Brianna L. Schofield & Jennifer M. Urban, Berkeley Digital Library Copyright Project Report: Takedown and Today’s Academic Digital Library
, U.C. Berkeley Pub. L. Research Pape
r No. 2694731 (November 2015), available at SSRN
A recent push to provide increased access to research, scholarship, and archival materials, as well as a desire to provide greater visibility to faculty and institutional work, have driven more and more academic libraries to create online repositories. These repositories have successfully generated greater visibility for scholarly work and archival collections and greatly enhanced access to these materials for researchers. Greater visibility and access, however, also bring greater potential for requests that libraries takedown materials either because of intellectual property rights claims or other claims, such as privacy.
Schofield and Urban studied the experience of academic libraries hosting open access repositories and their experience with notice and takedown requests, both under section 512(c) of the Digital Millennium Copyright Act (“DMCA”) and otherwise. They used a survey and targeted interviews to investigate how often takedown requests are received, for what type of content, the basis of the concern, and how the library responded to the takedown request. Schofield and Urban go on to provide recommendations on how libraries should respond to these takedown requests. Their findings have been published in Berkeley Digital Library Copyright Project Report: Takedown and Today’s Academic Digital Library. (available at SSRN) and will be presented at The Future of Libraries in the Digital Age conference.
Establishment of accessible repositories has been on the rise as the issue of access to research, particularly publicly-funded research, has gained attention in academia and the press. Traditionally the academic publishing model has been one where the author(s) sign over rights to their work in exchange for publication. Academic libraries then pay substantial sums of money to gain access to journals, and other publications, in which the research is published. The open access movement gained momentum particularly in the hard sciences and there are now federal, and sometimes state, restrictions requiring that certain types of publicly funded scientific research are made openly accessible for no cost. In addition, some academic institutions have begun encouraging, or even requiring, faculty to publish research in an open access format. These trends have fueled, in part, the rise in academic digital repositories.
Section 512(c) of DMCA (17 U.S.C. § 512(c) (2014)) provides protection for an “online service provider” (“OSP”) for copyright infringement by a user of the OSP. Schofield and Urban note that libraries developing and maintaining online, publicly accessible repositories may meet the definition of an OSP under the DMCA and become subject to both its requirements and protections. However, the authors also point out that the safe harbor provisions are only available for content loaded by a third party, such as a faculty member or student. Libraries who manage academic digital repositories often load the content on behalf of the author. As Schofield and Urban emphasize, that step, when performed by the library, eliminates the protections of § 512(c) since the library, through a librarian or staff member, did the actual loading of content.
The study revealed DMCA takedown requests are currently infrequent, although their incidence could rise as repositories become more prevalent. More common were non-DMCA takedown requests. While some of these did arise from copyright claims, the most frequent reasons given were privacy, embarrassment, and defamation concerns. Schofield and Urban found that many of these non-DMCA takedown requests were handled on a case-by-case basis, depending upon the cause for complaint.
Although the respondent pool for the study was small, see the article for why, the findings are intriguing and indicate that librarian managers of these repositories should be developing best practices for handling takedown requests as they are likely to grow in number and frequency. Some of the recommendations from the authors, for both DMCA and non-DMCA takedown requests, include author education about preserving rights, publication agreement transparency, and development of best practices within the academic library community. As the Schofield and Urban report highlights, authors, publishers, and academic institutions are likely to find the incidence of takedown requests on the rise. Academic libraries, as the developers, managers, and curators of digital repositories, should be prepared to respond.
James M. Donovan, Carol A. Watson & Caroline Osborne, The Open Access Advantage for American Law Reviews
(October 7, 2014), available at SSRN
Open access (OA) scholarship is available online, without fees, and free of restrictive copyright and licensing provisions. As institutions of higher education implement a more metrics-driven paradigm, law schools are increasingly attentive to the quantification of both individual faculty and aggregate law school impact. Citation counts are one means of quantifying these impacts. Donovan, Watson, and Osborne build on their 2011 article, Citation Advantage of Open Access Legal Scholarship, which demonstrated that open access resources have a great impact on legal scholarship, (103 Law Lib. J. 553, 557). In this article, they work to develop a systematic and scientific explanation for why open access scholarship has a citation advantage in the legal education context.
The authors’ research shows that articles published simultaneously as print and open access law review articles provide at least a 50% citation advantage over their print-only law review counterparts. More specifically, they find that the aggregate cumulative OA advantage for new and retrospective works combined is about 53%; the OA advantage of newer works published during the years 2007-2012 is about 60%. Their research also indicates that OA articles are more heavily cited in the years immediately following an article’s publication and that OA articles tend to “command greater attention over the lifespan of the work” (Donovan et al, at 8).
The authors also explore the measurement of the OA advantage to a law review as it relates to the institution’s ranking in the U.S. News & World Report. They conclude that the greatest OA advantage is for a journal whose home institution is in tier 2, 3, or 4 of the U.S. News & World Report law school ranking. For those tiers, the aggregate cumulative OA advantage for new and retrospective works combined is about 51% compared to an OA advantage of new works published during the years 2007-2012 of about 89% for tiers 2 and 3, and 81% for tier 4. For journals at tier 1 schools, the OA impact decreases significantly because journals at higher ranked institutions have high levels of exposure even without OA. In this tier, the aggregate cumulative OA advantage for new and retrospective works combined is about 11% compared to an OA advantage of new works published during the years 2007-2012 of about 16%.
As the authors point out in their conclusion, this article is a sobering reminder that readily available information on the Internet will often be the first, and in some cases the only, source consulted. Consequently, OA publishing offers faculty the potential opportunity to increase their work’s exposure in the field by being readily available, and therefore, is fertile ground for the OA citation advantage. According to Donovan, Watson, and Osborne, the OA citation advantage for a law review article is threefold: an OA article gets attention sooner; about half of the citations to an OA article will be from the first six years of the publication’s existence; and OA articles receive attention for a sustained period of time that exceeds the length of attention received by its non-OA counterparts. Depositing faculty scholarship in an open access repository, whether in SSRN or in an educational institution’s repository, is a simple, tasteful way that faculty can promote their scholarship while supporting the open access movement.
Every first-year law student is taught the importance of citing to the materials relied upon to make a legal argument. Opposing counsel and the judge should be able to retrieve the materials cited and determine whether the materials support the argument put forth. At first blush, citation to materials on the internet would seem to make retrieval easier. Lawyers no longer have to go to a library or dig through a database to find the cited materials. However, a fairly high number of these citations are being lost to link rot, that annoying instance when you click on a link and what you are looking for is no longer available.
Raizel Liebler and June Liebert conducted a study of internet links contained in Supreme Court of the United States (hereinafter “SCOTUS”) opinions from 1996–2010 and found that, shockingly, 29% of links in the opinions either led to nothing or did not lead to the information discussed in the opinion. With stare decisis as a foundational principle of American law, it is disturbing to consider how the underlying basis of court opinions may be disappearing at a rate much higher than anticipated. The consequences for lawyers and researchers seeking to understand the legal analysis contained in a court opinion are profound.
The authors begin, “Citations are the cornerstone upon which judicial opinions and law review articles stand.” (P. 275.) From there, Liebler and Liebert go on to discuss the increasing use of citations to links in SCOTUS opinions. The first citation to an internet link appeared in Justice Souter’s 1996 concurring opinion in Denver Area Educ. Telecomm. Consortium, Inc. v. FCC (518 U.S. 727, 777 n.4). The next use was not until a 1998 dissent by Justice Ginsburg in Muscarello v. United States (524 U.S. 125, 143 n.6). Since that time, citations to links has increased tremendously. Liebler and Liebert found 430 website citations in 144 SCOTUS opinions from the 1995–96 term to the 2009–10 term. Of these 430 URLs, 29% of them had suffered link rot. As the authors state, “Considering the preeminence of the United States Supreme Court, a link rot rate of almost one-third . . . is quite shocking.” (P. 298.)
The problem of link rot is not simply that a link may no longer work but can be much more complicated. For instance, a citation to an SSRN paper may later be changed as the author advances in the writing and publication process. The author of a blog post relied upon in a court opinion may later change her opinion and edit the blog post to reflect her current thinking. A lawyer or researcher examining the basis for a justice’s analysis may find material that is entirely different from what the justice cited. Statistics found on the internet, even on a government website, may be removed and replaced with newer statistics. In fact, the authors found no difference in the probability of link rot between government and non-government websites. Similarly, the probability of link rot did not depend on the format of the link (i.e., pdf, htm, html). The authors conclude that it is not possible to predict which links will rot.
There is, apparently, hope that this unfortunate state of affairs may find a solution. The authors highlight 2009 guidelines released by the Judicial Conference of the United States titled Citing To, Capturing, and Maintaining Internet Resources in Judicial Opinions/Using Hyperlinks in Judicial Opinions. The Conference advised judges to consider archiving a copy of internet materials relied upon in an opinion. The SCOTUS Clerk of Court does retain a print copy of cited materials in its case file but those materials are only available by visiting the Clerk’s office or contacting the National Archives and Records Administration (NARA) after the files are archived. The Ninth Circuit Library is also archiving websites cited by creating PDF files. As the authors insightfully point out, however, this works well for text files but does not capture audio, video, or software files.
Liebler and Liebert’s article offers a shocking illumination of the problem facing citation to internet links in a system of law based on precedent. With no solid plan in place to accurately preserve and make accessible cited material we risk losing the ability to see and understand the authority relied on by the Court.