Morning Ramble: The Open-Access "Debate"

A debate that has animated the scientific-publishing world for the past five years or so -- the push by an activist group to promote "open access" to the primary scientific and particularly biomedical literature -- has reached its proverbial tipping point, with the government's proposal to "recommend" that all papers based on NIH-funded research become freely available on the institutes' public archive, PubMed Central, within six months of publication. Numerous press outlets picked up the story -- or, more precisely, picked up and re-ran the highly favorable Washington Post report of the story, which emphasized the righteousness of the NIH cause, the need to empower the huddled masses yearning to read the latest paper on cellular signal transduction, and the callow and panicked response of the venal scientific publishing industry.

All right -- I exaggerate. In any event, the handwriting, as they say, is on the wall: With the recent NIH proposal, congressional support, and the issue's superficial "apple pie and motherhood" appeal, a move to require some kind of open access for NIH-funded scientific research has become pretty much inevitable. Moreover, if the NIH plan is adopted, it's unlikely to remain static -- it's only a matter of time before it's "recommended" that publishers make papers freely available not six months after publication, but three months after, then three weeks, then one -- and, ultimately, insist on free access immediately upon publication. Nor will the move remain confined to biomedicine: sooner or later, other government funding bodies such as NSF will start to ask just why papers reporting the research they fund shouldn't be free to the public as well. The result will be a sea change in the way scientific results are published.

As a "good liberal," I suppose I should be in favor of this. Still, there are two sides to every story, and it seems that the press may be too quick to dismiss the potentially damaging effects of the blunt instrument the NIH proposes to use to "solve" a real problem. I wish that I had more time to devote to this than this brief train ride to work, but a few points seem worth making here. (I would also counsel the reader to trust none of what I say: I have a vested interest in the current system, and am thus merely another mindless tool of a cynical, money-grubbing, albeit nonprofit, scientific publisher.)

First, I'd stress that there is a real problem here; academic libraries and similar institutions have seen subscription costs expand significantly in recent years. A large part of this, however, traces to the activities of a few large, for-profit aggregators that, by acquiring control of large numbers of scientific journals, have amassed quasi-monopolistic power, and have used that market power to boost subscription revenues. Thus an argument could be made (at least from my vantage point) that, rather than dismantling a publishing system that has worked for many years, society might be better served with antitrust action against this kind of abuse of market power.

Second, we need to ask ourselves who is really likely to benefit here, and how much. One ironic thing about this debate, to me, is that access to scientific literature is now more open than it has ever been before. Working scientists now can call up an unprecedented range of primary research from their computer desktops -- usually for free (at least to them), owing to the prevalence of online library subscriptions. Articles such as the Post item linked above invariably, and anecdotally, cite the desire of patients and the public to read the primary literature. But are most people even going understand that literature? Here's an example of the kind of thing the public is missing under the current "closed" system:

Hh signal transduction is unusual in that the unbound receptor (Ptc) is the active form (that is, it keeps the pathway switched 'off' by inhibiting the transducer (Smo)). Binding by the ligand (Hh) inactivates the receptor, releasing Smo from inhibition and turning the pathway on. The cell's perception of the amount of ambient Hh has been proposed to be determined solely by the number of unliganded (active) Ptc molecules. Thus, as Hh rises from nil to peak concentrations, Smo activity would increase merely as a consequence of the progressive depletion of the pool of active Ptc protein. According to this depletion model, liganded Ptc would be functionally equivalent to the absence of Ptc. Alternatively, liganded Ptc might titrate the inhibitory activity of unliganded Ptc so that a cell's perception of Hh concentration would depend on the ratio of the two forms.

Granted, less arcane material exists in the scientific literature (as does more arcane material). The point is that, arguably, what the public needs is not better access to primary scientific research; it needs better science journalism, better plain-language explanations, better value-added material to put scientific work into context -- and these are not things it will get from the open-access movement. (Indeed, to the extent that the move toward making research papers freely available puts pressure on the bottom lines of marginal players, it could limit those players' efforts/ability to devote scarce resources to publicity, press relations, and explanatory material, thereby limiting rather than broadening public access to these important gateways to the scientific world.) In light of the library crisis mentioned above, there may indeed be excellent arguments for some kind of open access. But for the sake of intellectual honesty, proponents of the idea should stop fronting the emotionally appealing but specious arguments about the public masses clamoring for instant desktop access to the latest papers on NF-κB.

Third, I think the argument of small scientific societies that they depend on their publishing revenues to fund their other operations shouldn't be tossed to the side in quite so cavalier a fashion as some in the OA community are wont to do. Many of these small nonprofits do fine work for their constituencies, but they operate on the financial razor's edge -- and arguably don't have the luxury of experimenting with an untested business model. (And make no mistake -- author-pays open access is an untested model. Remember that PLOS Biology, often cited as the shining example of open-access publishing "success," still enjoys the financial advantage of huge grants from the Gordon and Betty Moore Foundation and the Open Society Institute, and others.)

Fourth, open access (under an author-pays framework) seems unlikely to work outside of the exalted realms of biomedical research. There are many ironies in the open-access movement; one of the richest is that the arguments for free access are pushed most forcefully by privileged scientific castes in biomedicine whose funding is secure, who can attract large grants, and who are no longer in the stage of trying to build a career with limited funding that is premised on developing a publication record in top-tier, high-quality journals. For a front-rank researcher in molecular biology, pushing open access has no real risks; for workers in more marginal areas with limited funding, author-pays open access could effectively foreclose the opportunity to publish in first-rank journals -- which, the free market being what it is, would likely command the highest author fees for publication.

There is actually much, much more to this issue than I have time to write about now -- and I fear that I haven't really expressed myself well in what I have written. Time to shut up before doing myself any more damage.