individual researchers and journalists writing about OA.
I also still write the SPARC Open Access Newsletter (SOAN) and publish one issue a month.
At the moment, I’m also finishing up two books for MIT Press, one is a collection of some of my articles on OA, and the other is a concise introduction to OA.
Q: You also used to publish an OA blog didn’t you?
A: Yes, Open Access News . But I curtailed my blogging in mid-2009 to make time for my new work at the Berkman Center, and I laid down the blog altogether in the spring of 2010.
I still run a daily private crawl of the internet for OA-related news, just as I did in the days of the blog. But instead of sharing the results through the blog, I share them through the OA Tracking Project , the real-time alert service and social-tagging system that I launched in April 2009, and the Open Access Directory , the wiki I launched with Robin Peek [who writes the Focus in Publishing column in IT] in April 2008.
Q: It is 4 years since we last spoke . In broad terms, what’s changed for OA since 2007?
A: For the long answer, see my year-end reviews of OA in 2007 , 2008 , 2009 , and 2010 that I wrote for my newsletter. But the short answer is quite short: We’ve made progress every year on all four important fronts: more OA journals, more OA repositories, more OA policies, and more understanding of OA among researchers and policymakers.
We still have a long way to go on all four fronts, but our momentum has increased every year for a decade, and I don’t see that changing any time soon.
Q: Can you explain the reason why we need OA?
A: Authors need OA to reach all the readers who could build on their work, apply it, extend it, cite it, or make use of it. Readers need OA to find and retrieve everything they need to read and to allow their software prosthetics to process everything they need to process.
OA doesn’t merely share knowledge. It accelerates research by helping authors and readers find one another. It’s compatible with intermediaries but not with intermediaries who erect access barriers to keep authors and readers apart.
Basically we need it to solve a serious access problem hindering researchers both as authors and as readers. But we also need it to seize beautiful opportunities offered by the internet, especially the opportunity to distribute perfect copies of arbitrary files to a worldwide audience at zero marginal cost.
I understand why most novelists, musicians, and movie-makers don’t seize that opportunity. But scholars are not paid for their journal articles and can consent to OA without losing revenue. Any group that can seize this opportunity without losing revenue would be foolish not to do so. Authors who can’t, unfortunately, are in the hard spot of betting their livelihood against the internet.
Q: I guess the nub of the matter is that subscription publishers generally insist—as a condition of publication—that authors assign copyright (and thus, in effect, ownership) in their papers to the publisher. This allows publishers to sell bundled papers (i.e., journals) back to the research community at more or less whatever price they wish. As a result, journal subscriptions rise inexorably, while library budgets are generally static or falling. Since libraries are able to afford fewer and fewer journals a greater and greater access barrier arises between researchers and the research they need to do their jobs. OA removes that access barrier.
A: At least this is the nub of the serials pricing crisis. But that’s only one of the problems for which OA is the solution. Even if prices were low, we’d want to take full advantage of the internet for sharing research more widely and making it more useful.
Green and Gold?
Q: There are two main types of OA: There is green OA (OA through repositories) and there is gold OA (OA through journals). How would you characterize the current situation so far as green OA is concerned?
A: To me, there’s no single finish line for OA. But we’re rapidly approaching the point where green OA is the default for new research articles, even if it coexists with TA [toll access, or subscription access] from conventional journals. Green OA is already the default for physics worldwide, and for medicine in North America, for different reasons.
The pace of progress varies from field to field and country to country, but the direction of progress is the same everywhere. The curve is up everywhere. And the reason is the same nearly everywhere, namely, the rise of strong green OA policies at funding agencies and universities. Some policies are weak and have little effect, but the spread of strong policies is definitely enlarging the worldwide volume of OA research.
Q: It is now widely accepted that green OA policies don’t work unless they are mandated. Earlier this year when I spoke to the Sanford Bernstein financial analyst Claudio Aspesi (who writes reports on Elsevier for investors and so studies OA), he was less upbeat than you. He said , “[O]n mandates we have seen little progress. There have been several new mandates from both public and private institutions, but there is such a patchwork of mandates (and little transparency on the compliance) that I doubt there is a single research library that has changed its buying patterns to reflect the new or the cumulated mandates.” Are you seeing something Aspesi is not? And do you think he is right to say that library buying habits are not changing as a result of OA?
A: I liked much of Aspesi’s take in your interview. But on this point he’s using the wrong criterion of success. The immediate purpose of a green OA mandate is not to save libraries money but to make more research available to more people. Judged by the right criterion, OA mandates are very successful. They invariably increase deposits in OA repositories as well as the rate of deposits.
Rising levels of green OA don’t immediately or directly change library buying patterns. For that, we must look to the cancellation or conversion of TA journals. Green OA may eventually cause the cancellation or conversion of TA journals. But we should keep several things in mind as we watch for that effect.
First, as I said, it’s not the immediate purpose of green OA. For some people, it’s a secondary purpose of green OA, but for others, it’s not even a secondary purpose. Second, green OA hasn’t yet caused the cancellation of TA journals, even in physics, which has the longest history and highest levels of green OA. Third, if green OA eventually has this effect, it will be long-term and indirect. Green OA mandates can be very successful in growing the volume of OA literature long before they’re successful in saving libraries money.
Q: What about gold OA (OA through journals)?
A: We’re making progress here as well. The number of peer-reviewed OA journals is growing fast, both among new journals and established journals converting from TA to OA. There are more OA journals making profits or surpluses. There are more OA journals earning reputations for high quality, high impact, and high prestige.
There are more experiments with different business models for OA journals in different niches and more recognition that there are many different business models to experiment with. There are more universities and funding agencies willing to pay publication fees at fee-based OA journals—and significantly, these numbers are growing even in a deep recession.
There’s more recognition that supporting OA journals is an investment in a superior way to support research, researchers, research institutions, and peer review. More OA journals are documenting that their conversion to OA increased their submissions and citation impact.
The growing number of conversions from TA to OA suggests to me that small and medium-sized publishers are starting to see OA less as a threat and more as a survival strategy. The big deals are soaking up library budgets, library budgets are flat or declining, and journals excluded from the big deals have little future under the subscription model.
In fact, I see a growing recognition that the subscription model itself is unsustainable in a world in which the volume of published knowledge grows rapidly, and subscription prices grow faster than library budgets and inflation.
I see more funding agencies and governments—not just libraries and universities coming to the conclusion that price barriers to this critically useful literature create harmful access gaps and undermine the public investment in research and peer review. These institutions have long been committed to green OA, but they’re increasing their commitment to gold OA as well.
Q: How would you describe the relative advantages and disadvantage of green and gold OA so far as research funders, librarians, and scientists are concerned?
A: The advantages of green OA are that it’s faster and less expensive than gold OA. Green OA can be mandated without infringing academic freedom and gold OA cannot. A green OA policy at a university can cover the institution’s entire research output, while a gold OA policy can only cover the new articles that faculty members are willing to submit to OA journals.
The advantages of gold OA are that it can always be immediate rather than embargoed and can always be libre rather than gratis . It generally takes advantage of the first opportunity and is rarely embargoed. But unfortunately, it seldom takes advantage of the second, and today about 80% of OA journals are merely gratis rather than libre.
I should add that green OA can be immediate and libre as well, but it’s much easier for journals than repositories to obtain the permissions needed for immediate OA and libre OA.
Finally, gold OA performs its own peer review, without depending on TA journals to perform it, and gold OA can be self-sustaining, even profitable.
Gratis and libre
Q: As your answer indicates, there is more to OA than green and gold alone; there is also gratis and libre OA. In 2008, you produced a grid demonstrating the four-way relationship among the different types of OA. Can you expand on this, and outline the relative merits of gratis and libre OA?
A: Gratis OA is simply free of charge. But it’s not more free than that. Gratis literature may stand under all-rights-reserved copyrights and give users no more rights than they already had under fair use (or fair dealing).
Libre OA is free of charge and free of at least some copyright and licensing restrictions. Libre literature stands under some-rights-reserved copyrights, at most, and permits uses that exceed fair use. The advantage of libre OA is that researchers needn’t slow down to ask permission for legitimate scholarly uses that exceed fair use, needn’t take the risk of proceeding without permission, and needn’t err on the side of non-use. By the way, the grid you mentioned was merely a preview of a longer article, which explained the gratis/libre distinction in much more detail.
Q: Some OA advocates maintain that libre OA is the ideal form of OA. Would you agree?
A: The Budapest, Bethesda, and Berlin statements all call for libre OA, and I strongly support that call. But before I say that “the ideal form of OA” is libre, I’d make two qualifications.
First, gratis OA is attainable in many circumstances when libre OA is not. For example, most university and funder OA mandates are gratis policies, not libre policies. Pushing harder won’t change that, although I believe that time and momentum will. In those cases, we should celebrate gratis as a hard-won victory and major step forward. At the same time, we should continue to press for libre.
Second, libre is not just one thing. Because there are many ways to exceed fair use, there are many kinds of libre OA. I support the least restrictive Creative Commons license (CC-BY) for texts and I support the public domain (CC0) for data.
Q: We should also mention Hybrid OA , in which publishers of subscription journals agree to make individual papers OA on payment of an article processing charge. Is Hybrid OA an adequate solution in your view?
A: No, it isn’t. I summarized my reasons most recently in the March 2011 issue of my newsletter . “[H]ybrid OA journals do little or nothing to help researchers, libraries, or publishers. The recent, very comprehensive Study of Open Access Publishing (SOAP) showed that the average rate of author uptake at hybrid journals is just 2%. ... The chief virtue of hybrid OA journals is that they give publishers some first-hand experience with the economics and logistics of OA publishing. But the economics are artificial, since hybrid OA publishers have no incentive to increase author uptake and make the models succeed. The publishers always have subscriptions to fall back upon.”
Q: Your early interest in OA assumed a somewhat broader agenda than scholarly journals alone I think. You were initially advocating what you used to call Free Online Scholarship. Since the Budapest Open Access Initiative, however, I think you have been focused mainly on journal articles. Do you still think all forms of scholarship should be OA?
A: All research would be more useful if it were OA. But I’m an incrementalist. I’ll take what we can get when we can get it. I don’t expect OA to classified military research, and I don’t even argue for it. I don’t expect OA to patentable discoveries until after the discoverers voluntarily decide to publish. I don’t expect OA to copyrighted books except when OA would increase net sales or bring benefits that exceed royalties.
So there are many kinds of high-hanging fruit. But there are two kinds of low-hanging fruit: journal articles and publicly funded research, and of course, these two categories overlap significantly. Journal articles are low-hanging fruit because journals don’t pay authors for their articles, freeing authors to consent to OA without losing revenue. Publicly funded research is low-hanging fruit because public funds should be spent in the public interest, not to lock up valuable research where taxpayers must pay a second fee to read it.
Privately funded research is just as low-hanging, since private funders are charities whose motives are just as public-spirited as those of the public funders, namely, to fund useful research and make the results available to everyone who can make use of it.
By the way, when I say that journal articles and publicly funded research are low-hanging fruit, I mean relatively and not absolutely. They’re easier than royalty-producing novels, music, and movies. But they’re still out of reach in many niches and very hard to pluck. This is what I meant earlier when I said that many gratis OA policies are hard-won victories. They require long and patient advocacy, sometimes against well-funded opposition, and the fact that the target is royalty-free literature or publicly funded research doesn’t sweep away the difficulties.
I’m also very conscious of some difficult cross-breeds, such as publicly funded research resulting in monographs rather than journal articles or work in the public domain that hasn’t yet been digitized.
Q: Would you say that traditional publishers accept that OA (in some form or another) is now inevitable, maybe even that it is desirable?
A: I’d say that all publishers who follow OA progress now accept that it is here to stay and will keep growing. Many or most will say that it’s desirable for researchers. That’s not really in doubt. Fewer will say that that it’s desirable for publishers. Many publishers still fear it as a threat to their revenue. But as I mentioned, a growing number see it as a survival strategy, and a growing number see it as a way to increase citations, submissions, and quality.
In the U.S. and Europe, there’s still an aggressive publishing lobby working hard to block or weaken government OA policies. But it’s not as representative as it likes to depict itself. Despite this lobby’s fierce opposition to the NIH [National Institutes of Health] policy, all surveyed publishers accommodate NIH-funded authors. Moreover, many more publishers voluntarily do more than the policy requires than join the lobby against it.
Most publishers who oppose OA mandates permit green OA and experiment with gold OA. Five years ago, I’d have said that about half the TA publisher experiments with gold OA were in good faith and half were cynical. But the ratio of good-faith experiments to bad has grown steadily. I see some serious acceptance of the inevitability of green and gold OA, even if there’s also a serious desire to delay the inevitable as long as possible.
Don’t take my word that TA publishers are starting to accept OA. Last fall, Andrew Richardson, vice president for business development at Wolters Kluwer Health Medical Research, said in an interview that the publishing industry previously saw OA as a threat but now tends to see it as an opportunity.
In March 2011, the Publishers Association and Association of Learned and Professional Society Publishers—neither known as particularly warm to OA—sponsored a conference in London. Their publicity for the meeting started with the assertion that “Open access is here to stay, and has the support of our key partners. ...” They asked whether publishers could “respond more pro-actively and positively to open access” and whether publishers could “learn not just to live with open access, but to love it as well.” That was a remarkable sign of acceptance, one that I commented on in a recent issue of SOAN.
Q: You say that publishers who follow OA now accept that it will keep growing. No traditional publisher has embraced OA as wholeheartedly as Springer. Yet when I interviewed Springer CEO Derk Haank at the end of last year, he predicted that OA publishing will never be more than a niche activity. As he put it, “I expect it to remain between 5% and 10% at a maximum.” I assume you do not agree?
A: I know that Derk Haank understands the difference between green and gold OA. But the remark you quote is about gold OA alone. One way to supplement it is to point out the progress with green, which is well above 5% to 10% even in the absence of green OA mandates. On gold OA, there are many reasons to think that growth will continue.
For example, since you spoke with Haank, the Sponsoring Consortium for Open Access Publishing in Particle Physics ( SCOAP3) entered its implementation phase, which is huge. It has enough budget pledges to start the process of converting all the major TA journals in physics to OA, with the consent and cooperation of their publishers. If it works in physics, the model will surely spread to other fields. If Haank was pessimistic about SCOAP3, he should recalibrate.
I’d also add that even in a budget crisis, more and more universities are creating funds to pay publication fees at fee-based OA journals, and more and more funders who support green OA are thinking about ways to extend their support to gold OA. The recent report from JISC [Joint Information Systems Committee], PRC [Publishing Research Consortium], RIN [Research Information Network], RLUK [Research Libraries of the U.K.], and the Wellcome Trust endorses green and gold but favors gold, assuming publishing fees can be kept in check, as the most sustainable long-term solution.
Q: I think you accept that more prestigious journals such as Nature and Science will likely never migrate to an OA author-side payment model, since they would need to charge a prohibitive publication fee. And you will recall that in 2004 Richard Charkin, then CEO of the Nature Macmillan Publishing, told the UK Science & Technology Select Committee that it costs between $10,000 and $30,000 to publish a paper in Nature—a function, he explained, of the fact that the journal rejects 90% of the papers submitted to it. Do you have any views on the percentage of journals likely to remain subscription-based?
A: High-prestige journals should be able to charge subscriptions almost as long as they like. They’ll always be must-have titles for libraries. Or at least they’ll be must-have titles “during good behavior,” as the U.S. Constitution says of federal judges appointed for life.
The qualification is necessary. Last year, for example, Nature tried to raise the price of the University of California (UC)’s site license by 400% and may have thrown away its must-have status. UC not only threatened to cancel its NPG titles, but it was ready to encourage other institutions to do so as well . I say “may have” because we still don’t know the outcome of their negotiations. The point is that even a must-have journal with Nature’s cachet can become a must-drop journal with behavior like that.
By the way, it’s true that fee-based OA journals with high rejection rates must charge high fees. But in January, long after Charkin testified in 2004, Nature launched Scientific Reports , in imitation of PLoS ONE. It’s a full OA journal with a deliberately low rejection rate. Nature didn't let the economics Charkin described stop it from entering the world of gold OA, with the goal of publishing at the highest levels of quality. Neither will other publishers.
Q: When I spoke to Haank, he told me that Springer is one of the publishers that lobbies against green mandates. Explaining why, he said: “I draw a distinction between author archiving and mandatory open access requirements like the NIH Public Access Policy where they don’t allow for a sufficient embargo. OA mandates institutionalise the process of author archiving; and if the delay between publication and archiving is only a couple of months then there is a real danger of destroying the equilibrium that we have achieved over open access.” I assume his point is that unless there is a sufficient embargo period before a paper is made, OA mandates could significantly erode publishers’ profits and possibly put them out of business. Publishers undoubtedly remain important players in the publication process. Are you sympathetic to their anxieties about mandates?
A: Yes and no. I’ve long argued that publishers should accept OA mandates for publicly funded research and focus their concern on the length of the embargo. I’ve also conceded that allowing embargoes can be an acceptable compromise. In that sense, I’m sympathetic. But there are a couple of senses in which I’m not.
For example, I can’t tell whether Springer makes clear in its lobbying that it accepts the legitimacy of mandates and only opposes short embargoes. If so, then its campaign is appropriately surgical, and I welcome a public debate about the best length of the permissible embargo.
But if Springer actually lobbies against OA mandates for publicly funded research and not just against short embargoes, then it’s saying that public agencies should put the private interests of publishers ahead of the public interest in research. It’s saying that the public should compromise and publishers should not compromise. It’s saying that the public investment in research should create private property for corporations for an indefinitely long time.
Those who lobby against government green OA policies seem to believe that the prosperity of incumbent publishers is more important than the barrier-free circulation of publicly funded research. That’s an understandable position for an incumbent publisher. But it’s bad public policy, and publishers can’t expect that public agencies charged to act in the public interest will share their parochial position.
The fact that governments are amenable to lobbying, and frequently depart from the public interest, doesn’t justify any particular lobbying position and doesn’t justify any departures from the public interest. On the contrary, it’s a reason to reassert the public interest and save it from the money game that our politics has become.
Q: What is your message to those publishers still holding out against OA, especially those lobbying against self-archiving mandates?
A: I’d argue that they should accept the legitimacy of mandating OA for publicly funded research and focus their objections on the length of the permissible embargo. If they don’t, they’re putting their private interest ahead of the public interest and demanding that public agencies do the same.
If they don’t object to green OA mandates for publicly funded research and merely resist the idea of converting to gold OA themselves, then I have no objection. But I’d urge them to allow author-initiated green OA. I’d urge them to study the gold OA publishers who are paying their bills and making profits and study the real sustainability of the subscription business model in a world of rapidly growing research and flat or shrinking library budgets. But the choice is theirs, and I’ve never wanted government policy to go beyond regulating grantees to regulating publishers.
For me, the pitch to subscription-based publishers is fourfold. First, permit green OA. Second, study gold OA. Third, don’t stand in the way as public agencies act in the public interest. Fourth, don’t stand in the way as researchers and research institutions act in the interests of research.
Q: You pointed out earlier that the Big Deal is soaking up library budgets. Haank described the Big Deal to me as “the best thing since sliced bread.” He added: “The truth is that it is in the interests of everyone—publishers and librarians—to keep the Big Deal going.” And according to the U.K. Publishers Association, a recent study shows that (thanks to the Big Deal) 93% of researchers worldwide are satisfied with their current levels of access. I suspect you are a little sceptical about such claims?
A: Very. On the plus side, big deals give universities access to more titles than they had before and reduce the average cost per title. On the minus side, most bundles include titles that are low in quality, low in local usage, or both, titles which the institution would certainly cancel to save money if it were free to do so. But canceling titles only raises the price of the remaining titles, which is advantageous only to the publisher and harmful to libraries and users. Universities do want access to more titles, but they want to pick the titles most needed by their own patrons. They do want publishers to reduce the average cost per title, but they want to see it done through price cuts, not bundling.
Bundling may be the greatest thing since sliced bread for publishers. It creates a layer of artificial monopoly laid on top of the natural monopoly every journal already has by virtue of not duplicating other journals. It protects second-rate and under-used titles from cancellation. Big deals are too big to cancel, by design, at least without extraordinary pain, which gives publishers leverage to raise prices out of proportion to costs, size, impact, and quality. Big deals soak up library budgets, which hurts smaller publishers excluded from the bundles. In turn, this hurts research because in general journals from those smaller society publishers tend to be higher in quality and impact than journals in the big deals.
If you were hungry and had an allowance for one meal—roughly the situation of a university with its annual library budget—it wouldn’t help to be offered a smorgasbord of good and bad food mixed together for the price of four meals. You might scrounge to pay for the deal. You might even marvel at the size of the spread. But if you could bargain on equal terms with your food vendor, you’d pay one-fourth the price and get exactly the meal you wanted. Or having scrounged, you’d pay half the price and get twice the meal you wanted. At the same time, despite financial pressure, you’d put money aside to support an alternative system that took into account that you and other customers like you grew all the food, did most of the cooking, and gave it your vendor free of charge.
It’s true that libraries once welcomed big deals and even called for them. But now I think most libraries agree that big deals have backfired on them. Publishers like to speculate about the harmful unintended consequences of OA. It’s a healthy exercise, and I do it myself. But we could make a long list of the harmful unintended consequences of big deals, already realized. We could even make a second list of the harmful consequences that are no longer unintended.
Q: The most successful OA journal today is PLoS ONE. Indeed, OA advocates claim that it is now the largest journal in the world—having published more than 19,000 papers in the 4-plus years since it was launched. But PLoS ONE is not a typical journal as it has a number of very novel features. First, it is not subject specific but publishes papers in all disciplines within science and medicine. Second, papers are assessed only for technical soundness before publication. It is this recipe that has made PLoS ONE so successful, even though it charges a $1,350 publication fee. As a consequence, traditional publishers are now rushing to emulate it. As you pointed out, Nature has launched its Scientific Reports , and there is also Sage Open and the American Institute of Physics’ AIP Advances . Some now predict that PLoS ONE’s model will become the dominant one for scholarly journals. OA advocate Cameron Neylon, for instance, predicted that in future “most scholarly publishing will be in publication venues that place no value on a subjective assessment of ‘importance’.” Do you agree? Is PLoS ONE a picture of the future? What do you think the implications would be if its model became the primary way of publishing research papers?
A: I wrote about this in the April 2011 issue of my newsletter. Basically, yes, I agree. PLoS ONE is a significant model for a peer-reviewed journal, even against the background of other peer-reviewed OA journals. The rapid growth of PLoS ONE imitators is a significant development in journal publishing overall, especially in light of the fact that some of the publishers with PLoS ONE clones were formerly harsh critics of the PLoS ONE model.
In any science where there is some professional consensus on methodology, it makes sense to separate methodological soundness from significance and to focus prepublication review on soundness. To try to cover significance as well as soundness increases the time and cost of peer review and introduces a subjective element into editorial judgment.
Significance is better judged by the entire community in open discussion after publication than by a few referees in private before publication. The stakes are higher than they might appear, since the longer prepublication review takes, the longer we must wait for the peer-reviewed article to become OA.
The model will spread because it’s less expensive than traditional peer review. It may even have originated because it’s less expensive than traditional peer review. I appreciate the need to save money, especially at OA journals. But my own view is that the cost reduction is a weaker argument in its favor than its speed and focus on questions amenable to scientific judgment.
As with any kind of peer review, it can be done badly. But I don’t worry that it means the end of rigor. We can judge methodological soundness with more rigor than we can judge significance, at least in those fields where there is professional consensus on method. The idea that postponing judgments of significance will reduce rigor is nuts. Where we find lapses of rigor, therefore, we can’t attribute them to the model itself but only to its implementation.
I agree with Stuart Shieber that the PLoS ONE model is bringing some high-prestige publishers to the world of gold OA, which will increase the number of high-prestige OA titles. I also agree that the number of new PLoS ONE clones will increase competition for authors, which will tend to improve terms, for example, with lower publication fees and less restrictive licenses.
I also agree with Phil Davis that if PLoS continues to grant fee waivers no-questions-asked, and if the new PLoS ONE clones don’t, then PLoS could see a steady rise in the number of indigent authors, subtracting any savings it might currently realize from the model. I don’t see a good solution to this problem, except to make the case that all fee-based OA journals, including the new clones, should offer fee waivers in cases of economic hardship. But I don’t expect that argument to carry much weight with publishers who want to maximize profits and minimize the financial stability of a rival.
Q: As your answer implies, PLoS ONE’s novel review process (assessing papers for technical soundness alone) was posited on the assumption that published papers would also undergo open peer review after publication. The consensus is that in the vast majority of cases this post-publication review is not taking place. Does that matter?
A: First I agree with the consensus. Post-publication review doesn’t yet work as well as earlier boosters hoped, including me. It matters because effective post-publication review would allow OA journals to cut the delay of pre-publication review without giving up the benefits of peer review. Today the only way to do that is through preprint archiving. But not all OA preprints are undergoing peer review, even if they’re all en route to peer review.
I want to see post-publication review work. Lots of smart and motivated people have worked on the problem, and I don’t pretend to have a better fix on it. My only suggestion is for OA journals to assign reviews, as they do now for post-publication review, and not to depend on drive-by reviews, which tend to be brief and superficial.
Q: You earlier cited Wolters Kluwer’s Andrew Richardson saying that OA offered publishers an opportunity. At an event held in Oxford last year, another Wolters Kluwer executive said something similar, but then added that she viewed OA journals as a venue for low-quality research. Perhaps this is not surprising. It has long been assumed by some that OA journals will inevitably be of a lower-quality than their subscription competitors. There are, as you indicated earlier, OA journals about which that certainly cannot be said. Nevertheless, there remains a belief that OA implies lower quality. Fueling this belief is the fact that there are indeed some very inferior OA journals, and the number appears to be growing. Stevan Harnad talks of the “Dot.Gold Rush for Open Access.” As he put it in 2008: “There seems to be a growing epidemic of fast Gold-OA journal-fleet start-ups, based on next to no scholarly/scientific or publishing experience or expertise, and relying heavily on online spamming … Some of them do minimal, low-quality peer review, and some of them perhaps next to none.” There are undeniably a great many well-meaning and professional OA publishers, but doesn’t the flood of less-scrupulous OA publishers indicate that the author-side payment model is inherently problematic? Does it not pose the kind of conflict-of-interest issues Elsevier warned about in 2004 when giving evidence to the U.K. Science & Technology Select Committee?
A: There are two questions here. Are OA journals intrinsically lower in quality than TA journals? And are OA journals intrinsically prone to dishonest behavior? The answer in both cases in unequivocally no.
If the argument is that some OA journals of very low in quality, that’s true. But some TA journals are very low in quality as well. This shows nothing about what’s intrinsic to OA. If the argument is that some OA journals are scams, that’s true. But some TA journals are scams, such as Elsevier’s fake journals publishing Merck PR disguised as peer-reviewed journal articles. This shows nothing about what’s intrinsic to OA.
If the argument is that many OA journals charge author-side fees, which are supposed to corrupt peer review, then it overlooks the fact that 70% of OA journals charge no author-side fees at all, and that 75% of TA journals do charge author-side fees. I don’t think author-side fees must undermine peer review. But if they do, most TA journals are guilty, and most OA journals are not.
Most OA journals are new, and they endure all the disadvantages of being new, including lack of prestige in proportion to their quality . But there are OA journals at the highest levels of quality, prestige, and impact. This is an existence proof that nothing intrinsic to OA prevents the highest levels of quality, prestige, and impact. If so, then we should stop talking about keeping OA journals in their place and talk instead about the barriers to their wider success, such as the fact that the vast majority of the money needed to support peer-reviewed journals is tied up in subscriptions to TA journals.
In the end, the insinuation that OA journals are for second-rate work is just TA marketing by other means, unargued FUD, and wishful thinking.
Q: I get the feeling that some OA advocates are concluding that commercial publishers should have no role (or a significantly reduced role) in an OA environment? Is that your perception? Would you agree with that point of view?
A: No. BioMed Central and Hindawi are for-profit publishers. They’re even profitable for-profit publishers. By any definition of commercial publishers, they’re commercial publishers. But they’re also exemplary OA publishers. We shouldn’t confuse “commercial” or “for-profit” with “predominantly TA,” just as we shouldn’t confuse “online” with “OA.”
In my newsletter for March 2011, I noted that the three largest commercial publishers—Elsevier, Springer, and Wiley—now publish some full OA journals, as opposed to hybrid OA journals. I see no reason to oppose these developments. On the contrary, I applaud them, or at least I applaud the general direction.
Some of these gold OA programs could be improved. Some charge needlessly high publication fees. Some use needlessly restrictive licenses. But I’d judge a gold OA program by its features, not by the history of the company behind it. If a publisher notorious for hyperinflationary price increases, restrictive licensing terms, and oppressive bargaining tactics launched a good gold program, I’d hope we could see past its record and judge the new program on its merits. Really, we’re trying to create change, not deter change. Should we judge a company by its current policies or judge its current policies by its general reputation and past policies?
If you’re interested, I wrote about these issues more extensively in 2004. I praised Elsevier when it started to allow author-initiated self-archiving. Many readers wrote in to criticize my praise, not because they thought self-archiving was bad but because they thought Elsevier was bad.
I wrote an article-length response to that criticism. I support scrutiny, and even suspicion, of promising new developments. Looking closely at new policies—to see whether they might not be quite as good as they appear—can only help. But I hope we’re past the point when steps toward OA by large, predominantly TA publishers are not even recognized as steps toward OA.
The large, predominantly TA publishers have a lot of legacy overhead from the age of print that tends to make their OA journals more expensive for authors—or fee-payers—than OA journals of the same quality from lean, mean, dedicated-OA start-ups. Hence, the big TA publishers may not do well with OA journals, where there is not only competition for authors but competition to improve terms to appeal to authors. But that’s a reason to watch closely, not a reason to discourage them from trying.
For balance, I’d also point out that OA publishers aren’t doing all they could to support the cause either. I’m not talking about journal scams that take money and perform little or no peer review, since those scams exist on both the OA and TA sides of the line.
I’m talking about OA publishers that could offer libre OA but don’t. As I mentioned, it’s much easier for OA journals than OA repositories to obtain the permissions needed for libre OA. But about 80% of OA journals settle for gratis OA and don’t offer any degree of libre OA.
Some of them want to block commercial use, for example. But instead of offering a degree of libre OA with a some-rights-reserved license such as CC-BY-NC, many use no open license at all and block many legitimate scholarly uses with an all-rights-reserved copyright. They’re not serving their users as well as they could, and they leave in place many obstacles that OA was designed to remove.
Access vs. Affordability
Q: Stevan Harnad insists that the issue of costs is irrelevant to any discussion of OA. He likes to distinguish between what he calls the “access” problem and the “affordability” problem, and I think it would be fair to say that he believes the OA movement should not spend time fretting over affordability. Would you agree? What are the issues here?
A: I’d agree in part and disagree in part. I agree that affordability isn’t enough. Solving the affordability problem wouldn’t solve many of the problems that OA tries to solve.
One reason is that even affordable prices don’t scale with the rapidly growing literature. Even if prices were low today and guaranteed to remain low forever, the total price for the total literature would still be heading for exponential explosion.
This is easiest to see at the mythical University of Croesus, which can afford 100% of the literature today. In that respect, it’s far better off than any actual university in the world. Let’s suppose that journal prices and the Croesus library budget increase at the same rate forever. For simplicity, let’s assume that rate is zero. They never grow at all, not even at the rate of inflation. Let’s assume that the journal literature grows by 5% a year, a common industry estimate.
Croesus can afford full coverage today, but in 20 years, it would have to pay 2.7 times more for full coverage, in 60 years, 18.7 times more, and in 100 years, 131.5 times more. But since Croesus can’t pay more than it has, in 20 years, it couldn't afford 100% coverage but only 37.7% coverage, in 60 years, only 5.4% coverage, and in 100 years, only 0.8% coverage. We need a distribution model that scales with the growth of research itself, not one that scales negatively by shrinking the accessible percentage of research as research itself continues to grow.
Another reason why affordability isn’t enough is that most serious research today is mediated by sophisticated software, such as search engines, and even micropayments are macrobarriers to software. Of course, we could code our credit card numbers into our software, and it could pay for access whenever it acted on our behalf. That might work for browser access to individual articles, but it would kill the tools, such as search engines, on which we depend.
One fundamental purpose of OA is to bring more and more research within the scope of these powerful tools. OA is about access for our machines at least as much as it’s about access for ourselves.
On the other hand, if the claim is that trying to solve the affordability problem diverts OA activists away from OA, I have to disagree. The affordability problem motivates some of OA’s best friends, librarians. I’m the first to admit that some OA initiatives save libraries money and some don’t, just as some methods to save libraries money help OA and some don’t. See my 2005 article on this divergence.
But saving libraries money and advancing OA also intersect, and the common ground has created an alliance of librarians and researchers that lies behind our most significant success stories. It’s inconceivable to me that we could have made the progress we have without librarians, and it’s inconceivable that librarians would have supported OA to the extent they have without seeing its contribution to the affordability problem.
Q: What precise role do you see librarians playing in the alliance you spoke of?
A: Librarians lobby for OA mandates. They write to their representatives in the legislature. They make phone calls and visit. They network and organize. They communicate with one another, with their patrons, and with the public. They launch, maintain, and fill repositories. They write up their experiences, case studies, surveys, and best practices. They pay attention. On average, they understand the issues better than any other stakeholder group, including researchers, administrators, publishers, funders, and policymakers.
Even at universities where OA policies were enacted by faculty vote, you don’t have to look far to find that librarians gave a compelling presentation at a Faculty Senate meeting, pressed faculty colleagues at Library Committee meetings, or educated individual faculty one-on-one.
Q: When we spoke in 2007, you said you expected OA to be a cheaper way of publishing research. Is that still your view?
A: There are good reasons to think that OA publishing costs less, and will continue to cost less, than TA publishing at the same level of quality. There are several studies suggesting this.
However, there are also those who dispute the conclusion, generally without evidence or with misleading evidence, such as the experience of behemoth publishers with legacy overhead from the age of print and subscriptions. I’m happy to leave it an empirical question and wait for more decisive data to emerge. But my hypothesis based on present evidence is that OA publishing will cost less.
Q: You said some OA initiatives save librarians money. Can you give me some examples and say whether this is money saved by the research community at large, or whether it is simply shifting costs from the library to another cost centre within the university or perhaps to research funders?
A: If rising levels of green OA justify libraries in canceling journals, that would save libraries money. But this isn’t happening, at least not yet, despite some publisher fears and some activist hopes. However, I don’t rule out that it will happen.
New gold OA journals don’t cost libraries money, unless they’re published by libraries, and they don’t save libraries money, unless they justify some cancellations. But I haven’t heard of any cancellations triggered by the advent of new OA journals.
However, when TA journals convert to OA, that definitely saves money at the subscribing libraries, and the number of conversions has increased every year since I started keeping track. Unfortunately, I’ve never seen a study calculating the size of these savings. I’m ready to concede that it might be small today. But it’s growing.
The world’s first and largest mass-conversion project is succeeding. I’m talking about SCOAP3. Part of the rationale for SCOAP3 is that it will save money for the subscribing institutions. When CERN announced in April that SCOAP3 was entering its implementation phase, it made a point of saying that the “SCOAP3 partners reaffirmed the importance of a mutual understanding with the publishing industry on price reduction of large subscription packages for partner libraries in countries that are part of the initiative.”
It’s true that OA journals have their own expenses. So it’s fair to ask whether the conversion of TA journals to OA is simply shifting costs. It’s true that the costs shift. But the shift still saves libraries money if the cost of a newly converted OA journal isn’t paid by libraries at the same rate at which libraries paid for the previous TA journal.
This happens when we rope in other payers, such as funding agencies. And it happens when the costs of OA publication are lower than the costs of TA publication. We see both in the case of SCOAP3. We also see it beyond SCOAP3 whenever funding agencies are willing to pay publication fees at fee-based OA journals.
Q: I’d like to end with a couple of personal questions. Unlike most, if not all, other OA advocates you gave up a full-time job in order to promote the OA cause. What has been the cost to you of that decision, both personally and professionally, and do you have any regrets (Do you, for instance, sometimes feel that it might have been better to organize your advocacy work around a paid job?)
A: It cost me financial stability, since I gave up my salary and tenure to do this. It also cost me the second half of a career in philosophy, my primary field, which I still miss.
But my decision to leave philosophy for OA was voluntary, and I’d make it again. The fact that I miss philosophy doesn’t mean that I regret my decision. If I’d stayed in philosophy, I’d have missed playing a larger role in the unfolding of OA, and I know I’d regret that.
I can imagine another person taking a path similar to mine, but only loving one thing at a time, and therefore not missing anything from the previous stages. But I don’t think that person would be luckier than I am. If there were any salaried jobs to do what I do, I’d be the first in line to apply. But I haven’t seen any, and I would have noticed.
Q: How do you fund your activities?
A: SPARC pays me for the newsletter, and the rest comes from grants and fellowships. My Berkman stipend expired after 1 year, though I’m happy to say that the fellowship continues. I have a travel budget for my trips to Cambridge and a small stipend from Harvard’s Office for Scholarly Communication. I have several grant proposals in the works, and it’s possible that I’ll have good news before this interview is published. But I often think wistfully of what it was like to have a real job.