Socialism in science – Part 2 – I am wiser but not quite convinced
In my previous post I made a bold statement comparing Open Access publishing to socialism. The post was criticized in a number of insightful comments, but hey, I knew I had it coming. There are a lot of people very strongly emotionally attached to OA, and a voice pointing out the weaknesses of this publishing model was bound to cause an uproar. I am glad, because thanks to all the criticisms I heard I have been able to learn a great deal and to see the problem from a number of different angles. But, call it stubornness or whatever, I am still not convinced that OA publishing is on its way to take over the world of high impact scientific publishing. In fact, the extra mile the comments on the blog made me walk resulted in my being more set in my opinion that OA is not the way to go for top-tier biomedical research journals.
First, allow me to explain one thing. My post was not an advocacy piece, but an opinion piece. I am all for open dissemination of knowledge as long as it doesn’t interfere with the quality of the final product, ie. the research paper itself. I come from a country where access to the latest research literature was very limited, so nobody needs to tell me about how much difference OA would make for poorer scientific communities. That said, I don’t really see OA as a viable option for addressing all the problems with knowledge dissemination. I am also far from proclaiming that OA is dead, but I do think that it will not play a significant role in the prestigious high-impact journal area.
After that disclaimer, let me address some of the comments that I got to my previous post. I think Peter Suber made a few very important points. First, he says that most OA journals don’t charge author-side publication fees and that there are plenty of OA journal business models. I must agree with both these statements, but retort: “What difference does it make?”. Three major players in the OA field: Hindawi, BMC and PLoS all charge publication fees and neither uses any of the more exotic business models pointed out by Peter. What it probably means is that theirs are the only models that really work on a larger scale. I mean, OA has been around for a while, you would think that publishers would have come up with a better way to make money if there was one.
Peter also pointed out that BMC is actually making a profit and so it is possible to combine profitability for the publisher with the greater good of universal access. However, this profitability comes as a cost, and in this case it is the cost of lower impact. Again, let’s look at the three major players: PLoS, BMC, and Hindawi. The first one that reported profitability was Hindawi, but virtually all of their journals are super-low on the IF scale. The allegedly profitable BioMed Central has very few journals above IF 5, and only one above 6. They however, took over 9 years to turn a profit and I am guessing it is not exactly a cash cow for Springer. PLoS, on the other hand, is still bringing lossess despite huge endowments and a “bulk publishing” supporting e-journal PLoS ONE. Their journals, however, are very high impact and PLoS is sweating to keep them that way. I am not buying the argument that it takes time for a journal to gain impact, because PLoS Pathogens is a little over 3 years old and it sports a mighty IF of 9.3, while the flagship journal of BMC, Genome Biology, IF 6.59, is 9 years old. Also confirming the “high-impact=low OA profitability” hypothesis is the fact that two highly prestigious journals, British Medical Journal and Journal of Clinical Investigation, after flirting with OA for a while, reverted back to CA because they couldn’t make ends meet.
Peter, and especially Stevan Harnad brought up another issue. They revealed a seemingly fool-proof scheme to go around the limitations of OA journals: the OA archives. The scheme goes as follows: if all the authors agreed (or were more or less forcefully incited) to submit their final post-referee drafts of papers to their institutional archives, which is typically allowed by copyright policies of most journals, we could link all these archives together using a common search engine and circumvent the access barriers that the publishers set before us without recoursing to OA journals per se. While the system is still in the works, I don’t see a reason why it should not work out from a technical standpoint. However, this idea is still missing the main point of my discourse: these papers still must be reviewed and the review process still costs a lot of money, especially for high-quality journals. Why a publisher would incur all these costs just to see the final product of their hard work distributed for free in an OA archive is beyond me. The reason why publishers are lax with their post-publication archiving policies is precisely because the inter-archive search system is not yet up and running, and so it is almost impossible to find a paper you need online without subscribing to the journal. Once this system becomes operational, two possible scenarios are possible: (1) the publishers put their collective foot down and make post-print archiving policies much stricter, in effect bringing the situation back to where it is now, or (2) the powers that be prohibit the publishers from doing (1) and so the publishers pack their things and get out of the scientific journal business because it is not bringing them the bucks they expect from it. That, trust me, is not in the best interest of the scientific community. The 2003 Wellcome Trust Report on scientific publishing, which many OA enthusiasts bear almost like a flag before them without actually taking the time to read it, actually says something very similar, and gives an example of a very restrictive archive system in the Netherlands (see point 4.28) which is enough of a pain in the ass to use that most people will still want to get a subscription, but which, if need be, provides open access to all published manuscripts.
Finally, Anders said something I have a hard time agreeing with. He proposes that CA publishers are monopolies and that they are “government-enforced”. I hope that by now my dear readers see clearly that there is no need to enforce the CA publishing model – it is the one that works out quite well for all the players in the market. On the contrary, introduction of OA to top-tier scientific publishing would require quite a bit of enforcement and would probably end in a massive disaster, as many experiments not taking common sense into account do. Also, scientific publishing is an extremely complex market, as the Wellcome Trust Report analysis makes clear (3.41-3.44), and while it is characterized by price-inelasticity of demand, which puts publishers in a position of power over subscribers, it also puts a lot of competitive pressure on the publishers to deliver highest-quality product (papers) and services (revision/editing), which is clearly in the best interest of the scientific community. Funding agencies and governments can and do exert quite a lot of pressure on the publishers in order to prevent them from abusing their privileged position in relation to libraries. I wholeheartedly support these actions, as well as antitrust investigations of major mergers in the scientific publishing field, which might result in a true monopoly, but I would advise caution when squeezing the commercial publishers’ profits, lest they might leave the field altogether and leave us, the scientist, with a big void to fill.
Thus, it is with sadness that I must conclude I am still skeptical about the chances of the diverse OA initiatives to completely substitute for the current CA publishing models. Call me doubting Thomas, but I want to see proofs of the feasibility of profitable high-impact open access publishing before I jump on the bandwagon and sing songs in praise of OA. All I can see right now is a lot of unfounded hope and plenty of examples to the contrary. Convince me if you can.
Entry filed under: Computers in science, Philosophy of science, Scientific publishing. Tags: BioMed Central, BMC, Hindawi, open access, open access archives, PLOS, research paper, science 2.0, scientific publishing, university libraries, Wellcome Trust.