Elsevier has been sending takedown notices to any site hosting the final PDF version of its journal articles. The takedowns first became apparent on Academia.edu. Mike Taylor was one of the first to blog about it, takedown recipient Guy Leonard blogged about it, and there’s a link roundup on Confessions of a Science Librarian. Later it became clear that the takedown notices were more wide-ranging, going to hosting services like WordPress as well as universities. The blowback was enough to prompt a response from Elsevier.
Elsevier can send takedown notices since it owns the articles in its journals. It owns the articles because authors who publish in Elsevier journals sign away their copyright before publication. The license agreement allows for archiving of the author’s version, but not the journal’s published PDF. Authors should avoid posting the published version of their articles as a general rule, though a few publishers do allow it.
Here are my suggestions for avoiding this problem:
- Publish in an open access journal (see the Directory of Open Access Journals for a list by discipline). Many require only a license to publish, rather than a copyright transfer, and use a Creative Commons license.
- If you can’t publish in an open access journal, check a journal’s archiving policy in advance by searching it in SHERPA/RoMEO.
- Read the fine print regardless of where you are publishing. This is not like a software license where everyone just clicks “I Agree.” This is your work, so read licenses carefully. Copyright transfer gives complete ownership to the publisher, and your rights are limited to those listed in the license agreement.
- Archive your post-print if possible, since it is your final version incorporating changes from the peer review process. If not allowed, post the pre-print. Archive in a repository where your article is immediately accessible, such as VTechWorks. Research networking sites require membership (Academia.edu) and/or software download (Mendeley) that are barriers to immediate access.
- Make your archived version easy to read and reuse. If double spaced, revert to single spaced, and insert tables and figures in the appropriate places. Consider archiving your data as well so your work can be replicated and incorporated into larger studies. Attach a Creative Commons license to make it clear you are explicitly allowing reuse. [Update: if you transferred copyright you likely cannot assign a CC license- see discussions by Kevin Smith, Michael Carroll, and Charles Oppenheim.]
- If you have co-authors, come to agreement early on publishing venues and archiving so you don’t get locked into a result you don’t like. Remember that typically one author signs for all authors, so that person must understand group wishes.
- Learn about and download the author addendum which allows you to reserve rights, or use the addendum engine.
Above I briefly touch upon the fact that research networking sites do not provide open access, which is an aspect of this controversy I haven’t seen mentioned. By coincidence, at the time this became news I was searching for articles about DSpace and linked data and I found this article on Academia.edu. If you take a look, you’ll see that this article isn’t downloadable or printable without becoming a member of Academia.edu. All you can do is try to read the small print. Which, in my case, was enough to make me realize that I didn’t need it. But what if I did? This article isn’t available anywhere else.
Academia.edu added gasoline to the fire by taking such a combative (and calculated) attitude toward Elsevier in its own notice to users, linking to the Cost of Knowledge boycott and extolling its own support for open access (“Academia.edu is committed to enabling the transition to a world where there is open access to academic literature. Elsevier takes a different view…”). The e-mail signature of Richard Price, the CEO of Academia.edu, says “The goal of Academia.edu is to get every science PDF ever written on the internet, accessible for free.” I’m sure that would be good for Academia.edu, which is a for-profit business with an absurd domain name. Your participation on research networking sites will be monetized one way or another. If your article is available only on a research networking site, like the author above, do you want your work being used to attract members to a for-profit endeavor? Pro-open access statements by such companies should be considered with healthy skepticism, and in some cases they are just plain openwashing.
Most importantly, Academia.edu, ResearchGate, Mendeley (now owned by Elsevier) and others do not provide open access. Sign-up should not be required for access. Software download, in the case of Mendeley, should not be required for access. These services do not meet the definition of open access established by the Budapest Open Access Initiative:
By “open access” to [peer-reviewed research literature], we mean its free availability on the public internet, permitting any users to read, download, copy, distribute, print, search, or link to the full texts of these articles, crawl them for indexing, pass them as data to software, or use them for any other lawful purpose, without financial, legal, or technical barriers other than those inseparable from gaining access to the internet itself.
The point of this is not to be rigidly ideological for its own sake. It’s important to know what the term “open access” really means, otherwise it will get co-opted for private uses. If you choose to use a research networking service, please make sure you also provide a copy of your article to an institutional or disciplinary repository where it can be found and downloaded on the open internet.