Wednesday, December 16, 2009

Open Access in 2009: The Good, the Bad and the Ugly

As 2009 draws to a close advocates of Open Access (OA) will doubtless be looking back and weighing up the year's events. So what has been achieved, and what have been the main OA developments in 2009? Has it been a good year or a bad year for OA? Let's consider these questions.

Green OA/Self-archiving

First, what has 2009 been like for Green OA?

There are now 137 self-archiving mandates. OA advocate Stevan Harnad has estimated that 10,000 will be needed before OA is achieved. (Although he later added that a tipping point would be reached if the top 750 to 1,500 institutions introduced mandates).

The good news

So what's the good news? Mandates are currently accelerating at a fast rate (See Alma Swan's graph). They are also beginning to arrive in bunches: during the second week of December, for instance, 26 Finnish mandates were announced in one go.

Meanwhile, in the US legislators appear to be coming around to the idea that they have a responsibility to ensure that federally-funded researchers are compelled to make their scholarly papers freely available on the Internet. In June, for instance, the Federal Research Public Access Act (FRPAA) was reintroduced to the US Senate with this end in mind.

And then in December the Obama Administration announced that it was launching a public consultation process preparatory to requiring US federal science and technology funding agencies to introduce "public access" policies.

What about repositories? OpenDOAR currently lists over 1,500 repositories, and Scientific Commons — the repository aggregator service — has over 32 million items listed in its database.

Importantly, more and more researchers are now aware of the OA movement and increasingly inclined to embrace it. (Although many remain confused as to what exactly they should be doing, and fearful that OA could damage their careers).

The bad news

And the bad news? There remains some doubt as to the efficacy of the mandates being introduced. Indeed, it is not even clear that many of the claimed mandates amount to much more than pious declarations.

Certainly many mandates appear not to be what it says on the packet. Even the much lauded Harvard mandate has not borne close scrutiny, and in the wake of criticism its architect Stuart Shieber conceded on his blog: "[T]he Harvard open-access policy could not be, should not be, and is not a mandate. I've tried to be very careful never to refer to it as a mandate (though I can’t promise I've never slipped up)."

This clearly has implications for compliance, and indeed another area of concern relates to the failure of both researchers and publishers to abide by mandates. Canadian OA advocate Heather Morrison, for instance, recently published some figures on the National Institutes of Health (NIH) mandate. Compliance levels for some Springer and Elsevier journals, she reports, are as low as 9% to 37%.

Overall, says Morrison, "NIH-funded research that is freely available after 2 years is 41% for external researchers and 50% for internal researchers. This is definitely growing; but the figure should be 100% at 1 year."

There are also worries about the nature and quality of some of the content being deposited in repositories, much of which appears not to be OA's target content (peer-reviewed papers). Certainly a search on Scientific Commons reveals broken links, non-target content, abstracts rather than full text, and some links merely point to items locked behind toll walls.

Finally, there are signs that increasing take-up of Gold OA could be being achieved at the expense of Green OA. This could slow down adoption of self-archiving mandates, and thus of OA progress overall. (Assuming, like Harnad, one believes that Green OA offers a faster route to OA than Gold OA does).

Gold OA/Publishing

What about Gold OA?

There are now over 4,500 Gold journals. As there are estimated to be roughly around 25,000 peer-reviewed journals in total, this suggests that one fifth of all scholarly journals are now OA.

The good news

What has been the good news for Gold OA? Again, the rate of progress of Gold OA is clearly accelerating. The Directory of Open Access Journals added 700 journals to its database in 2009. This, points out Morrison, is equivalent to two journals a day.

In addition, most subscription journals now offer a Hybrid OA option for some or all their journals, enabling authors to pay to have individual papers made OA even when publishing in a subscription journal. And a growing number of funders and research institutions are creating Gold OA funds to support authors who wish to pay the article processing charges (APCs) needed to make their papers OA.

Finally, in January the Houghton Report (Economic Implications of Alternative Scholarly Publishing Models: Exploring the costs and benefits) concluded that the research community could make considerable cost savings if all scholarly journals converted to OA.

The bad news

The bad news? Today's 4,500 Gold OA journals are not the best fifth. Harnad explains: "A high proportion of Gold OA journals are lesser journals. I don't want to make it sound elitist, but they are not the journals that contain the research that everybody wants and needs the most. If you look at the top journals, the ones that are likely to capture 80% of citations, most of those are not Gold OA. (I'm not talking about the hybrid Gold option, which is a cheap option for publishers, but a pricey one for authors.) So if you are talking about a tipping point for Gold OA, it will be necessary to capture the top journals."

The problem, however, may be that high-impact subscription journals are the least likely to convert to Gold OA.

What has clearly not helped Gold OA is the fact that a growing number of start-up OA publishers have attracted criticism from the research community for their poor business practices — including bombarding researchers with spam invitations to submit papers, and providing an inadequate peer review process. This, it seems, is the ugly side of OA — a product perhaps of what Harnad calls "gold fever".

But perhaps the most serious long-term problem is that it is looking increasingly possible that OA publishing may fail to solve the affordability problem — as many had hoped, particularly those librarians suffering the pain of the serials crisis.

In other words, despite the predictions made in the Houghton Report, the research community may discover that OA delivers no cost savings, and simply replaces the serials crisis with an author crisis. (We should note that the APC rates levied by some OA publishers — and most Hybrid publishers — are already higher than the Houghton estimates, and rising).

The growing trend for publishers to offer institutional membership schemes to allow institutions to convert from a subscription model to an OA model has also raised concerns about pricing. (See, for instance, the experience of the Health Sciences Library at Columbia University, and criticism of the SCOAP3 and Springer experiments).

OA broadens

There was, however, some unexpected good news in 2009: where OA has historically been seen as something of relevance only to scientific and medical (STM) journals this year has seen growing interest in OA for the humanities, and for books.

"I think there will be an increasing movement to scholarly open access monographs," says Australian OA advocate Colin Steele. "In Australia we have recently seen the establishment of the Adelaide University E-Press (with a stirring speech on their website from John Coetzee, the Nobel Prize winner). A worldwide movement in this context, reflected in developments in the US (see for example the upcoming SPARC-ACRL conference) and Bloomsbury Academic in the UK, seems likely for the transformation of the scholarly monograph situation. We must remember books and that OA is not simply STM articles."

Publishing consultant Alma Swan agrees. "OA monographs are finding their feet now, with experiments by presses (even commercial ones), establishment of viable business propositions (e.g. Open Humanities Press) and a number of formal studies going on including OAPEN, which I think may produce some useful pointers as it works through to its conclusion. It's never been nice to have to keep waving humanities to one side for another day, but now the day is here, it seems, and that is a real step forward."

Whether overall 2009 will turn out to have been good or bad for the OA movement doubtless rests on what happens in 2010 and beyond. Will current momentum be maintained for instance? Will the Obama Administration come up with the right formula for the public access policies it wants to see, or will those that emerge simply ape the NIH mandate (which has its detractors). Will more governments realise what needs to be done, and do it? Will OA continue to spread behind STM journals? Will OA find a way of resolving the affordability problem? For now these are necessarily matters for speculation alone.

Key developments

Below is a list of 17 notable developments that took place during 2009 (in date order). Thanks to Leo Waaijers, Gavin Baker, Alma Swan, Colin Steele, David Prosser, Peter Suber, and Heather Joseph for their input to the list:

January 21st: University of California libraries join the Springer experiment to support open access publishing. As the press release put it, "Under the terms of the agreement, articles by UC-affiliated authors accepted for publication in a Springer journal beginning in 2009 will be published using Springer Open Choice with full and immediate open access. There will be no separate per-article charges, since costs have been factored into the overall license."

Essentially this is an agreement designed to convert UC from a subscription to an OA world. Whether it will prove successful will depend not just on how it develops, but how one judges success.

January 27th: The Houghton Report — Economic Implications of Alternative Scholarly Publishing Models: Exploring the costs and benefits — is published. This concludes that the UK higher education sector could save around £80 million a year by shifting from toll access to open access publishing. It also estimates that £115 million could be saved by moving from toll access to open access self-archiving. If one projects these figures internationally it would appear that, in theory at least, OA offers the research community significant savings.

February 25th: SciELO South Africa is launched, with the objective of providing free and open access to a range of top South African science journals in order to boost the international profile of South African scientific research.

In other words, South Africa has joined the Scientific Electronic Library Online (SciELO) initiative, which was pioneered in 1997 by the São Paulo Science Foundation (FAPESP) and the Latin America and Caribbean Center on Health Sciences Information (BIREME) in Brazil.

SciELO provides "a common methodology for the preparation, storage, dissemination and evaluation of scientific literature in electronic format" and it does so on an entirely OA basis. Currently SciELO hosts over 600 journals published in the developing world.

April 1st: The Directory of Open Access Journals (DOAJ) and e-Depot of KB Netherlands sign an agreement to ensure the long term preservation of DOAJ journals. The press release explains: "[L]ong term archiving of the journals listed in the DOAJ at KB’s e-Depot will become an integral part of the service provided by the DOAJ."

June 10th: Cornell postgraduate student Phil Davis reports on The Scholarly Kitchen blog that The Open Information Science Journal (published by Bentham Open) accepted a nonsense paper he had generated using a computer program. Davis had submitted the paper after being bombarded with spam messages from Bentham inviting him to publish a paper, at a cost of $800. The editor-in-chief of TOISJ Bambang Parmanto resigns as a result — announcing that he had not even seen the paper before it was accepted.

June 10th: US National student organisations call for Open Access to research, and release a statement on OA. Comments Open Access News' Gavin Baker: "This is the first broad-scale support for OA from student organisations."

June 12th: The Open Access Source Book (OASIS) is launched. OASIS is a new portal for educational materials on the "concept, principles, advantages, approaches and means to achieving Open Access". In particular, OASIS hopes to provide valuable information on OA to those in the developing world.

June 25th: The Federal Research Public Access Act (FRPAA, S.1373) is reintroduced in the US Senate. Commenting on the news de facto leader of the OA movement Peter Suber said: "This is an important development. FRPAA would essentially extend the NIH open-access policy across the federal government. Most federally-funded researchers would be required to deposit their peer reviewed manuscripts in a suitable open access repository at the time of acceptance for publication, and the repositories would be required to release the open access copies no later than six months after publication."

August 7th: The Open Humanities Press announces that it is partnering with the University of Michigan Library's Scholarly Publishing Office to produce Open Access books. Coupled with last year's launch of Bloomsbury Academic, many view this as a clear sign that OA is broadening beyond STM journals to encompass books, and the humanities too.

August 21st: PLoS Currents is launched. The goal of PLoS Currents is to provide a service where researchers can submit raw preprints, datasets and preliminary analyses. These are then made freely available online without first being subjected to in-depth peer review, allowing for the much more rapid dissemination of research.

Introducing the service PLoS co-founder and chairman Harold Varmus commented: "The successful development of open access publishing by organisations including the Public Library of Science (PLoS) in recent years is a dramatic illustration of how the Internet is revolutionising scientific communication. Today, after several months of work, I'm delighted to announce that PLoS is launching PLoS Currents (Beta) – a new and experimental website for the rapid communication of research results and ideas. In response to the recent worldwide H1N1 influenza outbreak, the first PLoS Currents research theme is influenza."

In many ways, it is suggested, this is the realisation of a dream Varmus first had when he was NIH director: In 1999 Varmus attempted to create a biomedical preprint server modelled on the physics preprint server arXiv. (Following intense criticism from publishers the project was significantly scaled back, the preprint component jettisoned, and the service launched in 2000 as PubMed Central).

September 15th: The Compact for Open Access Publishing (COPE) is launched. The brainchild of Harvard's Stuart Shieber, COPE founder members include Cornell, Dartmouth, Harvard, the Massachusetts Institute of Technology, and the University of California, Berkeley, who jointly make a commitment to a "compact for open-access publication". Subsequently The University of Ottawa and Columbia University join COPE.

Commenting on the launch of COPE Shieber said: "Universities and funding agencies ought to provide equitable support for open-access publishing by subsidising the processing fees that faculty incur when contributing to open-access publications. Right now, these fees are relatively rare. But if the research community supports open-access publishing and it gains in importance as we believe that it will, those fees could aggregate substantially over time. The compact ensures that support is available to eliminate these processing fees as a disincentive to open-access publishing."

Essentially COPE members are asked to commit to create a Gold OA fund for their researchers in order to help pay APCs to publish in OA journals.

September 23rd: Enabling Open Scholarship (EOS) is launched. Headed up by Bernard Rentier, the Rector of the University of Liège in Belgium, EOS is an information service and a forum committed to the "opening up of scholarship and research" in line with "the growing open access, open education, open science and open innovation movements."

Commented EOS Chairman Professor Rentier: "The world of research is changing and universities and other research-based institutions must drive the change, not sit back and let it happen. Having embarked upon implementing changes in thinking and practice at my own university, I want to encourage others in my position to join the discussion and help lead the way to a better future. We will be reaching out to universities and research institutes across the world to invite them to play an active role in building better systems of scholarship for the future. EOS will provide the forum and the voice for the research community on open scholarship issues and represents a very valuable resource for those who want to join in this endeavour."

EOS appears primarily focused on Green rather than Gold OA. As such, it may offer a valuable counterweight to the Open Access Scholarly Publishers Association (OASPA), which has been criticised for being more interested in promoting the interests of publishers than OA.

October 19th: The Wellcome Trust announces that it is committing an additional £2 million to its Gold OA fund. The announcement is made during Open Access Week.

October 21st: The Confederation of Open Access Repositories (COAR) is launched. Emerging from the European project DRIVER (Digital Repository Infrastructure Vision for European Research), which was funded by the EU Commission, the objective of COAR is to network "over 1,000 global scientific repositories comprising peer reviewed publications under the principle of Open Access."

November 9th: A petition is launched in Germany calling on the German National Parliament (Bundestag) to mandate OA for publicly-funded research. The initiative of a former chemist and now science journalist from Heidelberg called Lars Fischer, the petition catches the OA movement by surprise, and Fischer is criticised for the vague wording of the petition.

Speaking to me in November Fischer responded by saying that he had had difficulty obtaining help from OA advocates. He added that he had launched the petition "to demonstrate that there is broad public support for Open Access and to promote open debate about intellectual property laws in science. As far as I see it, Open Access has always been treated — even by its supporters — as a niche topic for experts. But that is wrong. It is an issue that in the long run concerns everyone, and many people understand that."

November 17th: It is reported that the Dutch science foundation has created a €5 million Gold OA fund.

December 9th: Last but far from least, as part of its Open Government Directive the Obama Administration launches a public consultation process preparatory to developing public access policies for science and technology funding agencies across the US Federal Government. Comments can be posted on the OSTP Blog, and must be received by 7th January.

This is undeniably potentially very big. But where does it fit with FRPAA? Does it perhaps make the FRPAA moot?

No, says Heather Joseph, executive director of the Scholarly Publishing and Academic Resources Coalition (SPARC). "If the Obama Administration takes action on the executive level, FRPAA would play an important role in codifying that action into law. FRPAA would be complementary to any executive order or letter that the Administration might issue."

Suber agrees: "The Obama policies could cover much the same territory as FRPAA, depending on what the public comments recommend and how the administration decides to interpret them and act. The policies could fall short of FRPAA or surpass it. But at best they still won't make FRPAA moot. The main reason is that Obama could only implement OA mandates by executive order, which could be reversed by the next president. We'd still want legislation to make the policies permanent. But whether the Obama policies exceed or fall short of FRPAA, they can be implemented immediately, while FRPAA waits its turn for legislative attention behind healthcare reform, financial regulation, and climate change."

The above list is in date order, but I wonder what OA advocates might feel to be the top five in order of importance? Have any important events been omitted? All comments welcomed.