The role of the “Listen in: Discussions from/in the field” column is to highlight interesting, exceptional, or provocative research published in LIS literature. This year we hope to emphasize a variety of research methods and the experiences of those often ignored in LIS literature.
Take it down… or else?: Privacy, freedom of information, and digital collections
With digital collections in academic libraries, our collections are available to a wider audience. For alumni or others whose personal information may appear in the collections, it may create privacy concerns, leading them to request that holdings be taken down. In the article “The Right to Be Forgotten and Implications on Digital Collections: A Survey of ARL Member Institution on Practice and Policy”, Virginia Dressier and Cindy Kristof examine how different libraries handle such requests as well as what policies and procedures are in place. Surprisingly, Dressier and Kristof found wide variations in how takedown notices are handled across libraries, and call for more libraries to discuss their procedures and policies more formally. Though the authors discuss the tension between the right to be forgotten and libraries’ duty to provide a complete public record, perhaps the field could benefit from a more thorough discussion of systemic forces at play.
“The main tension exists between the right to be forgotten, or right to privacy, and freedom of information.”
Though issues of privacy are not a new concept, the everlasting nature of digital collections, along with easier access to information thanks to the internet, have opened up new considerations for librarians. The main tension exists between the right to be forgotten, or right to privacy, and freedom of information. The right to be forgotten has been at the forefront of the news in the last decade, since the “Google Spain” case, which led to the formal adoption of Right to Be Forgotten law in the European Union. Since 2014, when the European Court of Justice found Google responsible for delisting private information about European citizens under certain circumstances, Google has now created a form where users can list the URLs they want removed from Google (only in place in EU countries). The authors discuss the difference in priorities between US and EU – whereas the EU tends to protect the privacy of individuals, the US is attached to its constitutional rights to freedom of expression. Dressler and Kristof also mention that most solutions to maintaining privacy in the digital world rely heavily on third party intervention, where entities that are the safe keepers of the individuals’ information ultimately make calls pertaining to their privacy.
The authors sent a survey to 124 institutions which were part of the Association of Research Libraries (ARL) network, with a response rate of 25.8%. The authors ended up using 28 responses as some responses were incomplete. The survey was a mix of multiple choice and open-ended questions. Multiple choice questions touched on demographics about their digital collections such as how long the collection existed, types of items in the collection, platforms, accessibility, staff in charge of the collection, size of the collection, and the existence of any policies. The rest were open-ended questions, where survey respondents were asked how they would handle hypothetical takedown request scenarios. Respondents were also asked to provide their policy, if possible. Dressier and Kristof did not expand on their reasons for choosing this methodology.
Results and Conclusion
This study was intent on giving exposure to starting a conversation around practices surrounding takedown notices, and perhaps start a basis for further research. The survey as a methodological instrument is conducive to exposing different practices, and given the wide range of responses, any further conclusions would be difficult to draw.
The survey provided some insight on the state of digital collections in the libraries surveyed. Over 90% of the collections had been online for 5 years or more, and approximately 60% for more than 10 years. Each library had digital collections with many different types of digital items, including images, newspapers, faculty research,student papers, and more. Respondents also indicated a variety of platforms, including commercial and locally created options. Most collections (61%) included more than 50,000 objects, though the authors conceded that it was hard for libraries with many collections to estimate the number of items. The authors emphasized that more established collections might be more inclined to develop policies. The multiplicity of platforms might also create obstacles to streamlined policies, and there currently isn’t an option that supports all types of digital collections.
More interestingly, the main findings of the survey were that the range of practices was much broader than the authors expected. Few libraries – only 25% – had actual policies that may relate to takedown request, and of those, only three had specific instructions to help navigate such situations.
The authors also found a wide range of responses to the hypothetical scenarios. The scenarios asked librarians to interpret takedown request situations. They were intentionally left open-ended and the survey respondents had no text limit to answer. Three scenarios were posited: the first for a name to be redacted from the student newspaper archive, the second for the removal of an article from the collection of a community newspaper, and finally a takedown notice from a publisher on account of a copyright infringement.
The responses were very thoughtful and provided insight into the complicated nature of takedown notices. Respondents mentioned having conversations with relevant stakeholders at their institutions. Some mentioned the distinction between published content and unpublished content, which would affect their decision for removal. Some mentioned the role of crawlers and indexing in search engines. Many also favored preserving the original version in an effort to maintain continuity of record. Where copyright infringement was concerned, most respondents expressed that they would ensure that the violation was valid before taking it down, and the threat of a lawsuit would make some respondent reach out to general counsel. Generally, the authors found that “institutions would err on the side of preserving the historical record” (Dressler & Kristof, 2018, p982), though the answers generally lacked uniformity. In fact, Dressler and Kristof cite a 2008 court case in which an alumnus from Cornell sued the university over a digitized newspaper article for defamation. The case was dismissed as the information contained in the article was truthful, an indication that the law is in favor of truth and freedom of information over individuals’ privacy.
Survey respondents were also asked about the job title of the person who would be consulted at their institutions, as well as to describe a similar situation to the hypothetical scenarios if one had ever happened.
The authors conclude by recommending the development of a high level, professional framework for dealing with takedown requests, in addition to more conversations around issues of copyright, publication, privacy and freedom of information. They encourage librarians to discuss the creation of policies within their institutions, whether or not it leads to a document. Finally, Dressler and Kristof warn us of the implications and long-term effect of taking down items, which they call the “swiss cheese” effect of creating holes in the digital archives.
I was drawn to this article after a conversation with the Scholarly Communications librarian at my institution. During a workshop where seniors learned to upload their thesis to the institutional repository, we came to talk about takedown requests from alumni who no longer wanted their undergraduate research to be available online. Following this conversation, I thought Dressler and Kristof’s article would provide some insight on how other institutions are handling such requests. Here are some takeaways:
- There are no current best practices across the field – each library is currently going with their own policy. This lack of consistency across the profession means that each institution needs to invest time to create their policy, or educate their staff about these issues.
- The answers to takedown requests are more complicated than it seems. The responses provided a wide range of possibilities. Sometimes, taking off someone’s name off is enough without having to take down the item. Items may be taken down from full access, but the record might stay up. Other times, the problem is with the item’s discoverability, and taking steps so that crawlers no longer index the item might be key. Sometimes, copyright issues might make take down requests easier – if copyright is infringed, taking it down is more straightforward. All of that must be balanced with a conversation regarding the ethics of privacy, and information access.
“When departments make uploading senior theses a requirement for graduation, students have little say about their thesis being available to the public. These power dynamics are important when we discuss privacy, freedom of information, and continuity of record.”
Overall, I appreciated the author’s thoughtful discussion of how different libraries approach this issue. I did wish for more details on their choice of methodology. In particular, I would have liked to know how they developed their questions, particularly for the open-ended hypothetical scenarios. Why did they use examples of newspaper articles? I also wished for a more complex conversation particularly surrounding student work. When departments make uploading senior theses a requirement for graduation, students have little say about their thesis being available to the public. These power dynamics are important when we discuss privacy, freedom of information, and continuity of record.
Keeping the conversation going
- As institutions of higher education encourage students to take risks as “growth learners”, while simultaneously requiring students to submit their thesis in institutional repositories, what should takedown policies be with regards to student research? As students did not willfully elect to publish their research, they may not want to be associated with their early corpus. Would taking down those items create significant holes in the collection?
- What “higher level” entity, if any, should lead the way in developing best practices for takedown requests? And whose interest should be protected – the general public’s right to information, or the individual’s right to privacy?
Dressler, V., & Kristof, C. (2018). The right to be forgotten and implications on digital collections: A survey of ARL member institutions on practice and policy. College & Research Libraries, 79(7), 972-990.
About the author
Charlotte Brun is a Social Sciences Librarian at The Claremont Colleges Library, in Claremont, CA. She is passionate about information access, feminism in the library, and social justice. Hobbies include furry friends, plants, knitting and reading. Find her on twitter: @cha_cjb
This work is licensed under a Creative Commons Attribution 4.0 International License
The expressions of writer do not reflect anyone’s views but their own