Interview With Stevan Harnad
A Prophet Whose Time Has Come
by Richard Poynder
[A shorter version of this interview appears in the current issue of Information Today (February 2010). This version provides more detail about the conversation between journalist Richard Poynder and open access advocate Stevan Harnad. –Ed.]
In June 1994, Stevan Harnad, a cognitive scientist at the University of Southampton in the U.K., posted a message on a mailing list that called on fellow researchers to make their papers freely available on the internet. The message became known as the Subversive Proposal.
“For centuries,” wrote Harnad, “it was only out of reluctant necessity that authors of esoteric publications made the Faustian bargain to allow a price-tag to be erected as a barrier between their work and its (tiny) intended readership because that was the only way to make their work public in the era when paper publication (and its substantial real expenses) were the only way to do so. But today there is another way. …”
In the online age, scientists can make their research freely available to all, he wrote, allowing them to better “build on one another’s work in that collaborative enterprise called learned inquiry.” And the self-archiving movement was born; this later became known as Green Open Access.
Concluding that the logical place for researchers to make their papers available was not in large discipline-based databases such as arXiv (the physics preprint server) but in a network of local repositories, Harnad also became a key figure in the development of the institutional repository movement.
But it has been far from smooth sailing. Harnad had assumed that researchers would immediately see the logic of self-archiving, not the least of which are the benefits they would gain from the greater visibility and impact that their research would have once it was freely available. However, few did (only about 15% of authors self-archive spontaneously).
Out of growing frustration with researchers’ passivity, Harnad became an ardent advocate for the introduction of self-archiving mandates. For the past several years, he has grown hoarse calling upon universities and research funders to require researchers to self-archive their papers, a message he feels has too often been diluted by growing interest in OA publishing (Gold Open Access), which Harnad believes to be a far less certain road to OA.
Fifteen years after he posted his Subversive Proposal, the self-styled “weary archivangelist” is a somewhat disappointed OA advocate today.
There are still only 139 mandates in the world now, and many of those that have been introduced lack teeth. Consequently, most institutional repositories remain all but empty. Meanwhile, the research community keeps succumbing to what Harnad calls “gold fever.”
But it is not all doom and gloom. “All of the U.K.’s Research Funding Councils and 14 U.K. universities have mandated OA,” says Harnad. “So have NIH [the National Institutes of Health], Harvard and MIT in the U.S.”
And with last year’s reintroduction of the Federal Research Public Access Act (FRPAA) and President Obama’s recent public consultation on requiring U.S. federal science and technology funding agencies to introduce public access policies, Harnad is hopeful that in 2010 we will finally see the tipping point needed to usher in universal OA.
Harnad discusses this and more in the interview that follows.
Q: Last year was the 15th anniversary of the Subversive Proposal , an online message you posted in June 1994 (and which subsequently formed the basis of a book ) calling on fellow researchers to make their papers freely available online by “ self-archiving ” them. This was 7 years before the term “open access” (OA) was coined. What would you say has been achieved since 1994?
A: The first point to make is that my call was not heeded! NIH Director and Nobel Laureate Harold Varmus voiced a similar hope in 1999 (E-biomed), another milestone, but far more influential. (His call was not heeded either.)
There have been a number of other milestones. The first was the creation of the Cogprints repository in 1997 in order to help researchers self-archive. CogPrints was modeled on the physics preprint repository arXiv in the hope that if people wouldn’t put their papers into their own local ftp archive (as I had proposed), they might put them in a central one such as arXiv (i.e., Cogprints).
Today, Cogprints has still only attracted about 2,500 deposits, so it is a miserable failure. What I didn’t realize at the time was that arXiv was successful as result of the physics community’s unique preprints culture, not because a repository was provided.
Q: The challenge was to get other disciplines to copy physicists?
A: Yes. And Cogprints reflects the temporary belief I had that a central archive was the solution. Another milestone I would mention was the launch in 1999 of the Open Archives Initiative [OAI], which established an interoperability protocol to enable repositories to interact. My hope was that if providing a central repository wasn’t enough to get researchers tapping their keyboards to deposit their papers then maybe providing free interoperable software to allow the content in multiple institutional repositories to be aggregated would do (EPrints 2000). That was the start of the institutional repository movement.
Q: That was a better solution than asking researchers to put their papers in an FTP archive?
A: Sure, but the repository movement headed off in all directions with people trying to create digital libraries and putting all kinds of extraneous content into repositories. Soon we ended up with a lot of repositories but mostly empty of OA’s intended target: peer-reviewed papers.
Q: Today still only around 15% of the peer-reviewed literature is being self-archived, I believe?
A: And so it became apparent that it was necessary to mandate that researchers self-archive. That this was the right solution became apparent after mandates were introduced at Southampton University (2003), Australia’s Queensland University of Technology and Portugal’s Minho University (2004), which worked.
Q: What you learned was that it is no good waiting for researchers to start tapping their keyboards, but you have to persuade the administrators of research institutions to require them to do so, by means of a mandate?
A: Because it is a tad easier to persuade administrators to mandate than to persuade researchers to tap their keyboards, and there are fewer of them to persuade, each covering much more research. It has always been evident what is needed, to get the 2.5 million articles a year published in the 25,000 peer-reviewed journals all freely accessible online.
But problems came up with mandates too (even with the word “mandate”!) At some points, even the meaning of “open access” created obstacles, some, like me, using it to mean free online access to refereed research, others to include further rights such as republication and “mash-up” rights that would just have made mandates and free access itself harder to come by.
(The rest will come as a natural matter, of course, but the history of OA so far has been one of gratuitous over-reaching that has not only netted little, but it has failed even to grasp what has already been well within reach for some time: free online access to refereed research.)
Q: Yes, there has been considerable confusion. I note you did not mention the 2001 Budapest Open Access Initiative (BOAI), where the term “open access” was coined and where the two strategies of Green OA (self-archiving) and Gold OA (open access publishing) were devised. I don’t think anyone would dispute that you have played a very important role in the OA movement, but perhaps your most important contribution was to persuade other BOAI attendees of the need for Green OA in the first place?
A: I think there are plenty of people who would dispute that my role has been important (I among them, if importance is gauged by success in persuasion to self-archive)!
Q: Here’s a quote that might surprise you. One of those who attended BOAI was Jean-Claude Guédon . He said to me, “The great majority of people at Budapest were really on the side of journals (i.e., Gold OA), and I would say retrospectively, ‘Thank God Stevan held his ground,’ because in the end, he forced us to introduce the Green road, which turned out to be an extraordinarily powerful weapon in the development of open access.” What are your thoughts on this?
A: It is true that at the outset just about everybody else at Budapest—with the notable exception of Peter Suber — greatly underestimated the value and importance of self-archiving (later called BOAI Strategy 1, and then “Green OA”) and were focused single-mindedly on creating “new journals” (later called BOAI Strategy 2, and then “Gold OA”). This is clear in the persistent notion that Green OA is a “weapon” (presumably in the pursuit of Gold OA) whereas it is not: It is OA!
So, my own impression of the outcome of the Budapest meeting was slightly different: I had tried (and failed) to make Green the sole priority, not just to put it on the agenda (though Green did end up with more votes than Gold).
Q: What we learn from this is that had you not been at Budapest there would be no such thing as Green OA?
A: But we still don’t have much Green OA!
Q: True, but without you, there would also be no talk of mandates I suspect?
A: Perhaps, though we don’t have many mandates yet either!
But you are right: Frustrated at the glacially slow rate of spontaneous self-archiving, I started talking about funders and institutions needing to mandate self-archiving back in 2000 (or earlier), even before I was quite sure I was saying anything realistic. Bachrach et al.’s 1998 call for copyright retention had a similar motivation, but as we have seen, mandating copyright retention is far more complicated and faces far more obstacles than just mandating self-archiving (and, most important, it is not necessary for self-archiving).
More About Mandates
Q: In 2004, I wrote a 10-year retrospective of the Subversive Proposal for Information Today. At that point, the U.K. Science and Technology Select Committee had just recommended that all U.K. researchers should be mandated to make their papers freely available on the web. When news of the Select Committee recommendation broke, one Australian OA advocate emailed you to say, “You must feel like a prophet whose time has come.” That year, the U.S. National Institutes of Health (NIH) seemed about to introduce a mandate, and mandates were being discussed in a number of other countries. At the time, you commented to me that we were witnessing “a historic race to see which nation actually implements a mandate.” Today, there are 13 9 mandates worldwide. In retrospect, would you say that 2004 was a tipping point or a false dawn?
A: It was a false dawn. Indeed, it has been all false dawns until now. By the way, it was the Wellcome Trust that introduced the first funder mandate (in 2005); NIH’s policy was just a request until it was upgraded to a mandate after 2 years of failure.
Q: You said that problems came up with mandates. What went wrong?
A: Wellcome Trust and RCUK [Research Councils UK] did eventually introduce a mandate. But ambiguity developed over who should deposit what, where, and when. The Gold OA issue somehow got intertwined too. So funder mandates began to be emulated, but so were their (easily remediable) flaws.
And the issue of Gold OA journals somehow started to get intertwined with the issue of access. (Everybody keeps getting seduced and side-tracked by golden fantasies. Even opponents of Gold OA have tuned into the fallacy that OA = Gold OA publishing.)
But at least funder mandates began to get the topic in the public eye. Nobody paid any attention when it was just Southampton, QUT and Minho mandating; but when talk began about an NIH mandate, everyone perked up. The problem was that they perked up to the wrong model.
Q: As you say, the NIH mandate was initially only a recommendation. And even after it was made compulsory publishers were able to insist on a 12-month embargo before papers were made freely accessible. The mandate also required deposit in the central archive PubMed Central , rather than in the fundee’s institutional repository; and it allowed publishers to do the deposit in place of the fundees. All these aspects you have criticized. The next most significant mandate was perhaps the one voted for by the Faculty of Arts & Sciences at Harvard in February 2008 . This too you have criticized. Why?
A: Harvard definitely set institutional mandates into motion just as NIH had set funder mandates into motion. But, like NIH, Harvard got some crucial details wrong.
Nobody is quite sure yet whether the Harvard mandate is indeed a mandate, because instead of requiring all researchers to deposit their final refereed draft, Harvard asked authors to adopt the author addendum , an agreement with the publisher that the author and institution can deposit their paper, and re-use it in various ways. In my view some of these extra reuses are not only difficult to negotiate with publishers, but they are not even necessary.
Q: In fact, it is not a mandate. You will know that Stuart Shieber wrote on his blog , “The Harvard open-access policy could not be, should not be, and is not a mandate. I’ve tried to be very careful never to refer to it as a mandate.” That’s pretty clear, isn’t it?
A: It’s pretty muddy! Stuart doesn’t say why Harvard’s is not a mandate (nor why it shouldn’t be). Other universities have mandated.
Q: As I recall, Shieber argued that it is not possible for a faculty to introduce a mandate. Only funders and governments can do that?
A: Which is incorrect because universities have done it, and Alma Swan’s surveys have shown that the vast majority of researchers are OK with mandates. Meanwhile, Arthur Sale has shown that they comply with mandates.
Look, I want to give full credit to Stuart. He did something no one else had done: Before Harvard’s, all mandates were top-down—a provost or VC would adopt it. At Harvard, it was self-imposed by the faculty, unanimously.
So it was a first, and it was Harvard, which is glorious, invaluable. But if it had also been the right mandate, it would have been a tipping point because people are now imitating the Harvard mandate. They might instead have been imitating a model that works, and scales.
Q: What is the right kind of mandate?
A: The right mandate is the one that Professor Rentier adopted at Liège University last year.
Liège is not slavishly imitated in the way that Harvard is, but it is the optimal mandate: Professor Rentier linked Liège’s ID/OA mandate with its submission procedure for annual performance review: “Henceforth, dear researchers, you will need to deposit your final drafts in our institutional repository, for when we do our annual performance review, we are going to extract the data from our repository. If your data aren’t in there, we won’t see them!”
And it has worked.
Q: As you indicate, this is what you call an ID/OA , or Immediate-Deposit/Optional-Access, mandate. An ID/OS mandate requires all researchers, without exception, to deposit their papers immediately on publication. But if they want, they can set a “Closed Access” flag against any paper. This caters for publisher embargoes, but ensures that the paper is nevertheless discoverable (via its metadata). Researchers webwide can then hit the institutional repository’s automatic “email eprint request” button to ask the author for an electronic copy to be forwarded to them for research purposes. Correct?
A: Correct. The point is that even if the article is made Closed Access during an embargo, it doesn’t matter greatly. While 37% of journals embargo OA to various degrees today, 63% endorse immediate OA. The 37% must still be deposited immediately so 100%, without exception, is being captured. The eprint button can tide over access needs for embargoed deposits.
And I am as sure as I am of anything that once ID/OA is universally mandated, the rest of the dominoes will quickly fall (including the demise of access embargoes).
Q: As you said, there is a widespread view that Green OA is a weapon; a tool, if you like, for achieving OA, not OA in itself.
A: OA as a tool for achieving OA? This uses the term OA twice, referring to itself by bootstrap.
How can Green OA be a tool for OA? That implies that OA is something other than OA; that Green OA therefore isn't OA. (Is this not the usual conflation of OA with publishing reform, i.e., Gold OA?)
I see OA as an end in itself, for research and researchers; not just a means to some other end (though it might turn out to be that too, if only we give it a chance to happen).
Q: Perhaps, but many nevertheless see Green OA as a lever for forcing publishers to embrace Gold OA. Since they fear that self-archiving will destroy their business, the theory says, publishers will adopt Gold in the hope of thereby averting Green OA being mandated. Do you think Green OA threatens the business of scholarly publishers?
A: Well, I know that you too, Richard, believe publishers have taken Gold OA more seriously because they were scared of Green OA. But I’m not sure that hypothesis is true. The majority of publishers went Green in response to demands for OA; it seems to me it was Gold OA they feared then. (And they bet—rightly—that researchers wouldn’t bother to provide Green. But they probably hadn’t reckoned on Green mandates ...)
Q: I think it is fair to conclude that publishers fear Green OA from what the chief executive of the Association of Learned and Professional Society Publishers (ALPSP) Ian Russell said to the Times Higher Education last November, “Repositories are parasitic on the existing journal structure for their peer-review process ... If the subscription journals are unable to sustain themselves, then what will provide that authority and badge of trustworthiness?”
It seems to me that implicit in Russell’s statement is a claim that Green OA will destroy scholarly publishing. And it is surely a credible claim: Self-archiving consists of researchers making papers freely available on the web that would otherwise only be available to fellow researchers if their institution had a subscription for the journal in which the paper had been published.
A: I’m still not sure the reason publishers are taking Gold seriously is just because they’re scared of Green (or Green mandates), although it could be.
Another hypothesis is that publishers thought, “Look, if the research community is clamouring for OA, and that’s what they really want, then let them put their money where their mouth is, and maybe we can make some more money.” That would be a natural thing for publishers to do. (But my own interest is in the research community’s motivations and interests, not the publishing community’s.)
Q: You have yourself described Green OA as a tool. You frequently say, for instance, that if all institutions introduce self-archiving mandates publishers will be forced to downsize.
A: I periodically voice that counterspeculation but only in response to the speculation that self-archiving will destroy peer-reviewed journals.
Q: Do you really think that the research community, research funders, and governments, will sit on their hands if, instead of downsizing, publishers started to exit the market?
A: I’m not sure what you mean by sit on their hands. Until now, the 750 to 1,500 key institutions have been subscribing to the main journals. This has created an inelastic market that has allowed publishers to make large profits.
Green OA mandates from funders and institutions will reach a tipping point where self-archiving becomes universally mandated. Once universal Green OA is achieved (i.e., the final drafts of all 2.5 million of the articles published in the 25,000 journals annually will be freely accessible online), cash-strapped libraries might cancel subscriptions more and more.
Publishers will then say, “We’re not making the revenues we did before! What can we do to cut costs?”
Q: Or perhaps simply exit the market?
A: Because mandate adoption and compliance will grow anarchically and randomly, by the article—rather than systematically, by the journal or by the publisher—cancellation pressure will be gradual, as the overall OA percentage grows. What do you do when demand is declining? You try to cut costs. In the beginning, journals may jettison their paper edition (and its costs), because if institutions are cancelling, that means they don’t feel they need the paper version anymore; so journals will be available online only.
And then publishers will realize that nobody even wants the online version anymore because it is available free. Piece-by-piece, they will cut other obsolete products and services (and their costs).
The point is that in a world of universal Green OA, you don’t need publishers to provide access, as that can be provided through the network of institutional repositories. Repositories can also archive the papers. Indeed, all search and dissemination can be done by Green OA repositories (and their OAI harvesters). Publishers just implement peer-review.
Publishers do, of course, still provide a print edition today, whose cost—for as long as there is a market for it—somebody has to pay for. But it’s ridiculous to wrap that cost into the overhead costs of publishing when charging the author-institution for OA by the individual outgoing article—as some hybrid OA publishers are currently trying to do. A print edition is not part of the cost of publishing in the OA era.
And yes, some publishers will decide to leave the business, but that is perfectly fine. What is a journal after all? It is an editorial board, a stable of referees, an authorship and a readership—a title, in other words. And titles have migrated from publisher to publisher for decades. It’s the same journal. It just changes hands. And there will be plenty of Gold OA publishers ready to take over migrant titles.
Q: If your scenario is correct, then what do we make of the current rush to create Gold OA funds ? And what do you make of the Compact for Open Access Publishing Equity (COPE)? When COPE was launched its architect—Harvard’s Stuart Shieber —published an article arguing that it is important to “send a signal to publishers and scholarly societies that research universities and funders appreciate and value their contributions and that universities and funders promoting self-archiving have every intention of continuing to fund publication, albeit within a different model.” In effect, COPE says that publishers have to be thrown a financial lifeline.
A: COPE is based on the illusion that there is enough money available in institutions today to pay for OA publication in all the must-have journals— Nature, Science, APS, and all the other top journals—while continuing to subscribe to those journals (and we don’t as yet have OA for their contents, so it’s premature to cancel). I doubt researchers would want to see their already-scarce research money redirected to pay for that either.
The only way the money would be available to continue giving those journals the income to which they have become accustomed is if current subscriptions were cancelled and that money was redirected toward paying OA publication costs. So COPE is simply throwing more money at publishers, pre-emptively (before the money is needed or available), and prematurely (before the asking price is right). But only a bit of money. COPE is just a token, covering only a fraction of what is being spent on subscriptions today.
Q: So you believe that mandating Green OA is a better solution than Gold OA not just because it can provide OA more quickly, but because it will squeeze costs out of the system by forcing publishers to downsize?
A: Quicker and surer. (I don’t care about downsizing.)
Q: Time will tell, but there is no doubt that more and more institutions and funders are providing money to pay for Gold OA on top of subscriptions. At the end of last year, for instance, Wellcome announced that it was adding another £2m to its Gold OA pot. Likewise, the Dutch Science Foundation announced that it was making €5 million available. Here’s the fear that publishers are going to downsize, here’s the money, and here is Shieber calling on the research community to prop up publishers. Your thoughts?
A: Yes. Stuart was incredibly effective getting faculty consensus; a historic break-through. I couldn’t have done it; I would have alienated everybody.
But when you are successful, you tend to get superstitious about the reasons for your success. Maybe at the back of Stuart’s mind was the thought that the way he had succeeded in convincing faculty to agree was by allaying their fears by promising to prop them up with extra money.
Q: Which goes to my question: Is the research community really going to sit on its hands while publishers exit the market? Today, it is just a fear. What happens when they do start exiting?
A: As I said, publishers exiting does not mean titles dying.
Stuart replied to faculty who expressed worries about publishers’ future, “Look we will find money somehow, somewhere, to keep paying publishers, so don’t worry about that”—and he founded COPE.
Instead, Stuart could have replied, “Don’t worry. Green OA grows anarchically, by the article, not by the journal. An institution can’t cancel its journals until its users have access to those journals’ articles. Once all journal articles are Green OA, the institutions can cancel, and use (a fraction of) those cancellation savings to pay for publication, by the article, via Gold OA. To pay now, before Green OA, would be gratuitous. If some publishers drop their titles, others will pick them up.”
The point is that we need Green OA first, to generate the cancellation pressure that will cause costs to be cut down to only what is still necessary in the OA era, as well as to release the subscription money to pay those remaining costs (instead of poaching scarce funds from elsewhere to pay today’s needless and inflated asking prices for Gold OA). All those inessential things that we think we need to prop up now (paper edition, online edition, dissemination, archiving) don’t need to be propped up.
Q: So I guess the question is whether it is Shieber’s response or Harnad’s response that ends up gaining greater credence. And on that question hangs the likelihood of whether the research community will continue to throw more money at publishers, or eventually say to them, “Downsize, or go away?”
A: Clearly, it’s not just what one says, but how one says it. I evidently say things in ways that don’t convince. (At least not in the short-term. People do have a habit of coming around eventually, though ...)
The Hybrid OA
Q: We can speculate as to whether publishers really view Green OA as a threat, but they have undoubtedly responded to it. Publishers such as Hindawi have converted entire portfolios of journals from subscription to Gold OA. As you said, others have begun offering so-called Hybrid OA. In fact, most large subscription journal publishers offer a hybrid option today. What is the difference between Hybrid OA and Gold OA?
A: Hybrid OA is where a journal continues to publish both its paper and online editions and continues to sell it all on a subscription basis. But over and above the revenues it earns from subscriptions, it also offers authors the option to pay extra to make their individual article OA. Pure Gold OA, by contrast, is where all a journal’s articles are published OA.
Q: Is Hybrid OA a good thing?
A: I first thought Hybrid OA would help Green OA because it would be so obvious that instead of spending $3,000 to make an article OA. an author could just spend 10 minutes clicking on a keyboard.
Q: In other words, they don’t need to pay to publish because they could continue publishing in subscription journals and then self-archive their papers, at no cost?
A: Yes. But in the end, Hybrid OA distracted people. The research (and library) community kept getting more caught up in golden dreams about publishing reform than they are in the original dream of just getting online access free for all, which, after all, is what OA is supposedly all about, and for.
Yet, the question researchers really ought to be asking themselves is, “Why try and reform publishing first, when we can all have OA right now, without further ado?” For years now, we just keep over-reaching, while failing to grasp what’s already within our reach.
Q: Would it be correct to say that you would have no problem with Hybrid OA if it had not proved a distraction to achieving Green OA?
A: None at all. As I say, I was happy that they offered Hybrid OA because I didn’t believe for 1 minute that people would waste their money buying it, and I assumed it would make Green OA even more attractive. All you can do by overpricing something —especially if people want it—is to make them wonder if there isn’t a better way of getting it.
And remember that the amount that publishers are asking for publishing articles OA today, whether hybrid or pure Gold OA, is vastly inflated, relative to what it would cost if we already had global Green OA.
Q: Yes, I note that in the Times Higher Education article, you estimated the cost of publishing a paper at about $500. Yet today, Gold OA journals such as PLoS Biology and PLoS Medicine charge $2,900, and hybrid publishers such as Springer and OUP (Oxford University Press) charge $3,000. Cell Press, meanwhile, charges $5,000 per paper. Thoughts?
A: That’s because these publishers are co-bundling all kinds of extras that cost (or earn) money today, that are not intrinsically necessary. Green OA can shake out all these inessential add-ons.
Q: As always, everything you say is totally logical. Nevertheless, it seems increasingly possible that Gold OA will prevail long before Green OA is able to trigger this shake out. And at the point where universal Gold OA has prevailed Green OA becomes moot, does it not?
A: That’s news! Where are the growth curve and tipping point for this scenario?
Q: As we discussed, we are currently witnessing a rush by research institutions to create Gold OA funds to help authors pay article processing charges . One can envisage a situation in which the research community eventually persuades all publishers to convert to pure Gold OA. Comments?
A: Some extra funds are indeed being offered, but count the money in them: The funds are tiny. And count the uptake: It too is tiny. And where do you imagine the rest of that extra pre-emptive cash is going to come from?
We can fantasize about publishing reform, but the fact is that researchers still need to have access to the journals they use and—in the absence of OA—what is providing that access today is institutional subscriptions. That means the potential money to pay for Hybrid Gold is currently tied up in subscriptions. Until institutions have OA, they can’t cancel subscriptions in order to release the money to pay directly for OA. Green OA would release it; Gold OA demands it in advance.
Does Not Scale
Q: But couldn’t subscriptions gradually be converted into OA institutional membership schemes such as those offered by BioMed Central (BMC) and Hindawi , or indeed Hybrid publishers such as Springer? Isn’t that what the deals with the Max Planck Society and with the University of California are about: converting subscription publishing to OA publishing?
A: That is indeed being tried locally (by publisher, institution, or field), but set aside whether it’s necessary: can it scale? Can global subscriptions morph coherently into global “memberships”? If you look at SCOAP 3 [Sponsoring Consortium for Open Access Publishing in Particle Physics], you will see that it too is basically a membership scheme. Essentially, the HEP [high-energy physics] community has said to publishers, “Let’s agree on what it is that you are doing now, and let’s re-baptize payments as ‘OA memberships’ instead of ‘subscriptions.’ We will agree to a consortial membership fee and everything will continue as before, except we will be paying for consortial membership instead of subscription and you will make it all OA.”
The difference between BMC and Hindawi membership and SCOAP 3 membership, however, is that SCOAP 3 membership is focused on one homogenous community, the consumers of high-energy physics—most of whom, we should add, already make their content (Green) OA today, (thanks to the longstanding HEP preprint culture). For them this lock-in and re-baptism is particularly foolish because they are just committing to keep paying for what they already have: subscriptions and OA.
But while it works, by fiat, at present prices, in high-energy physics, it does not scale. Institutions such as CERN subscribe only to HEP journals. Harvard can join by renaming its current physics journal subscription fees “membership dues,” but it still has to pay for all the other journals it needs. So why wouldn’t Harvard just opt out of “membership” next year?
Q: Every publisher could go to every institution and say, “Let’s re-baptize what we do. Pay us the same amount of money, let’s now call it an institutional membership and make it OA.”
A: Yes, but remember that the similarity between annual memberships and annual subscriptions is that both can be cancelled at the end of each year. Once all the complete contents of a journal are being published OA, any institution can say, “Why do we need to keep paying for this institutional ‘membership’? Everyone’s getting access anyway!” What is the answer to that for publishers? (In biology, this is called an “ evolutionarily unstable strategy.” It can be quickly invaded by freeloaders.)
The natural way to pay for Gold OA is by the individual submitted article, not by annual bulk institutional membership.
Q: That, presumably, is why institutional membership schemes are sold as pre-paid article processing charges, and usually calculated on the basis of individual articles. And institutions are told that they are no longer paying for access, but for publishing a certain number of articles.
A: 10,000 institutions (or, if you like, the top 1,000) with “memberships” in 25,000 journals (or the top 2,500), each journal with an annual quota of accepted articles from each institution? Even with a journal-fleet publisher oligopoly negotiating collectively with a global institutional consortium and even allowing for annual pro-rated quota adjustments and even setting aside the problem of how to “peer-review” submissions from “members” with paid annual quotas, this Escherian edifice looks more promising for global prepaid medical insurance or meal plans than for the needs of 10 million individual researchers (or the top 1 million) accustomed to picking their own individual journals, paper by paper, not according to prepaid journal-free quotas—and peer-reviewers who are accustomed to deciding on acceptance based on quality, paper by paper, without any constraint from quotas. ... And it is on this basis, Richard, that you think it “seems increasingly possible that Gold OA will prevail long before Green OA”?
Looking for the Tipping Point
Q: OK, let’s leave that question hanging. What would you say were the most significant OA developments in 2009?
A: I am not sure that there have been that many.
Q: Was the Houghton Report not significant?
A: It is a mixed blessing. I am just as happy about the Houghton Report as I was about the NIH and Harvard mandates: They are all welcome, timely things to happen, and they all help, but all of them have their (easily remediable) flaws.
Q: What is the flaw in the Houghton Report?
A: In the process of making a strong positive case for OA and how it is going to save money, it frames the discussion largely in terms of publication costs and savings. It hardly reckons in terms of the economic and epistemic benefits of having research more accessible to researchers, which is in my view the real rationale for OA.The report should have put much more emphasis on the benefits of OA for research productivity and progress.
Also, in their calculations they more or less take for granted the current costs of the products and services that publishers (either Hybrid or full Gold OA) co-bundle today.
Q: Even so, I note that while the report estimates the cost of publishing a paper at $2,536, PLoS Biology and PLoS Medicine are already charging more than that?
A: Because that includes many services that will become completely unnecessary if and when Green OA prevails.
Q: Publishers were nevertheless very upset about the report. They argue that it underestimates the economic cost of publishing an article. Comments?
A: Whereas it actually overestimates the costs, because it doesn’t take into account the downsizing potential of universal Green.
Q: That, of course, is to assume something that might never happen?
A: You’re right: At the current rate, the heat death of the universe may well come first ...
Q: What about the Federal Research Public Access Act (FRPAA). I assume you welcome its reintroduction in the U.S. Senate?
A: Very much! If FRPAA required all top U.S. funders to mandate institutional deposit immediately upon acceptance for publication—an ID/OA mandate—then that would soon be reciprocated by emulators worldwide, not only funders but the all-important institutions too. So where today we see flawed mandates being cloned, we would start to see an optimal mandate being cloned. The tipping point we’ve all awaited so long.
Q: And the public consultation on a Public Access Policy launched in December by the Obama administration … that too could be good news?
A: That was completely unexpected, a wonderful surprise, and another sign of the responsive and thoughtful approach President Obama promised, and seems to be delivering. (I’ve again made a lot of detailed suggestions on the OSTP website—as I’ve done many times before, only to have them ignored: but this time I am optimistic that the outcome may be different.)
Q: Another development we saw in 2009 was the launch of a new OA organization—EnablingOpenScholarship (EOS). EOS is headed up by the architect of the Liège University mandate Bernard Rentier and seems focused on promoting Green OA. That’s good, yes?
A: It certainly is!. It will provide informed and focused OA policy guidance for universities. And it couldn’t be in better hands: Bernard Rentier, and Alma Swan, who is the convenor.
Q: So there are 139 mandates today. You have said that we need 10,000 to achieve universal OA. There’s a way to go, isn’t there?
A: Remember I said thatthe breakeven point for a journal looking to be viable is to sign up the top institutions in the world, which is somewhere between 750 and 1,500. The same applies to mandates. Once we have mandates at the top 750 to 1,500 institutions, we’re safely past the tipping point; the others will all follow suit soon enough.
(137/750 is about 18%, so it’s coming along, though ironically that also happens to be about the global spontaneous [unmandated] self-archiving rate.)
Q: Meanwhile, there are currently around 4,500 Gold OA journals. You say there are 25,000 peer-reviewed journals?
A: So roughly 20% are already Gold OA (not counting the Hybrid Gold option, which is a cheap option for publishers, but a pricey one for authors). But the problem is that that 20% includes very few of the top journals in each field, the ones that publish the research that everybody wants and needs the most. If you look at the top journals, the ones that are likely to capture 80% of citations, most of those are not Gold OA. A tipping point for Gold OA would require capturing the top journals.
Q: What do you hope for OA in 2010?
A: I hope FRPAA succeeds. With the help of the Obama OSTP initiative, it just might! I hope the new EOS and OASIS organizations, perhaps with the help of the Houghton Report, will manage to promote the ID/OA green mandate to lots of institutions in Europe and worldwide, taking us to the green tipping point at long last.
The only point I have my eye on today is that green tipping point. None of the other stuff matters at this time. The only thing standing between us and universal OA is author keystrokes, and all that’s needed is that funder and institutional mandates should set those fingers into motion at last.