Monday, 22 October 2012

Is Open Access for Free Too Much to Ask?

Google is now the second most valuable IT company. They made the bulk of their fortune by providing fast and accurate internet searches for free and are funded almost entirely by advertising. If you had told me this would happen a decade or two ago I would’ve thought it was ludicrous! Similarly, the 6th most visited website on the internet is an encyclopedia, called Wikipedia. Wikipedia is written entirely by amateurs and volunteers and is funded entirely by donations. This would have also seemed crazy a decade ago. Wikipedia is supported by a non-profit organisation called the Wikimedia Foundation that employs just 50 people.

Yet this IS the world we live in. Google and the Wikimedia Foundation are remarkably successful and influential businesses. They show that unusual business models can be remarkably successful. However, the major academic publishing houses have languished. Large publishing houses continue to lock vital medical, basic science and engineering literature behind paywalls. A few, major publishers provide a variety of author-pays open access (OA) models. The costs of which range from $300 to $5000 USD, even with a strong NZ dollar, the average cost is equivalent to a few Summer Scholarships (to test wild research ideas), a new computer or 60GB of next-gen sequencing data, at current costs (this is roughly 20 human genomes-worth of sequence data). Some, rare, publishers make all their articles open access (with the author’s permission) after 1-2 years.
A few new publishing models have been proposed. One of particular interest is the child of major German, UK and US funding agencies. These are The Max Planck Society (a publicly funded NGO named after theoretical physicist, Max Planck), The Wellcome Trust (founded by pharmaceutical magnate, Sir Henry Wellcome in 1936) and The Howard Hughes Medical Institute (founded by businessman, Howard Hughes in 1953). One can only assume that these charities have become tired of their donations being used to line the pockets of publishers. Therefore, in a cost cutting exercise, they have launched their own journal, eLife. A new open access journal, that initially is experimenting with free OA publishing. The first edition of the journal was released this week!
Other models include a hybrid of traditional publishing and preprint archiving pioneered by PeerJ, with a very reasonably priced Lifetime Subscription model. While on the subject of preprints, there is also the free (physics) preprint archive epitomised by I’ve recently converted to using and have been very impressed by the near instantaneous indexing by GoogleScholar. Also, is ranked very well by GoogleScholar’s H5-index. This appears to be a great option for freeing your research, if your field is eligible (thank you qBIO).
As a follower of Impact Factors and other (better) measures of journal quality I’m not a fan of new journals. Personally I think there are already too many journals. However, the eLife model is so novel and has the backing three of the most powerful funding agencies in the world, therefore they may have the traction to build a successful new journal.
So, what about the other publishing groups? What fees do they charge for OA? Are any providing good value for money? To investigate this I have obtained OA fee ranges from a probably biased table prepared by BMC. Then for each of these publishers I have used the ISI Web of Science(TM) database to look up the range of impact factors for each publisher’s top 5 journals. See figure 1 for a visualisation of this data.

Figure 1: The figure on the left shows the range of OA costs charged by each publishing house. The fi gure on the right shows the Impact Factor range for each publishing house’s top 5 journals.
Then I became curious: Which publishers are providing the best OA deals in terms of dollars per impact factor point? (figure 2). Ignoring eLife for now, Wiley-Blackwell (W-B) may be charging $18.18USD/IF if one can really publish in the insanely high-impact (101.78) journal, “CA: A Cancer Journal for Clinicians” for $1850USD. However, checking the guidelines for this journal, I found that the only OA option costs $3000USD. This drops W-B to 5th place. Next is the American Chemical Society (ACS), with a potential charge of $24.88USD/IF if one can publish in Chemical Reviews (IF:40.197) for $1000USD. This option is available to ACS Members and Affliated Subscribers. Dues are $148.00 per year. This is surprisingly reasonable deal for a publisher with a history of a strong anti-OA stance.
Figure 2: This figure shows the potential range of cost/IF-point for each publisher’s top 5 journals.
Now, the other end of the spectrum. Who is providing the worst deal? This is a difficult question to answer: almost all the publishers support journals with impact factors near zero that charge for OA. Consequently, any monetary value divided by a small IF  results in a large value. However, let’s look at the worst deals in the data I have. The list is topped by America’s National Academy  of Science (which only publishes 3 journals indexed by ISI). “Transportation Research Record” (IF:0.471), as far as I can tell from their website, has no open access policy at all. That is no OA deal at all. Next down the list is “mBio” (IF:5.3) which, according to  BMC’s table, may be charging $3285USD for OA publishing. Checking their website ASM members can publish for $2000USD, non-members for $3000USD. ASM membership costs $50, this is probably a worthwhile investment. $2050/5.3 drops mBio to 7th worst OA deal on my list. Next on my hitlist is Hindawi–a bunch of academic spammers if my inbox is anything to go by. Hindawi are potentially charging $1500USD to publish in “Journal of Biomedicine and Biotechnology” (IF:2.436). A quick trip to their website confirms this is the case. We finally have a winner for the worst deal award! Next down the list is, surprisingly, PLoS. PLoS’ 5-th ranked journal is “PLoS Computational Biology” (IF:5.215). According to their fee page, publication for research in middle to high income country incurs a fee of $2250USD. This is cheaper than anticipated so PLoS moves down to 8th worst slot. Phew! Anthony Poole and I have just had an article accepted in PLoS CB. I think I’ll finish this tiresome game there.

I have been terribly unfair to the publishers I have mentioned here (and probably the ones I haven’t). The rules of my game have been rather arbitrary. If I was to do this fairly I would survey a large number of academics from a number of disciplines to find out what OA fees they are really paying in which journals. I haven’t the time to do this, however, this is something that could be added to the next SOAP initiative.
Another issue I haven’t discussed is copyright. Some of the publishers still retain the copyright on OA articles, others do not. This topic is covered by other blogposts in the series
In summary, open access is great but can be extremely expensive. Not all publishers are equal, therefore, it is worth shopping around. Preprint archives can provide a nice intermediate solution. Finally, please buy Paul’s Patented Cognitive Enhancement Vitamin Formula, produced in association with Placeboceuticals.
Conflicts of interest:

1. I am an Assistant Editor in Chief for the Landes Bioscience journal, RNA Biology. I regularly invite
contributions and referee articles for the journal. I receive no salary for this position. They did send me an iPad nearly 2 years ago after I cheekily asked for one when they advertised their new iPad App. This was very nice, my kids regularly use it for watching YouTube clips about ”Lego” and ”Thomas the Tank Engine”. I’m sometimes permitted to use it for email, Facebook, Twitter and as a Kindle.

2. I was funded by the Wellcome Trust for 4 years. They were wonderful, my contract specified that all my articles must be deposited in UKPMC within 6 months of publication. They happily paid the OA fees when I managed to publish my work.

Abbreviations used in the above figures: T&F=Taylor & Francis, ACS=American Chemical Society, NAS=National Academy of Science, PLoS-Public Library of Science, BMC=BioMed Central, W-B=Wiley-Blackwell, BMJ=British Medical Journal, CUP=Cambridge University Press, OUP=Oxford University Press, NPG=Nature Publishing Group.

A version of this article is cross posted on the NZ Creative Commons Blog

Open Access Week, 2012

Together with a number of other open access advocates, I have written a blog post for Open Access week, 2012.  To read it and some other fantastic posts from my colleagues visit the NZ Creative Commons website.