User:SMiller
From Open Annotation Collaboration
Contents |
Contributors
Submitted by: Sylvia Miller, Project Director, "Publishing the Long Civil Rights Movement" University of North Carolina Press
Presented by: Jenn Riley, Head, Carolina Digital Library and Archives The University of North Carolina at Chapel Hill
Use Case: The Long Civil Rights Movement Project
Description
What
The "Publishing the Long Civil Rights Movement" Project (LCRM Project) is a collaboration of UNC Press, the Southern Oral History Program in the Center for the Study of the American South, and the UNC Library. It has been funded by the Andrew W. Mellon Foundation's Scholarly Communications division since 2008. In April 2010, the project partners launched an innovative online collection of civil rights scholarship with an annotation feature. Starting with 33 UNC Press monographs central to the study of the history of segregation, desegregation, and resegregation in U.S. public education, the collection has since grown to include more than 60 scholarly books, articles, papers, and reports. During its test period, currently extended through June 2011, this experimental site is free to invited users, who are encouraged to annotate the content.
The combination of Open Source software and frameworks that form the architecture of the site is called the Voice Publishing Platform; it ingests content on an ongoing basis in TEI XML form. Once loaded into the Voice system, the content is annotatable at the paragraph level, providing a rich and growing body of annotation targets.
To achieve financial sustainability, the LCRM Project plans to partner with an online publisher to make the collection available to academic libraries by subscription. At that point, some of the content will become firewalled by the subscription-access barrier, while other content will remain Open Access.
Scope
The types of annotations envisioned by the LCRM Project team are listed in the Commenting Guidelines. In addition to these, we are open to other types that users might devise; for example, one author has used the annotation feature to add discussion questions to his book.
The type of annotation that we consider central to the LCRM Project is the type that connects the target, to online resources hosted elsewhere. We see this as a use case that may not yet be fully supported by the OAC Alpha 3 data model. To date, the targets in the LCRM implementation are text-based; however, the outside materials are multimedia. In the future we would like to be able to annotate photographs, tables, notes, and bibliography entries as well as paragraphs. Annotation is currently done at the paragraph level, as each paragraph in the site has a unique ID. Some examples are as follows:
- Annotation with multiple outbound links https://lcrm.lib.unc.edu/voice/works/w/freedomsteacher/
s/2#p-1
- Annotation with outbound link to search-results list in archival image database https://lcrm.lib.unc.edu/voice/works/w/to-right-these-wrongs/s/6#p-370
- Annotation with outbound link to image intended to serve as discovery portal to image database hosted by outside institution https://lcrm.lib.unc.edu/voice/works/w/freedoms-teacher/s/12#p-841
- Annotation with internal link to content within the LCRM site https://lcrm.lib.unc.edu/voice/works/w/to-right-these-wrongs/s/6#p-363
Please note: In order to access these examples, you must first register on the site; we will turn on premium access for you.
Who
Robert Korstad of the Terry Sanford School of Public Policy and Duke University and James Leloudis of the UNC Chapel Hill History Department were the first to recommend linking to primary sources in order to provide users with opportunities for further research and for critique and discussion of scholarly authors' narrative interpretations of those primary sources. Such primary sources might be regarded as the historian’s "data set."
The project team has surveyed faculty, librarians, and students and continues to hold focus groups with groups in these categories. Sociologists, and specialists in the history of segregation, desegregation, and resegregation of education; the history of African-Americans in the United States, women’s history, the history of other social movements, and the history of law; and oral historians are some of the scholars who are interested in annotating scholarly works or using annotations in their research and teaching.
In focus groups, it has been useful to focus on archivists, who are interested in contributing annotations with links to the collections that they know well; they see the annotation as a connector that (1) provides a portal to digital resources and (2) provides context for those resources via the scholarly narrative . The LCRM Project provides an opportunity for publishers and archivists to collaborate on an ongoing basis in adding value to scholarship and creating new avenues of discovery for archives.
Details
Relevance
The LCRM Project annotations provide commentary at a granular level, which fits the OAC model. The LCRM pilot site uses RESTful architecture, assigning persistent URIs to each paragraph of text, which we believe is solid preparation for an annotation model in which all core entities are URI-addressable.
Another OAC principle of particular interest to the LCRM Project is the one that recommends allowing both the target and the annotation to be of any media type; as shown above, we already have examples of links to multimedia, and we are interested in developing the capability to embed multimedia in the annotation itself.
Interoperability is important to the LCRM Project. A core principle of the project is service to scholarly researchers via optimization of connectedness, transparency, and discoverability of all related scholarly content via both incoming and outgoing links, whether the content is firewalled via subscription or available Open Access. Our main purpose in joining CrossRef was to enable outbound linking to full text in bibliographies. However, knowing that DOIs link primarily to paid content, we have simultaneously implemented OpenURL. The LCRM project brings to OAC discussions a perspective that treats annotations as mechanisms that link a large variety of resources in a variety of ways, beyond simply the annotation-target relationship. We are therefore interested in learning how the OAC model might integrate with OAI-ORE (Object Reuse and Exchange), so that the annotation itself becomes part of the widespread connections among related objects.
Our live pilot site contains a small but growing body of actual annotations contributed by multiple users; our ability to provide real-life examples with the potential to test and refine the OAC model makes our use case especially compelling. We are currently in the process of refining the functional requirements that we will provide to our future partner publisher, who will help us transform the R&D pilot site into a self-sustaining subscription resource. Using these updated functional requirements, the partner publisher will either transfer our technology to their hosting site or re-create the annotation functionality within their own technology platform. The publishing platform we describe has the potential to support other collections of scholarly content on other topics in the humanities and social sciences in the future.
Perpetual access will be a commitment that the LCRM Project and its partner publisher will make to library customers. We are very interested in learning how the OAC ontology might help us plan how annotations will be stored, preserved, accessed, and shared over time.
Challenges
During our research and current pilot phase, we found that annotations seemed to be grouped, potentially, into types. Feedback in focus groups has indicated that it could be useful to record and store more metadata about annotations than our system currently allows (it currently stores author, date, and target). Categorization, tagging, and separation of the annotations into components (for example, commentary vs. links) have been suggested by users as well as a professional consultant who is currently working with us to refine our functional requirements for the project’s next phase. Richer metadata would allow us to provide services such as rich browsing and searching of annotations, both with and without the targets. It would also allow a variety of sharing mechanisms including private, group-only, registered user-only, or Open Access annotations. We are interested in ways of surfacing content and annotations beyond the subscription level.
Another issue of concern to users is the authority of the annotator. It would potentially be useful to rank them or categorize them in a recognizable and systematic way in addition to the optional profile now included in the commenting functionality.
The Django framework upon which our annotation feature is built does not at this time provide direct support for the OAC model.
The LCRM Project is not entirely ready for OAC compliance at present but would like to continue to develop in ways that might allow this in the future. Our program officer at the Mellon Foundation has encouraged us to follow developments in the OAC model.
