April 17, 2014—In November 2013, Harvard received 23 takedown notices from Elsevier, a publisher of academic journals.
A takedown notice is a request from a copyright holder to remove a work from the internet because of alleged copyright infringement. To comply with the Digital Millennium Copyright Act (DMCA), internet hosts like Harvard must comply with takedown notices even if the recipient may choose to put the work back up again.
All 23 of the takedown notices targeted published editions of articles from Elsevier journals posted to websites on the Harvard.edu domain, including for example lab sites, faculty sites, and course websites hosted on iSites. All 23 articles were promptly taken down.
None of the takedowns targeted articles in DASH (Digital Access to Scholarship at Harvard), the open-access repository maintained by Harvard’s Office for Scholarly Communication (OSC). As Sarah Thomas, vice president for the Harvard Library, put it, "The OSC is part of the solution, not part of the problem."
We explore the issue with Peter Suber, the director of Harvard’s Office for Scholarly Communication.
Q. Why did the 23 Harvard faculty members receive takedown notices from Elsevier?
A. Elsevier believed that the posted copies infringed its copyrights.
Q. Was Elsevier within its rights to demand these takedowns?
When authors publish in a journal, they typically transfer all or some of their rights to the publisher. These rights typically allow the publisher to do what Elsevier has done.
Sometimes authors hold the right to post published editions of their articles to the internet. If any of these authors were in that position, then they could respond to the Elsevier takedown notice with a “counter-notice” and Harvard could put their articles back up again. But when authors publish in non-open-access journals, they rarely hold such rights.
Q. Some authors say that posting their own articles to their own web sites is commonplace and promotes research. They are angry that Elsevier is interfering with this practice. Can you comment?
A. It is commonplace and it does promote research. I’m sure that publishers have long known about the practice. But until recently they tended to tolerate it. These takedown notices mean that Elsevier is starting to enforce its rights to stop this practice.
I sympathize with author frustration. They wrote these articles, and they have good reasons to want to share them with anyone who wishes to read, apply, or build upon them.
At the same time, while they wrote these articles, they no longer own them, or no longer own all the rights to them. Elsevier acquired certain rights from authors and is entitled to act on the rights it has acquired.
Speaking personally, authors can’t blame Elsevier for enforcing the rights they gave it. If they dislike what Elsevier is doing, they should submit future work to a different publisher.
If they want to give the world free access to their articles, and I hope they do, they should take advantage of the lawful alternative Harvard has created through its open-access policies and open-access repository.
Q. What is this lawful alternative? How can Harvard authors make their articles open access without risking infringement or inviting takedown notices?
Many Harvard schools have adopted open-access policies granting Harvard non-exclusive rights to the future scholarly articles authored or co-authored by faculty, including the non-exclusive right to distribute a version through DASH (Digital Access to Scholarship at Harvard), the university’s open-access repository. It is this grant of rights that enables Harvard to avoid copyright infringement when implementing its open-access policies.
The policies also allow faculty to obtain a waiver or opt-out from the grant of non-exclusive rights to the institution. Hence, to be more precise, Harvard has permission to host open-access copies unless the authors choose to obtain waivers. Harvard only hosts open-access copies in DASH when it has permission to do so.
We now have open-access policies at the Faculty of Arts and Sciences (February 2008), the Law School (May 2008), the Kennedy School of Government (March 2009), the Graduate School of Education (June 2009), the Business School (February 2010), the Divinity School (November 2010), the Graduate School of Design (March 2011), and the School of Public Health (November 2012). All the policies are available on the web site of the Harvard Office for Scholarly Communication.
Incidentally, all the Harvard OA policies were adopted by faculty vote, not administrative fiat. At several of the schools the votes were unanimous, and at the rest the votes were by significant majorities or supermajorities.
Although Harvard’s non-exclusive rights allow it to distribute covered works through DASH, authors should take care not to enter agreements with publishers that conflict with Harvard’s non-exclusive rights. The OSC web site includes a way to amend publishing contracts in order to avoid this problem.
Q. What are some other benefits of depositing articles in DASH?
DASH provides persistent URLs so that links to its articles never break, and Harvard preserves the contents of DASH in a state-of-the-art digital preservation system.
DASH is regularly crawled or indexed by major search engines, like Google. Of course the contents are already open access, and easy to download or retrieve. But this indexing also makes them easy to discover.
If authors wish to deposit early, unrefereed versions of their research results (sometimes called “preprints”), then DASH will provide authoritative public time-stamps to establish their priority over others who may be working the same problems.
DASH sends authors regular stats on how often their papers are downloaded, and from what countries. It also collects first-person accounts from users about how they benefited from the open-access articles they discovered in DASH. Some of these stories are very moving.
The major advantage, of course, is to make articles open access, even when the journals in which they were published did not make them open access. This enlarges the audience for the articles, increases their impact, and even increases their citations.
Q. How does a faculty member or researcher deposit a paper into DASH?
The Harvard open-access policies only apply to Harvard faculty, but DASH will accept deposits from anyone with a valid Harvard ID, including students, staff, and fellows. Similarly, the policies only apply to scholarly articles, but DASH will accept many other kinds of content as well, such as theses and dissertations and multimedia files.