 Thank you for joining us for the University of Pittsburgh Library System Briefing, Improving Stewardship, Opportunities and Tensions. My name is Carrie May and I'm the Digital Archivist and Preservation Librarian for the University of Pittsburgh. I will be co-presenting today with Clinton Graham, Systems Development Lead for the ULS Information Technology. We will be presenting information on Implementing Full Digital Preservation in the University of Pittsburgh Library System, ULS, Selecting Preservica, Setting Standards and Drafting a Digital Preservation Policy, Integrating Systems Using APIs, Opening New Pathways with Authenticated Access. So let's get started. The ULS has acquired, created and provided access to digital content for 20 years using various platforms and standard business backups. During 2019, I conducted extensive research into seven possible digital preservation solutions. In June of 2020, the ULS began using Preservica to provide active digital preservation to its archival assets. We began looking into Preservica because it clearly provided the four core aspects we had set as requirements. One, the potential for applying more uniform standards to all ULS digital assets. Two, risk mitigation practices. Three, scheduled consistent health checks. And four, reporting across all content. But Preservica also offered two further capabilities of great interest to us. One, there are existing APIs that could be used to integrate platforms already in use by the ULS. And two, there is a potential for allowing authorized access to specified collections. With digital content stored over various repositories posted on various platforms. The integration of these platforms was recognized early on as an important goal for the improvement of the storage and maintenance of ULS assets as a whole. APIs could allow the transfer of preservation copies from these repositories to a single secure system for full preservation. Online publication of open access materials is already available through the ULS digital collections website using Island Dora. But the ULS holds several collections with access potential affected by copyright limitations. By combining the user portal provided by Preservica with the capability to apply user authentication to specific collections. ULS could offer a viable new path for online access to these materials. But before moving forward with these exciting possibilities, we need to create a foundation for our newly initiated digital preservation program. This means we had to start setting some standards. My position is situated in the archives and special collections department. So my work of setting standards began with a focus on archival collections attract the journey of a digital asset from acquisition to ingest for preservation. Next, I documented these steps and the digital preservation workflow. When seeing this workflow colleagues were surprised by the number of individuals and separate units involved. This chart also made the complexity of the path abundantly clear. So setting standards to be used throughout this process started at the very beginning with defining the method for creating disk image files. Then moved through specifications for documenting physical media. Next, the acceptable base level contents of a submission information package were determined. Formats to be used for normalization were selected. With this information, the ULS digital preservation team was created and using these standards began drafting a digital preservation policy. Members of this team are not only from the ANSC, it includes staff from administration, ULS IT, the metadata and discovery unit, digital collections, university archives, and digital scholarship. The first steps in creating the program had focused on archival collections, a diverse membership in this team was necessary to represent the needs that will arise as content from other repositories is ingested. In February of 2021, the ULS digital preservation policy was distributed to all ULS colleagues. Now I'd like to hand over to Clinton to present the information on API's. With the digital preservation policy in place, we can begin thinking about the technical connections which will need in order to build the connections between our systems to help facilitate that policy. This begins with one philosophy, open source in support of open access. The ULS has a deep investment in leveraging and contributing to open source tools to facilitate making scholarly and archival materials publicly available. This gives us a strong foundation of existing API's for interoperability, and the option of building out new API's if we find any that are missing or faulty in functionality. This presents us with two challenges, which I'll describe as preservation first and preservation for preservation first represents the work that is done in Preservica, where access copies can be made publicly available to other systems. We have a substantial backlog of uncurated digital content, which is sitting on original or duplicate storage media, and will be assessed, organized and described in Preservica. If that content is appropriate for public release, the rest and content and link data APIs in Preservica, in concert with API's or other tooling within our dissemination systems, will help us to build a workflow for getting the dissemination information package from Preservica ingested into the appropriate public facing repository. Preservation for represents the existing content, which has been published in another system, which Preservica is then used to capture and preserve. We have a substantial amount of existing published content, which is deserving of full digital preservation. As that content is identified, API's or ETL tooling within the existing publication system will help us to streamline the creation and processing of submission information packages destined for Preservica. We'll do that primarily through three platforms. Islandora, the public knowledge project and ePrince. Within Islandora, we have 50 plus terabytes of digital content of archival and distinctive collections. The ULS has been an active member of the Islandora community for the past five years, and this platform will be a primary recipient of that preservation first digital content, which we also may want to make available publicly. Islandora has a robust set of community maintained modules extending the underlying Fedora data store, not only with rich presentation capabilities, but also with API and ETL functions that we'll need to make use of. The public knowledge project software represents 100 plus gigabytes of scholarly journal and monograph content of which we are the publisher of record. The ULS has a long standing relationship with PKP as the organization's first major development partner. The scholarly peer reviewed journals published in the Pitt Open Library Publishing Program are covered by industry standard backups and by membership within both public and private locks networks, but the preservation options afforded by Preservica may also come into play in the future. When that is the case, the ULS will also become a publisher of a new plugin for PKP to facilitate that connection. And finally, ePrince or frankly, another future repository platform. We have 500 plus gigabyte of author self archive content in subject based repositories and in the institutional repository. The ULS has a long history of using ePrince as a platform for author self archiving repositories, but we're currently evaluating our commitment to the platform. We do act as a steward of some very unique content within these repositories, which should be specially considered for long term preservation, most notably the University of Pittsburgh theses and dissertations. If our platform remains as ePrince, we'll be looking to collaborate on existing integration efforts. And if a new platform is judged to be preferred, the ability to leverage or build out APIs to facilitate the interaction with Preservica will be a key consideration. As part of our open source commitment, we'll be publishing our work using and building the API connections in our GitHub repositories for community reuse and remixing. There remains, however, a subset of artifacts which we can't fully make public yet. And we need a mechanism for the limited sharing that we can offer within copyright guidelines. For this, Carrie will now present on our last item, authenticated access. Carrie. Thank you, Clinton. The overall collecting policy of the ANSC does not simply focus on acquiring and preserving final products, but also seeks to maintain and present the entire creative process leading to those products. Two collections with contents that clearly document the creative process are the George A Romero archival collection and the August Wilson archive. Both collections contain drafts, notes, revisions, correspondence, photographs and audio visual documentation related to completed published works and unpublished works. The ULS quickly recognize the challenge of providing access to these significant personal archives, since the use of much of the content is affected by copyright licenses, which disallow open online access. Romero collection is a core component to the ULS horse study special collection and allows students filmmakers and fans to trace Romero's projects from inception to completion. The Wilson archive presents researchers with the ability to trace productions from conception through drafting to completion, and even through varying performances. Currently, researchers wishing to utilize these materials must submit a completed request form, schedule an appointment in the archives reading room, travel to the physical location, and work on a single standalone PC. The ULS seeks to initiate a new option for accessing these materials by combining specialized permissions on selected assets and a preset user account with a user portal requiring authentication for login. These are standard capabilities of the preservative system, but our idea is to use them in a slightly novel way. The new process will require the researcher to complete that request form and specify the materials that they would like to use. The preservation librarian will then modify the permissions for those assets so that they can be viewed in the user portal. It will then set up a user account and negotiate a date and time with the researcher to access the materials. The preservation librarian will send the requester the username password and URL for access when appropriate. The user account will then be enabled for the time arranged and disabled when that time has ended. Asset permissions will also be returned to standard settings for security purposes. In essence, the process is simple, but it does rely on some on several manual steps. Initial testing of account creation and modifying permissions have been successful. Next steps will include further testing of methods for managing the rendering of specific content to ensure that only requested material are accessed as agreed. Tests that include examining access capabilities for various storage locations. Examining the possibilities and challenges of temporarily altering storage locations of specific assets and also determining the risks of taking any of these actions. We will also then need to design a research request form to gather all necessary information to provide a successful access experience for the requester. Finally, we will draft the full workflow for fulfilling these requests and to find the individuals responsible for each step. In summary, in moving to full digital preservation, the ULS has taken a solid step in improving the stewardship of its digital assets. Though complex, using APIs to integrate the various repositories in use will be another solid step in improved stewardship. Finally, the creation of a method to allow authenticated access to limited access collections will provide improved potential for research in such collections and the creation of scholarly outputs. We thank you for attending the University of Pittsburgh library system briefing. If you have any questions, please feel free to contact us.