Editor’s note: In June, we published the report Social Feed Manager: Guide for Building Social Media Archives, written by Christopher Prom of the University of Illinois with support from our grant from NHPRC. To highlight this work and the insights it provides to the community, we’re sharing excerpts of the report as blog posts. This is the third in a series of posts we’re calling “SFM for Archivists”.
Research Consultation Service
Social media collection is not a ‘set it and forget it’ function, no more so that other web archiving efforts. Even in the simplest possible implementation scenarios, library and archives staff will need to develop an appropriate policy framework and will need to monitor and actively manage the service.
The need of active management is even more true if the collecting organization intends to support end user captures, to complement its own collecting activities. This use of the tool seems very likely in many repositories. Partnerships between archivists, students, and faculty will be necessary for both groups to get the most value from SFM. Archivists can help students and scholars understand how the application work, explain privacy issues, help them seed collections, review whether the tool is collecting the data they need, and help them package it for potential export and use. In turn, users can help identify collection areas and shape tool implementation.
In fact, GW Libraries originally built the service to meet faculty and student needs. With a strong campus focus on public policy/administration, media studies, politics, and public health, many GW faculty and students had interest in monitoring, capturing, and then analyzing social media data for potential insight into particular research questions. GW Libraries consult with potential users, help them registering to use the tool, and lets them collect social media records, which the users can export for analysis. Other repositories could do likewise, with three projected outcomes, in order of increasing complexity and support demands:
Researcher Collects, Uses, and Discards
In this scenario, the Library would simply act as facilitator. A researcher creates their own collections for their own use. Data stays in SFM as long needed for the current project and is exported from the UI by the user or by an administrator from the command line. When the project is over, the data is then is deactivated or deleted by SFM staff—a function that is available only to administrators of the system.
Researcher Collects and Library Preserves
A researcher creates a dataset for their own use, but makes an agreement that the archives will accession and preserve it for its potential future research value. In this case, a specific agreement to donate or transfer should be made with the collecting researcher. This could be via a simple email but ideally would be formalized in deed of gift or other legal instrument, in which the researcher gives whatever right he or she has in the data to the repository. Alternatively, if the donor does not wish to donate the dataset, the archives could be provided a license to distribute it or make copies available, either immediately or after a set period of time. Such an arrangement could be suitable in the case of a collector who does not wish to immediately share data, to protect a forthcoming research work, but who wishes to share it longer-term.
Library/Archives Creates and Preserves
In this case, the archives or library would directly manage the development of new collection sets, collections, and seeds, seeking to develop research resources that need an immediate need, but which also might have some longer term research value. A strategy like this could complement the building of subject-based collections mentioned in the last section. It would build partnerships with Library users to identify and capture collections with near and potentially long-term research use. In this respect, students and faculty members can help archives identify relevant collecting areas and shape uses of the tool. For example, users may be able to provide some direction concerning tool development and access methods. As Ian Milligan has noted, future historians will need to develop text mining and manipulation skills to interpret the voluminous records generated by web archiving tools (Milligan, “The Promise of WebARChive Files.”). As additional archives work with SFM, users can help shape future version of the tools, in particular any access methods that might to developed for SFM data.
When an archives or library is leading the collecting efforts, several potential approaches suggest themselves. For example, a subject librarian may be co-teaching or providing support services for a faculty member, say a professor who specialises in health care economics. When the students are assigned a project on consumer attitudes toward drug prices, emergency room costs, or some other topic, the faculty member and subject librarian consult with each other and decide to create a dataset that (hopefully) speaks to these topics. After collection has been completed, they export some datasets for the student groups to analyze. If the data proves useful, it could be retained for future use, or the harvesting could be extended over one or more years.
As another example, consider an archivist who is supporting a graduate history seminar about protest movements past and present. After reading widely in the secondary literature, the students are asked to examine and reporting on primary source datasets. Referred by their professor, student meets with the archivist and asks if they could collect social media data relating to protest groups on opposite sides of a campus issue. The archivist suggests some potential twitter search and filter harvests, then configures SFM to collect the data. Provided a system login, students download and analyze datasets for their class paper, and after the course is over, the archivist continues to harvest the data for potential future use.
As a final example, consider the case of an institution that holds strong library collections related to a particularly country, such as the Philippines, including books, newspapers, and periodicals in Filipino, English, Spanish, and other languages. In addition, the Library’s special collections department has acquired well-known and heavily-used manuscript collections from Filipino immigrant groups. To support faculty and student research, the library began collecting web resources in 2008, and it now wishes to supplement those collecting efforts by capturing social media, relating to both the Philipines proper and the local immigrants. SFM provides Library staff the means to do so, and even a tool to foster the local community’s digital self-archiving.
Obviously, there are many other potential uses to which SFM can be put within a particular archives. While the foregoing discussion mentions some general possibilities, the collecting organization will need to consider carefully which uses it wishes to pursue, translate those uses into specific goals, and then develop policies and procedures to ensure that the activities rest on a solid foundation.