The conflict between tobacco behemoth Philip Morris International and the University of Stirling (and I agree with Heather Brooke that information should be free for businesses too) is the latest in a string of clashes over access to research and the Freedom of Information Act. Professor Mike Baillie, a dendrochronologist, was ordered to hand over data to a city banker and climate crank, while UEA staff were caught out dealing rather unprofessionally with requests from conspiracy theorists.
There's a clear link between these incidents and the complaints in recent days from George Monbiot and Ben Goldacre regarding the lack of open access to research papers. Monbiot suggested that academic publishers were "the most ruthless capitalists in the western world", while Goldacre highlighted recent cases of modern day "Robin Hood nerds", using their skills to 'liberate' papers for the public to read.
In theory the Act could be applied to paywalls as well as data1, but whether you're after papers or data, FOI requests are completely unsuitable as a means of access. If your goal is to free scientific data, then getting people to make requests for it under the Freedom of Information Act is, as effective strategies go, on a par with sending the Jedward twins out to kill a buffalo with a rudimentary flint spear.
Freedom of Information requests are slow, tedious, and as a means of gathering information roughly equal in efficiency to the game Battleships. You fire off a formal request, wait a month for a reply telling you your request was rubbish and off-target, move two squares to the left and fire again, and repeat until you get what you were after, or you've given up, or you've died of old age. Apply this to research data and there's a good chance that even if it's available it may be unintelligible, or in a proprietary format, or undocumented.
I don't want to get into a three-month formal e-mail exchange just so I can see some data for a paper. The only request I want to submit is an HTTP request. I want to be able to look up a paper on the internet, read it for free, and then click a button next to it to download the data that was used (where appropriate) in a standard format so I can have a play with it. It seems like a pretty basic thing to ask.
The problem is that peer-review is a privatized industry in which public interest is an externality. The public pay for raw research to be performed, but we don't pay for the peer-review or publishing necessary to turn it into the finished article - published research. Instead, academic journals are in the business of refining raw product and selling the result. In this case, the refined product is sold back to the research institutions who subscribe to it. Nobody pays for public access, and there's no great incentive for publishers to provide it.
Monbiot talks about liberating "the research that belongs to us," but part of the problem is that it doesn't belong to us, at least not in that refined form. It belongs to the companies who create it, a small number of whom have a strangehold on the industry. Elsevier run over 2,000 journals, while Elsevier, Springer and Wiley publish an Adamsian 42% of research papers. Academics have to read and publish to sustain their careers, and if you want to do either you have to deal with Big Journal.
The insanely high prices that Monbiot highlights - ("Reading a single article published by one of Elsevier's journals will cost you $31.50. Springer charges €34.95, Wiley-Blackwell, $42") - are just one of the problems caused by this unhealthy monopoly.
Ivan Oransky meticulously documents some of the others on his twin blogs, RetractionWatch (co-edited with Adam Marcus) and EmbargoWatch. Journals are often less than transparent about papers that are retracted post-publication due to problems or fraud, while embargoes have been criticized for their potential impact on science journalism - some suggesting that embargoes allow journals to "[control] the way scientists and the press conduct the business of public discourse."
On occasion multiple issues combine to form one giant clusterfuck, as happened recently with the journal "Molecular Biology and Evolution" who, brilliantly, published a retraction notice behind a paywall. Without paying, the public could read:
This article has…
But could pay $32 to see the full text:
This article has been permanently retracted from publication by the authors.
Genius.
Monbiot outlines a few possible solutions in his recent article, but some are better than others. A move to "insist that all papers arising from publicly funded research are placed in a free public database" seems to attack publishing without providing any constructive alternative.
Another suggestion is "a single global archive of academic literature and data," funded by the money saved from library subscriptions, but who would run it? Such an international institution would, I imagine, end up with a lot of politics involved, and it seems perverse to argue that peer review needs more competition, then suggest another 'benevolent' monopoly. Besides, if academics have a choice between publishing in Nature, or publishing in some new free archive, they will rightly choose the more prestigious outlet for their work.
On the other hand, I agree that "governments should refer the academic publishers to their competition watchdogs" to encourage a more open market. We don't necessarily need to dismantle the publishing industry, or nationalize peer-review, but we do need the government to accept that there's a problem. There ought to be some way to work with publishers to free science, and put it back in the public domain where it belongs.
- - - - - - - - - -
Note 1 Papers are certainly covered by the legislation, and I assume the response would hinge on whether papers that cost twenty-odd quid to get a copy of would be deemed 'reasonably accessible' by the information overlords (a key exemption test - if information requested is 'reasonably accessible' then you don't have to provide it). Another question is how many papers you could reasonably ask for - "everything your department has published since 1990" might be a bit excessive. Even if you could use the FOI to mass-order papers, I'm not sure what the rules on republishing them would be. I'd be interested to know if anyone has tried this, or knows more.(back to text)
- - - - - - - - - -
@mjrobbins | layscience@googlemail.com
Comments (…)
Sign in or create your Guardian account to join the discussion