See the letters to Nature

Here are the letters, in the order in which they were sent to Nature:

[1] Text of letter sent to Nature by Professor Stevan Harnad, Canada Research Chair at the University of Quebec at Montreal, and Professor of Cognitive Science at the University of Southampton, UK

Sir,

Open Access (OA) means free online access to peer-reviewed journal articles. There are two ways to provide OA:

(1) either by publishing one's article in an OA journal that makes all articles free online ("Gold OA") http://www.nature.com/nature/focus/accessdebate/21.html

(2) or by publishing one's article in a non-OA journal and self-archiving it to make it OA ("Green OA") http://eprints.ecs.soton.ac.uk/13309/

R. Gadagkar (Letter to Nature, 22 May 2008) suggests that although denying access to users because of unaffordable subscription fees to the user-institution is bad, denying publishing to authors because of unaffordable OA publishing fees to the author-institution is worse, especially in the Developing World. http://www.nature.com/nature/journal/v453/n7194/full/453450c.html

The usual reply is that (1) many Gold OA journals do not charge a publishing fee and (2) exceptions are made for authors who cannot pay. More important, there is also Green OA self-archiving, and the self-archiving mandates increasingly being adopted by universities (e.g. Harvard) and research funders (e.g. NIH). http://www.eprints.org/openaccess/policysignup/

Self-archiving costs nothing, and if it ever makes subscriptions unsustainable it will by the very same token generate the windfall institutional savings out of which to pay for OA publishing instead.

Nor are the costs of publishing likely to remain the same under self-archiving: If journal subscriptions are ever no longer in demand (because users all use authors' self-archived drafts rather than publishers' subscription-based versions) journals will not convert to OA publishing under its current terms (where journals still provide most of the products and services of conventional journal publishing), but under substantially scaled-down terms.

Current costs of providing the print and PDF edition, of access-provision and of archiving will all vanish (for the publisher). Those functions will have been off-loaded onto the distributed network of OA institutional repositories, each hosting its own peer-reviewed, published output. The only service that peer-reviewed journal publishers will still need to provide then will be peer review itself http://www.nature.com/nature/webmatters/invisible/invisible.html and the windfall institutional cancellation savings will be more than enough to pay for that. http://www.publications.parliament.uk/pa/cm200304/cmselect/cmsctech/399/399we152.htm

But until then, Green OA is OA enough -- and free.

Sincerely,

Stevan Harnad

[2] Letter sent to Nature by Trustees of the Electronic Publishing Trust for Development

Sir,

As Trustees of the Electronic Publishing Trust for Development working with research scientists and publishers in developing countries* for over a decade, we write to correct misunderstandings conveyed in the correspondence from Raghavendra Gadagkar (Nature, 453, 450, May 22nd, 2008).

First, the choice for researchers in the economically poor regions is not between 'pay to publish' versus 'pay to read' since by far the majority of 'Gold' Open Access (OA) journals make no charge to authors whatsoever. Most are therefore free to both authors and readers.

Second, the alternative 'Green' route to OA for universities is to create low-cost institutional repositories (IRs) -- in which their researchers can self-archive their publications to make them freely available to all users with Internet access -- and this has already been adopted by about 1300 institutions worldwide.

A growing number (44) of universities and funding organisations (including Harvard, Southampton, Liège, CERN, NIH, Wellcome Trust, 6 of the 7 UK research councils, and India's National Institute of Technology) have already gone on to officially mandate Green OA self-archiving for all their research publications.

Usage of these resources by developing countries is now well recorded. As examples, usage of journals published in developing countries (and making no charge to authors or readers) was recorded by Bioline International as having reached 3.5 million full text downloads in 2007. Usage of research publications archived in IRs shows India, China, Brazil and South Africa among the top15 most active user-countries, and smaller developing countries to a lesser degree. Full text downloads from just one of the 1300 registered repositories showed UK: 10,174; India: 5,733; China: 5,070; South Africa: 1155. Detailed usage of 4 such IRs by 6 countries is shown in the EPT Blog.

It is clear from these small but representative examples of usage that OA has huge benefits for the progress of research in the developing world, and advances steadily.

Sincerely,

Subbiah Arunachalam, Flat No. 1, Raagas Apts, 66 Venkatakrishna Road, Chennai 600 028, India. Tel: +91 44‰? 2461 3224, Mobile: 97909 23941

Leslie Chan, University of Toronto, Department of Social Sciences, 1265 Military Trail, Scarborough, Ontario, M1C1A4, Canada, Tel: +1 416 287 7505

Barbara Kirsop, Electronic Publishing Trust for Development, Wilmots, Elmton, Worksop, S80 4LS, UK Tel: +44 1909 724184, Mobile 07773677650

Electronic Publishing Trust for Development, http://www.epublishingtrust.org University of Otago, New Zealand, http://eprints.otago.ac.nz/es/ Bioline International, http://www.bioline.org.br) EPT Blog: http://epublishingtrust.blogspot.com/2008/03/bring-on-irs.html

[3] Letter sent to Nature by Peter Suber, Professor of Philosophy, Earlham College, USA

Dear Sir/Madam,

Re: Raghavendra Gadagkar's letter in the May 22 issue, Open-access more harm than good in developing world. http://www.nature.com/nature/journal/v453/n7194/full/453450c.html

Nature gave Gadagkar's letter a misleading title. His argument is not against open access (OA) as such, or even OA journals as such, but against fee-based OA journals or "the 'pay to publish and read for free' business model".

However, Gadagkar's argument is misleading in its own right. He is apparently unaware that most OA journals charge no publication fees [1]. As of late 2007, 67% of the journals listed in the Directory of Open Access Journals charged no publication fees [2], and 83% of OA journals from society publishers charged no publication fees [3].

Gadagkar writes that "A 'publish for free, read for free' model may one day prove to be viable..." as if it were untried, when in fact it is the majority model around the world. Moreover, it's the exclusive model in his own country. All OA journals published in India are of the no-fee variety.

He also fails to mention that OA archiving already follows the model of no fees for readers and no fees for authors. In the same week that Nature published Gadagkar's letter, the OA repository at his institution, the Indian Institute of Science, passed the milestone of 10,000 deposits.

Peter Suber Research Professor of Philosophy, Earlham College Visiting Fellow, Yale Law School

1. http://www.earlham.edu/%7Epeters/fos/newsletter/11-02-06.htm#nofee

2. http://www.earlham.edu/%7Epeters/fos/2007/12/new-data-showing-that-most- oa-journals.html

3. http://www.earlham.edu/%7Epeters/fos/newsletter/11-02-07.htm#list