Open Access News

From May onwards some Elsevier journals will be offering to their authors the option to pay a sponsorship fee to ensure that their article, already accepted for publication, is made freely available to non-subscribers via ScienceDirect.

Worldwide approximately 10 million researchers can already access these journals through institutional subscriptions. In a few instances, authors publishing in these journals have requested an option to make their articles freely available online to non-subscribers.

Six journals in Physics are the first to offer such an option. These are:

Nuclear Physics A
Nuclear Physics B
Nuclear Physics B Proceedings Supplements
Nuclear Instruments and Methods A
Physics Letters B
Astroparticle Physics

Thirty more journals across other fields such as Life and Health
sciences also plan to offer this option in the next two months.

The author charge for article sponsorship is $3,000. The fee excludes taxes and other potential author fees such as color charges which are additional. Information about selecting this option is now available on the journal homepages at www.elsevier.com as well as Elsevier's author gateway site, authors.elsevier.com. The availability of this option will be offered to authors of the above-mentioned journals only after receiving notification that their article has been accepted for publication. This prevents a potential conflict of interest where a journal would have a financial incentive to accept an article.

Comment. This is important. A few comments now and more to come.

1. This is a large, welcome step. Hybrid journals provide genuine OA for authors who select the OA option (or they can, depending on the fine print). The step is welcome for providing more OA and welcome for putting Elsevier's weight behind it. Just as Elsevier's decision to permit postprint archiving in June 2004 broke the ice for many publishers who were not already green, this decision may also break the ice for those that are not already offering a hybrid option. (Note that Springer, Oxford, Blackwell and others already offer a hybrid option and broke the ice for Elsevier.)
2. The step is welcome even though the program is flawed. It has essentially the same defects that the Springer Open Choice program had when it was first announced. Elsevier's processing fee is very high (the same as Springer's), and may generate a low uptake by authors, especially since traditional page charges will be laid on top of it (same as at Springer). A low uptake will not indicate low interest in OA. Nothing in the announcement or at the journal sites suggests that Elsevier will waive the fee in cases of economic hardship. Further, Elsevier appears to demand transfer of copyright even for authors who select the new option (more below).
3. Like other publishers who have decided to accept author-side fees, Elsevier will have to stop arguing that these fees corrupt peer review. Like PLoS, BMC, and Hindawi, and others, Elsevier must erect a firewall between the editorial and financial sides of the enterprise so that peer-review judgments are not affected by the financial incentives. The PAMnet announcement gives a misleading picture of Elsevier's firewall and makes it seem porous and paradoxical: don't tell authors that the fee-based OA option even exists until the paper is accepted --but at the same time, tell them (through this announcement and the journal web sites) that the option exists. The explanation at the web site for Nuclear Physics A is much clearer: Elsevier won't ask authors for their access decision until it notifies them that their paper has been accepted. This makes sense.
4. The page at Nuclear Physics A adds a detail missing from the PAMnet announcement: "When calculating subscription prices we plan to only take into account content published under the subscription model. We do not plan to charge subscribers for author sponsored content." This policy, pioneered by Springer, is becoming customary for hybrid OA journals.
5. It appears that authors who select the OA option must still sign Elsevier's standard copyright transfer agreement. At least the order form for requesting the OA option makes no reference to an alternative. Springer required copyright transfer even for its "Open Choice" authors until October 2005 when it let authors retain copyright and adopted an home-grown equivalent to the Creative Commons Attribution-NoCommercial license. I hope Elsevier can learn from Springer in this regard.
6. The chief strength of hybrid OA journals and the chief weakness are the same: because only some authors in a given issue will select the OA option, libraries cannot justify cancelling their subscriptions. This postpones the day that libraries and universities will save money from OA journals, but it also reduces the risk for publishers and encourages them to try the experiment.
7. Elsevier doesn't have a name for its new program, like Springer's Open Choice, Blackwell's Online Open, or OUP's Oxford Open. I'm calling it the "OA option" for now, but largely from optimism and lack of an alternative. When Elsevier finally gives it a name (say, the XY program), then I'll gladly trade in "OA option" for "XY option", especially if we learn that the option removes no permission barriers.

Update. Soon after the Carl Schwarz announcement appeared on PAMnet, the same announcement, now signed by Tony McSean and Daviess Menefee (Elsevier, Library Relations), began appearing on other lists.