Open access in 2005
SPARC Open Access Newsletter, issue #93
January 2, 2006
by Peter Suber
* 2005 was the best year to date for university actions in support of OA.  We saw major OA policies or resolutions in 2005 from (in alphabetical order) Case Western Reserve University, Columbia University, Cornell University, Indiana University - Purdue University at Indianapolis, Lund University, Oregon State University, University of Bielefeld, University of California at Berkeley, University of California at Santa Cruz, University of Kansas, University of North Carolina at Chapel Hill, and the University of Wisconson.  Before 2005, Queensland University of Technology, University of Minho, and the University of Southampton Department of ECS were alone in mandating OA to their research outputs, but in 2005 they were joined by CERN and the University of Zurich.  The Eprints Institutional Self-Archiving Policy Registry now lists 17 institutions with strong self-archiving policies, most of them added in 2005 as a result of the Berlin3 meeting.  The 19 major research universities in the UK's Russell Group endorsed OA in June, and all UK institutions did so through Universities UK three months later. 

* 2005 was the year that funding agency OA policies made the transition from proposal to practice.  The NIH policy took effect on May 2 and the Wellcome Trust policy took effect on October 1.  The RCUK policy won't take effect until 2006, but was drafted, released for public comment, and revised in 2005.  (See Predictions for 2006, below, for more.)  In 2004, the OECD recommended public access to publicly-funded research data and in 2005 extended the recommendation to research literature. 

The OA policies at universities and funding agencies are even more important together than they are individually.  Since only authors can make OA happen, it's very important that institutions in a position to enlighten, assist, or nudge authors wake up to their own interests in doing so.  We're now seeing this from employers and funders, the two institutions in the best position to influence author decisions.  It's equally important that the percentage of authors who would willingly comply with an OA mandate from their employer or funder, according to a Key Perspectives report in May 2005, is over 81% this year, up over 10 points from last year.

* It's no surprise, then, that we're seeing an increase in author knowledge of OA and author actions in pursuit of it.  In the same Key Perspectives report, Alma Swan and Sheridan Brown found that the rate of OA archiving in institutional repositories doubled since their 2004 survey, and that the rate of OA archiving in disciplinary repositories rose by 60%.  The CIBER team also found an increase in OA activity since last year.  Authors who said they knew "quite a lot" about OA rose 10 points over 2004, and authors who knew nothing at all about it dropped 25 points.  The percentage of authors who had published in OA journals rose 11 points to 29%. 

* OA archiving continued to worry some publishers, who fear that it will undermine subscriptions.  However, publishers have so far been unable to provide evidence that their fears are justified, and 2005 was the year in which we saw strong counter-evidence that archiving is either harmless or helpful to journals.  If OA archiving undermined subscriptions, the effect would show first or show most in physics, where OA archiving is most widespread and longstanding.  But Key Perspectives reported in May that the American Physical Society (APS) and the Institute of Physics Publishing Ltd (IOPP) were unable to identify any subscriptions lost in the 14 years of arXiv's existence.  Both publishers support OA archiving by accepting submissions directly from arXiv, which encourages authors to deposit their preprints there.  In 1999, the APS went so far as to help launch an arXiv mirror at the Brookhaven National Laboratory and in 2005 we learned that the IOPP is the process of launching an arXiv mirror of its own.  Publishers who fear OA archiving, but who have less experience with it than the APS and IOPP, have a new burden to meet if they want to show that it harms subscriptions, especially if they want to stop experiments, like archiving mandates from funders, that would generate new evidence in other fields.

* OA journals picked up speed in 2005.  PLoS actually launched more new journals in 2005 (three) than in 2004 (two).  Hindawi launched, acquired, or converted even more, moving into a leadership position in OA publishing with PLoS and BMC.  I haven't counted the number of subscription-based journals that converted to OA during 2005, but it's at least a dozen, and therefore at least as large as the number of conversions in all previous years combined (see postscript below).  The author-choice OA model made more gains in 2005 than in any previous year:  Blackwell introduced Online Open in February, Oxford introduced Oxford Open in May, and Springer strengthened its existing Open Choice model in October.  In December all three publishers struck a deal with the Wellcome Trust to provide immediate OA to articles based on WT-funded research (more in Top Stories below).  Springer even created a new position, Director of Open Access, for which it hired Jan Velterop, former publisher of BMC.  The American Institute of Physics Author Select program was announced in 2004 but published its first OA article in October 2005.  The number of publishers putting all or most of their emphasis on OA journals has grown meteorically.  Some are very small but their proliferation is a sign of fecundity.  Apart from BMC, PLoS, and Hindawi, we have (in alphabetical order) ADHO, Allied Academies,  Bepress, Copernicus, DiPP, ElectraPress, Flying Publisher, HSRC Press, ICAAP, Internet Scientific Publications, JMIR, Libertas Academica, Library Publishing Media, MedRounds Publications, MedKnow Publications, ODINPubAfrica, and Petroleum Journals Online.  If you count Highwire, then there's another heavyweight to add to the list.

PLoS and BMC earned some stunning impact factors in 2005; five BMC journals ranked in the top five journals in their specialties, and PLoS Biology, in its first year with an impact factor, scored a 13.9, making it the number one journal in the category of general biology.  BMC and PLoS are among the first to argue that impact factors are misused and misleading, but their own high scores show that journals don't need to charge subscriptions or demand the author's copyright in order to deliver high impact and high prestige.

* 2005 was the first year in which draft treaties explicitly mandated OA to publicly-funded research.  The Medical Research and Development Treaty was submitted to the World Health Organization in February and a draft of the Access to Knowledge (A2K) Treaty was released for the WIPO Development Agenda in May.  (Disclosure:  I participated in the drafting of each.)  Previously we had only the OECD ministerial-level agreement on OA to research data in 2004, and the WSIS Phase 1 documents in 2003.  If you plot these over time, the OA-related international agreements are becoming more substantive, more specific, and more wide-ranging.

* 2005 was another big year for OA to data.  One very active front was access to clinical drug trial data.  The International Committee of Medical Journal Editors (ICMJE) first made OA to clinical trial data a condition of publishing articles about clinical trials in 2004, but it strengthened and clarified its requirements in 2005.  The FACT Act (Fair Access to Clinical Trials Act), was introduced by Senators Grassley and Dodd in early 2005, and Senator Lieberman introduced the CURES Act last month; both would mandate OA to clinical trials data.  PLoS Clinical Trials issued its first call for papers.  Beyond clinical trials, more journals, like Nature, are now requiring OA to data files as a condition of publishing articles.  (It's noteworthy that Nature and the ICMJE journals are not OA; for them, mandating OA to data is a significant concession that OA is in the interests of science even if some publishers have to give top priority to the bottom line.)  While there were OA databases launched or lauded in nearly field of science, 2005 was a particularly big year for OA to geospatial data.  The Electronic Geophysical Year published the Declaration for a Geoscience Information Commons and the Open Knowledge Foundation Network published a manifesto calling for Open Access to State-Collected Geospatial Data.  OA geospatial tools like Google Earth provided material assistance after Hurricane Katrina and the Pakistan earthquake but also raised security alarms from a handful of countries worried about automated spying.  In genomics, the HapMap is the biggest OA breakthrough since the Human Genome Project.

* In 2005 we saw more profit-seeking (or "surplus"-seeking) businesses demand that the government stop providing public access to publicly-funded information.  This is a demand to profit at the expense of taxpayers or a demand that taxpayers pay twice in order to guarantee a revenue stream for a private-sector organization.  In support of their demand, the organizations argue that free information from the government is unfair competition, as if what they really wanted was a free market.  The leading example was the American Chemical Society's attempt to shut down or scale back PubChem, the OA database launched by the NIH.  Congress denied the ACS request to defund PubChem and now the ACS and NIH are working out the terms of their coexistence, with the NIH holding all the cards.  AccuWeather was at least as brazen as the ACS, wanting to repackage and sell government-collected weather data to the public without "competition" from the National Weather Service.  AccuWeather even paid Senator Rick Santorum (R-PA) to introduce a bill in Congress to protect the company's revenue stream from government OA.  I'm glad to say that the Santorum bill doesn't appear to be going anywhere.  You can criticize these demands as harmful in their substance and dishonest in their rhetoric, but you can't dismiss them as losing propositions.  In 2002 a trade association of publishers (Software & Information Industry Association) successfully used similar arguments to persuade Congress to kill PubScience.  These arguments might succeed again if we are not paying attention.

* 2005 saw OA projects take to wikis in order to collect, organize, and share their information.  Ari Friedman launched his self-archiving wiki in June and Arthur Sale launched his AuseAccess (Australian repositories) wiki in November.  Individual OA projects with wikis --not all of them open to the public-- include DigiWiki, DSpace, NDLTD, Ockham, Open Source Anthropology, Science Commons, the University of Maine Commons, Wex, Wikibooks, Wikilaw, and the WSIS Scientific Information Working Group.  The Open Business wiki collects business models for open content enterprises and includes some OA projects. 

We also saw the rise of mutant wikis designed to take the most distinctive feature of traditional wikis --open contributions from anyone-- and combine it with more rigorous forms of quality control and peer review.  Wex limits contributions to approved contributors.  Digital Universe will allow contributions from anyone but subject them to review by hand-picked experts.  Wikipedia is adopting a number of measures to limit headline-grabbing inaccuracies:  freezing a stable edition, not letting unregistered users start new pages, introducing time-delays before edits take effect on high-traffic pages.  OA doesn't at all depend on wiki-like ease of editing or wiki-like freedom from expert peer review.  But the more wikis incorporate serious quality controls --innovative or traditional-- they more they are likely to become important vehicles for OA scholarship.

* 2005 saw the rise of social indexing (social bookmarking, social tagging, folksonomies) for online scholarship.  The trend started with services like and Flickr, which were not specifically academic, but these quickly gave rise to CiteULike and Connotea, the two leading tools for making and sharing tags on works of web-based scholarship.  The Library Thing and Amazon allow online tagging of print books, and Google Base allows tagging of all its projects.  Folksonomy tagging services work best for OA content even if they can in principle apply to any kind of content.  They provide a layer of indexing beyond search-engine indexing; they cost the publisher nothing; they harness collective intelligence; they improve over time; and after a certain point they lend themselves to integration with the semantic web.  So their advent is significant as a low-cost and powerful way to enhance OA literature.  Like good search engines before them, they also undercut the argument that conventional indexing, which is still largely limited to conventional journals, is so unique and useful that it's a reason to publish in conventional journals.

* The discussion of OA has always favored journals, listservs, newsletters, and blogs.  In fact, the first print book on OA was probably Okerson and O'Donnell's, _Scholarly Journals at the Crossroads_ (ARL, 1995), which is a collection of listserv postings by and about Stevan Harnad's subversive proposal for OA archiving.  But in the 10 years since that collection appeared, OA discussion has made very few appearances in books.  In 2005, however, OA reappeared in books, in force.  John Willinsky published _The Access Principle_ (MIT Press) and Francis André published _Libre accès aux savoirs_ (Futuribles).  Anthologies on OA were announced by Chandos,  Middlebury, and CLACSO.  Polimetrica launched an entire book series on OA.  We saw a book-length study of author attitudes from Key Perspectives, a book-length report on OA journals by the Kaufman-Wills group, a book-length bibliography of OA by Charles W. Bailey, Jr., and more book-length studies of OA to data from the National Academies of Sciences.  New books on OA are in the works from at least three OA activists, including myself.

My take on the timing:  1995 was a good time for a book on the tantalizing possibility of OA.  But the next 10 years were full of development, accomplishment, adoption, and refinement.  Change was taking place too quickly for a book, but not too quickly for more bite-sized contributions to the conversation.  By 2005, if not a few years earlier, we had reached a plateau of what could be called understanding without consensus.  The concept of OA was widely implemented and no longer hypothetical or quixotic, even if OA was still far from the default method of dissemination.  The objections were well-answered even if they did not disappear.  There were enough supporters to assure continued momentum, even if there were also enough dissenters to assure continued contention.  The stakeholders who cared about it, either from hope or fear, were numerous enough to interest book publishers.

* Ignorance and misunderstanding have always been obstacles to OA, but for at least two years now a single misunderstanding has been a major player.  We need a name for the mistake of thinking that a mandate to deposit work in an OA repository is really a mandate to submit work to an OA journal.  Sometimes the mistake is to think that a mandate to deposit work in an OA repository is really a mandate to phase out subscription-based journals in favor of OA journals.  Let's call it the Journal-Archive Mixup (JAM).  In 2004 most publishers with an opinion about the NIH public-access policy made JAM in their responses to it, and were repeatedly corrected by OA activists and the NIH itself.  The UK government made JAM in response to the OA recommendations by House of Commons Science and Technology Committee and was repeatedly corrected by journalists, OA activists, and the House of Commons.  Both episodes should have driven a stake through heart of JAM, but they stopped it only temporarily.  JAM arose again in 2005, walked the land, and struck terror (or at least tried to strike terror) in the hearts of people who would actually benefit from OA archiving.  In 2005 JAM was seen at a handful of journal publishers, the UK Under-Secretary of State for Science and Innovation  (one of those criticized and corrected for making JAM the previous year), Members of Parliament like Ed Vaizey from Oxfordshire, and the leaders of the Royal Society.  The persistence of JAM is about careless reading, even if it's also about interest and strategy.  Fortunately, most of the JAM-makers don't have to read subtle manuscripts, understand them, and decide whether they are worthy of publication.  Unfortunately, they do have to read policies (some, like the RCUK draft, not subtle), understand them, and decide whether they are worthy of adoption.

* 2005 was definitely the biggest year to date for book scanning and digitization.  In fact, the book-scanning news, even when it was not about OA, swamped the OA news and persuaded many people that it really was about OA.  Google got more headlines than all the other projects combined but it was far from alone.  The other big projects in 2005 were the Open Content Alliance (including the Internet Archive, Yahoo, and Microsoft), the EU (i2010 Digital Libraries Project), the Library of Congress, Amazon, the Million Book Project, and some private scanning projects by individual book publishers.  Together, the projects drew attention to the value of indexing all information, the fuzziness of fair use, the evidence that free online full-text increases net sales of print editions (at least for some kinds of books), and the urgency of protecting the public domain from further shrinkage and encroachments.  They invited readers to dream about free online searching of all the world's book literature, which even critics acknowledge will be an unprecedented boon to scholarship and education.  They forced lawyers to acknowledge that copyright law had no ready classification for the act of copying full texts without permission but only using them to display fair-use excerpts.  They divided authors into those who thought that Google snippets were invaluable free advertising and those who thought they were piratical threats to profit and control.  Then they divided publishes into the same two camps.  They forced OA activists to acknowledge that free online book searching, and free even online book reading, might approach 100% faster than progress toward OA for journals.  (More in Predictions for 2006, below.)  They also brought access issues to a much larger public than OA activists could ever persuade to care about peer-reviewed journal literature.  If anyone still tried to look knowing at parties by dismissing the internet as spam, pornography, and teenage drivel, the book-scanning projects finally shut them up.

There were two big misunderstandings of Google's book-scanning projects that were individually bigger than JAM in their inexcusable sloppiness, even if smaller for hard-core OA interests:  (1) that Google's opt-out Library project was the same thing as its opt-in Publisher project and (2) that the the Google Library program would reprint or republish the copyrighted books that it only proposed to index.

* The textbook pricing crisis joined the journal pricing crisis on the radar of a growing number of universities, policy-makers, and news organizations.  Serious initiatives to produce OA textbooks multiplied and now include BookPower, California Open Source Textbook Project, CommonText, Free High School Science Texts, Libertas Academica,
Medical Approaches, MedRounds Publications, next\text, Open Textbook Project, Potto Project, and Wikibooks.  We even have a searchable portal from Textbook Revolution.  Textbook authors usually expect royalties, which may be why we're not seeing conversions from TA textbooks to OA.  Instead we're seeing new textbooks that are OA from birth, from a new generation of authors with a new set of priorities.

* The term "open access" is starting to seep out into the general scholarly culture that isn't working for OA so much as simply using it.  I've seen this from a special perspective as one who routinely searches for new OA developments.  About five years ago, OA news was very difficult to find.  New developments were scarce, reports or news stories about them were even scarcer, and there was no generally accepted term to use in a search.  Ever since the BOAI popularized the term "open acces" in 2002, this aspect of life has been easier.  But during 2005 finding OA-related news became hard again, or at least it became more time-consuming to weed out the true positives from the false positives.  One reason is that the term is now used for many other purposes, like power-grid and cable TV interoperability, wifi hot spots, and even a new method for scheduling doctor appointments.  But another reason is directly related to our success.  Search hits now bring up someone's recommendation of "an open access article about spinal cord injury" or a discussion of "an open access article about hydrological modelling".  But if you read the references, you find that they are really about spinal cord injuries or hydrological modelling, not OA.  Of course it's good that there are more OA articles to recommend and discuss, and good that more people are mentioning that articles are OA when they recommend or discuss them.  But five years ago, finding relevant new developments meant reading 20-30 times more than turned out to be relevant.  In the past few years this Chaff Ratio dropped to 2-3.  But now it's rising again to 4-5.

* Postscript:  Which was the first TA journal to convert to OA? 

In preparing this review of 2005, I was struck by Richard Roberts' claim that Nucleic Acids Research (NAR), published by Oxford University Press (OUP) was the first subscription-based journal to convert to full OA.  NAR's conversion took effect in 2005.  But it didn't seem right to me that 2005 was the first year in which any journals had ever converted.

Investigating further, I got caught up in the question what counts as a conversion to OA.  To take NAR as an example, it announced its partial conversion in August 2003, to take effect in January 2004.  Then it announced its full conversion in June 2004 to take effect in January 2005.  So do you date its conversion in 2003, 2004, or 2005?

Nucleic Acids Research

NAR's August 2003 announcement of its January 2004 conversion

NAR's June 2004 announcement of its January 2005 conversion

Environmental Health Perspectives (EHP), published by the National Institute of Environmental Health Sciences (NIEHS), one of institutes of the National Institutes of Health (NIH), converted to full OA in January 2004.

Environmental Health Perspectives

EHP's December 2003 announcement of its January 2004 conversion

BTW, it's possible that EHP will convert back.  In September the NIEHS asked for public comment on the possibility of privatizing the journal.  Whether EHP continues as OA after privatization would depend entirely on the policies of its next publisher.  Although the comment period on privatization ended on October 28, the NIEHS has not yet made a decision, according to a message from editor Thomas Goehl in the December issue.

Whether NAR converted to full OA before EHP depends on when you date the NAR conversion.  If 2003, yes; if 2004, it ties with EHP; if 2005, no.

But the question is moot, since neither journal was the first to convert.  Ulrich Pöschl said in the September 2004 issue of _Research Information_ that the European Geosciences Union (EGU) started to convert all its journals to OA "long before" the Berlin Declaration in October 2003.  (I don't have conversion dates for any of the 11 individual EGU journals, but I'd like to get them sometime.)

European Geosciences Union

If a subscription-based journal changes its name and publisher at the same time that it drops subscriptions and switches to OA, then is it one journal undergoing a conversion or two journals in succession?  If it counts as a conversion, then the three earliest conversions I can find are all of this type. 

In July 2000 Henry Hagedorn resigned as editor of the Archives of Insect Biochemistry & Physiology (Wiley-Liss) in order to form the OA Journal of Insect Science (University of Arizona library).  Early in 2001, a handful of editors of Topology and Its Applications (Elsevier) resigned in order to create the OA Algebraic and Geometric Topology (University of Warwick and International Press).  Finally, over a nine month period in 2001, forty editors of Machine Learning (Kluwer) resigned from the editorial board; one of them, Leslie Pack Kaelbling, created the OA Journal of Machine Learning Research (MIT Press).  For details on all three, see my list of journal declarations of independence.

No matter how you slice it, subscription-based journals went green before they went gold, just as born-gold journals preceded gold converts.  While there were certainly some conversions before 2005, there were more in 2005 than in all preceding years combined.

As long as I'm on the subject of journal conversions, here are two guides, one from 2004 and one from 2005.

Raym Crow and Howard Goldstein, Guide to Business Planning for Converting a Subscription-based Journal to Open Access, Open Society Institute, Edition 3, February 2004.

Jan Velterop, Guide to Open Access Publishing and Scholarly Societies, Open Society Institute, July 2005.  (Focused on helping society publishers convert their journals to OA.)

* PPS.  Here are some related year-end reviews by others:

StevenB, Top Stories of 2005 For Academic Librarians, ACRLog, December 30, 2005.

Top Ten Stories that Shaped 2005, LIS News, December 30, 2005.

Michael Geist, Canada A to Z 2005 review, P2Pnet, December 27, 2005.

* PPPS.  Before your memory of the past year fades, have a look at my timeline.  Let me know if I've omitted anything significant from the section on 2005.


Read this issue online

SOAN is published and sponsored by the Scholarly Publishing and Academic Resources Coalition (SPARC).

Additional support is provided by Data Conversion Laboratory (DCL), experts in converting research documents to XML.


This is the SPARC Open Access Newsletter (ISSN 1546-7821), written by Peter Suber and published by SPARC.  The views I express in this newsletter are my own and do not necessarily reflect those of SPARC or other sponsors.

To unsubscribe, send any message (from the subscribed address) to <>.

Please feel free to forward any issue of the newsletter to interested colleagues.  If you are reading a forwarded copy, see the instructions for subscribing at either of the next two sites below.

SPARC home page for the Open Access Newsletter and Open Access Forum

Peter Suber's page of related information, including the newsletter editorial position

Newsletter, archived back issues

Forum, archived postings

Conferences Related to the Open Access Movement

Timeline of the Open Access Movement

Open Access Overview

Open Access News blog

Peter Suber

SOAN is licensed under a Creative Commons Attribution 3.0 United States License.

Return to the Newsletter archive