PLoS ONE: Take Two

Declan Butler’s article about PLoS financials in this week’s Nature has provoked a predictable – and many ways understandable – backlash from open access fans (see Bora’s blog for links and summaries).

First, to deal with a few of the gripes raised in the various blog posts:

  1. Nature isn’t anti-open access. Its coverage of everything from ChemSpider to PubMed Central and Wikipedia ought to make that clear. Heck, it even broke the original story about Eric Dezenhall’s involvement in the debacle subsequently known as PRISM.
  2. Declan isn’t anti-open access either. But like me, he’s a realist. Here’s what he wrote in 2003:

    Few people would disagree, in principle, with the ideal of open access. The question is whether the economics can be made to work.

    Quite so. We’ll come back to this later.

  3. The idea that commercial publishers are inherently evil lingers on. I can understand why, but it’s completely untrue. BioMed Central is for-profit and the American Chemical Society is not-for-profit. Now, tell me which is more progressive.
  4. There were also quite a few comments to the effect that the article was self-serving given Nature‘s business interests. That completely ignores the overall balance of Nature‘s coverage, which in my opinion is broadly supportive of open access even though Nature itself charges subscription fees. The calls for a disclosure of competing interests sound reasonable to me if there really is a danger that any reader might not notice the link. But in that case let’s also have them on all those PLoS editorials espousing open access.

Anyway, enough quibbling. Let me come at last to my main point: in some important respects PLoS is indeed failing. However, I’m not sure that Declan’s article made it sufficiently clear why, so let me try.

PLoS’s original goal wasn’t to show that author-pays publishing can be made to work for low-end journals. (By “low-end” I mean relatively low rejection rate and low editorial input. I don’t intend it to mean “beneath contempt” or anything similarly pejorative – these journals have an important place in the overall mix.) At the point PLoS came into existence as a publisher in 2002, BioMed Central had already been working on this for three years – and without shedloads of charitable funding. Instead, PLoS’s goal was to show that something similar could work for high-end journals (again, defined by high rejection rates and high levels of editorial input). As Helen Doyle, a PLoS director, put it in 2003:

PLoS Biology must be more than a solid scientific journal; it must prove that a new open-access journal can generate top-notch papers, be supported by excellent reviewers, develop a following among diverse and discerning readers, and [Helen’s emphasis] become a sustainable business.

This was a much more ambitious goal because at this end of the market the economics are, to put it mildly, challenging. PLoS’s own expectations in 2004 that it would break even by 2006 were not fulfilled [764K PDF]. What Declan’s article (and an earlier, equally controversial piece by him) shows is that PLoS hasn’t even come close, at least with respect to its high-end journals.

This doesn’t matter, argue some open-access fans, because PLoS doesn’t exist to make money. And in any case, PLoS ONE is making money and taking PLoS as a whole towards break-even. Actually it matters a lot, mainly because of effects beyond PLoS. To understand this, consider its aims. Here is the mission statement from PLoS’s 2003 accounts (available from

To realize [the potential of unrestricted online access to scientific information], a new business model for scientific publication is required that treats the costs of publication as the final integral step of the funding of a research project. To demonstrate that this publishing model will be successful for the publication of the very best research, PLoS has published its own journal, beginning with PLoS Biology in [sic] October 13, 2003 and PLoS Medicine in October 2004.

PLoS is working with scientists, their societies, funding agencies, and other publishers to pursue our broader goal of ensuring an open-access home for every published article and to develop tools to make the literature useful to scientists and the public.

And in case any of this is unclear, here is a snippet for PLoS’s current FAQ:

PLoS is launching journals to demonstrate the enormous benefits of open access publishing. Our goal is to make the scientific and medical literature a freely accessible resource, but the literature is huge, and we cannot (nor do we want to) do it all by ourselves. Using the success of our own journals as a template, we hope to encourage other publishers to adopt the open access model. This has always been our goal.

That’s a great goal, and with millions of dollars from the Gordon & Betty Moore Foundation to cover the costs of making this transition, maybe even an achievable one. But it hasn’t been achieved, and in some ways quite the opposite has happened.

The situation back in 2002 was that existing publishers (Nature Publishing Group among them) were very skeptical that author-pays economics could ever work for high-end journals. Fast-forward to 2008 and far from bringing other publishers along with them, PLoS has helped to confirm these suspicions. Moreover, it has created heavily and perpetually subsidised journals that act as a strong disincentive for anyone else to try a similar publishing model. Which other publisher will enter the high-end author-pays segment now that there is an incumbent with ‘private income’ that will always be able undercut any competitor who needs to be economically self-sustaining? In short, the effect has been to reduce competition in an industry that, for the good of science as a whole, ought to have more competition.

Since the economics for high-end author-pays journals may never be workable, it’s legitimate to argue that PLoS has merely put some barbed wire and a big “Keep Out!” sign on top of a barrier that was already insurmountable. But that hardly makes it a catalyst for positive change in publishing. And although this outcome was never the intention, that doesn’t make it any less unfortunate.

I’m not trying to criticise the principle of open access (on the contrary, I’m a fan), or diss a competing publisher (in my opinion organisations such as BioMed Central are doing genuinely good things), or even pick a fight with PLoS (which has noble aspirations and some employees whom I consider friends). But we need to be able to discuss these things openly because anyone who doesn’t care about the economics of open access doesn’t care about making it a reality. And it has to be possible to be critical without the risk of knee-jerk counterattacks. (Right now I don’t think this is possible, but I guess we’ll find out.)

Certainly some ‘traditional’ publishers have created barriers to progress, and it’s right to criticise those. But we also need to be alert to the Law of Unintended Consequences among organisations with good intentions. To me, this is just such an example.

Consider for a moment what you’d do if you really wanted to hinder the spread of author-pays, open-access journals. A few publishers tried PRISM, which was understandably described as “evil” (and the Dr. Evil pose is pretty funny). But that word (even when unaccompanied by its frequent companion, “genius”) conjures up the image of someone with an IQ score that’s, well, at least into double figures. For reasons I’ve covered before, PRISM is just too dumb to be classed as evil; “brain-dead” would be closer to the mark.

No, if you really were Dr. Evil, Science Publisher, and intent on imposing limits to the spread of all this open access nonsense then you’d have a much more cunning plan up your grey flannel sleeve. For example, you might imagine setting up a small series of loss-making open-access journals, thus killing off any chance of a competitive market emerging in that segment. You’d fund them through donations for as long as possible, but such income is vulnerable to the vagaries of donors so eventually you’d have to cross-subsidise them from a profitable business line. And all the while you’d explain that you’re doing the world a favour. At this point in your imaginings you’d stroke your hairless cat and laugh uproariously, content in the knowledge that you need not lift even your little finger to implement any of this because PLoS has already done the work.

Of course, I’m not saying that this was PLoS’s intent, only that it’s the result so far. Just as PRISM – an initiative designed to reinforce the importance of publishers and undermine open access – succeeded only in making the participating organisations look like fools while at the same time emboldening the open-access camp, so PLoS is creating effects that run directly against its stated aims of spreading the open access model across all of science publishing. Personally I find this very sad: since I work for a high-end science publisher I’d like there to be a workable economic model that we could adopt too. Unfortunately we’re further away from that than ever.

To look on the bright side, none of this may matter very much in the longer run since truly widespread open access to scientific content is coming about through funder-mandated archiving, not open-access publishing. Nevertheless, the ironies and misunderstandings are just too stark to pass them by without comment.


Comments are closed.