2026 DSpace North American User Group Meeting

Registration and Breakfast (provided)

Check-in and enjoy coffee and refreshments

Welcome! Opening Remarks and Logistics

Welcome to Harvard Library and the DSpace North American User Group meeting

The State of the Program: DSpace's Current Trends, Strategic Direction, And the Road Ahead

Presenters: Holger Lenz and Lia Eggleston (Lyrasis)

The DSpace Program has undergone significant growth in recent years and continues to evolve in response to changing expectations around open scholarship, research information management, and repository interoperability. Institutional support through membership has steadily increased, new staff members have expanded the program team, and the recent approval of the merger of DSpace and DSpace-CRIS has established a clear, forward-looking strategic direction for the platform’s technical development for the next year.

How have these developments shaped the DSpace Program? How is the program positioned amid increasing uncertainty around funding streams and broader fiscal volatility affecting the scholarly infrastructure landscape? What trends have emerged, and how do these trends reflect DSpace’s commitment to long-term sustainability? 

This presentation will provide an overview of the current state of the DSpace Program, including recent developments, emerging trends, and the broader context within which the program operates. Attendees will also gain insight into the structure of the DSpace Program and its community, the roles of program team members, and the collaborative processes that support the program’s ongoing development. 

Designed for repository managers, developers, and institutional stakeholders, this session offers both a high-level program update and an opportunity to better understand how community members can engage with and influence the future of DSpace.

Break / Networking

Charting a collective course towards Entities: Reflections from exploring and implementing Configurable Entities as a community

Presenter: Julia Gilmore (University of Toronto) 

When Configurable Entities were introduced in DSpace 7, this feature presented exciting opportunities to enhance and expand the capability of IRs via new data models, virtual metadata, and richer relationships between repository objects. Alongside this excitement, however, came many questions about how best to implement and leverage this new functionality.

This presentation will share how the Scholaris community in Canada, including the Scholaris service team, participating academic institutions, and expert group members, came together to explore Entities and build local capacity through a combination of shared approaches and strategies. These efforts included hosting monthly working hours, dedicated testing sessions, and conducting community outreach to develop documentation and guidance, share local implementation roadmaps, and establish flexible workflows for adopting and using Entities.

This presentation will highlight key takeaways and outcomes of our collective Entities adventure, share challenges, successes, and lessons learned, and describe some of the technical requirements and institutional considerations for implementing this feature. This presentation will also include brief case studies of how several Scholaris repository teams have adopted Entities at their institutions and provide practical recommendations and insights for attendees who are considering, implementing, and/or supporting Configurable Entities in DSpace.

Advancing Persistent Identifiers in DSpace: Global Community Feedback and Next Steps

Presenter: Sheila Rabun (Lyrasis)

This session will present findings from a 2025 global survey of the DSpace community focused on the use and future development of ORCID and other persistent identifiers (PIDs) such as DOIs and ROR. Conducted on behalf of the ORCID US Community and ORCID consortia counterparts across the globe, the survey gathered 93 responses from 78 organizations across 26 countries, offering a broad view of current practices, shared challenges, and topics for further discussion. We will discuss ideas for potential next steps, and attendees will be encouraged to provide their own insight and feedback related to further prioritization and development of PID integrations and functionality in DSpace moving forward. More information about the survey results can be found at https://lyrasis.org/advancing-orcid-and-persistent-identifiers-pids-in-dspace-community-driven-priorities/

The Entities are Coming: Lessons Learned from a Pilot Project to Integrate ORCID with Configurable Entities in a DSpace Repository

Presenter: Nicholas Woodward (Texas Digital Library)

In summer, 2026, the Texas Digital Library conducted a pilot project with the University of Texas at Austin to explore integrating their DSpace repository with the ORCID API. The project included testing various features of the ORCID integration and assessing how they can be implemented in their repository. As part of analyzing the ORCID functionality, pilot participants investigated incorporating Configurable Entities into the DSpace repository, particularly Publications and Person entities that could be associated with Researcher Profiles. This work included improving and fine-tuning a workflow for generating Person entities from existing authority metadata that was previously developed in an earlier DSpace pilot project. In this presentation, I will chronicle our experiences during the pilot project and what we learned about integrating a DSpace repository with the ORCID API, enabling Configurable Entities to broaden the functionality associated with ORCID, and incorporating a workflow for migrating authorities managed in earlier versions of DSpace into the entities framework.

Lunch (provided)

A to Z Space

Presenter: Claire Blechman (Harvard Library)

Join Claire as she highlights the A to Z of "Spaces" from around the globe! 

Update from Lyrasis Director of Community Supported Technologies 

Presenter: Bridget Almas (Lyrasis)

Towards the integration of CORE into DSpace

Presenter: Federico Verlicchi (4Science)

This presentation outlines key areas for collaboration where 4Science’s advanced repository solutions, CRIS systems, and data management capabilities can integrate with CORE’s vast aggregation of open access content to maximise research visibility, interoperability, and impact.

Collecting and indexing research from repositories and journals from one side and 4Science, a leading provider of DSpace and DSpace-CRIS solutions (repositories/RIMS software platform) from the other side, initiated a strategic collaboration to explore tighter integration of CORE functionality within repository systems.

The need for integration:
The Open Research infrastructure, of which we are all part, is supported by numerous distributed components that must function cohesively to achieve success. Communication between the various components is essential for the vitality of such an ecosystem. Various initiatives, such as the OAI-PMH protocol and, more recently, COAR Notify, enable stakeholders to communicate in a distributed manner. However, whilst these services exist in principle, effective communication remains dependent on the individual components' capacity to interact with one another. On various occasions in the past, communication failures have given rise to a narrative suggesting that the system would benefit from having fewer actors and adopting a more centralised structure. However, we contend that such an approach is detrimental to the overall future of Open Research, and we are working towards a more FAIR and open paradigm. Both CORE and 4Science represent significant actors in this domain, and we share common objectives. This collaboration does not constitute a closed ecosystem; rather, it is undertaken with a commitment to openness and the delivery of improvements to the community from multiple perspectives. Both CORE and 4Science benefit from this collaboration: CORE, as an aggregator, faces limitations in integrating with workflows necessary for a research paper to materialise, whilst 4Science operates at the centre of these processes but may lack the comprehensive overview and resources that an aggregator can provide.

DSpace as a Living Repository: Integrating Low-Code Platforms to Drive Research Workflows

Presenter: Małgorzata Paszkowska (PCG Academia)

Most institutions treat DSpace as a destination, a place where finished work is deposited after the fact. This presentation argues for a different model: DSpace as a living data backbone, actively integrated with a low-code workflow platform to drive day-to-day research operations.
We will share a practical implementation built around two core use cases. In the first, DSpace serves as the authoritative source for faculty performance review: rather than manually entering publications and project outputs, the low-code platform queries DSpace via API and pre-populates evaluation forms with verified bibliographic records. The approval workflow (faculty member, department chair, Vice Provost for Research) is configured directly in the low-code environment, with automated notifications and resubmission at each stage.

In the second use case, research projects are created and managed entirely within the low-code platform: team composition, task tracking, budget monitoring, and deliverable registration. Here, DSpace is not just a deposit target, it is a data source. Researcher profiles and equipment records held in DSpace feed directly into project setup, eliminating redundant data entry. When a project reaches a defined milestone, the platform automatically triggers a deposit to DSpace, pre-filled with metadata generated throughout the project lifecycle. The integration layer can extend further to HR and ERP systems, making DSpace a hub in a broader institutional data ecosystem.

The integration is built on DSpace's REST API, configured through the low-code platform's native API capabilities and our team's hands-on knowledge of DSpace's data model. No middleware products. The effort went into the things that actually matter: metadata mapping, workflow design, and permission modeling.
Attendees will leave with a clear architecture pattern, a practical metadata mapping framework, and honest lessons learned, including what to plan for before you start.

Break / Networking

Tour of Widener and Houghton Libraries

Attendees will be split into separate groups for two 45 minute tours of both Harvard's Widener and Houghton Libraries - attendees will see both libraries. 

Widener Library

The Harry Elkins Widener Memorial Library opened in 1915. The library was built as a gift from Eleanor Elkins Widener to her son, Harry, who passed away on the maiden voyage of the Titanic. Widener Library is the largest library on Harvard’s campus, and arguably, the most beautiful. Widener has 57 miles of shelf space and the capacity to hold over three million volumes. At the building's heart are the Widener Memorial Rooms, displaying papers and mementos recalling the life and death of Harry Widener, as well as the Harry Elkins Widener Collec­tion, "the precious group of rare and wonder­fully interesting books brought together by Mr. Widener," to which was later added one of the few perfect Gutenberg Bibles

Houghton Library 

Join us for an introduction to Houghton Library, Harvard’s primary rare book and manuscript library. The tour includes visits to our exhibition spaces and display rooms dedicated to the English writer Samuel Johnson and his circle, Romantic poet John Keats, American poets Emily Dickinson and Amy Lowell, as well as the library of Harvard collector William King Richardson. A history of the building and an overview of services available to library patrons will also be provided.‍

Conference Reception

Location: TBD

Breakfast (provided)

A Vibe Check for our Glow Up: Gathering User Feedback During an Upgrade

Presenter: Anne Shelley (Iowa State University)

In this presentation, we will describe our team’s efforts to involve users in a DSpace upgrade project. We wanted to better understand users’ experiences navigating a test instance of the redesigned repository and potentially incorporate their feedback into the upgrade process. From the usability sessions we held with seven participants, we gained a deeper understanding of how users search and browse DSpace, what tasks they find easy and challenging to accomplish, and how they adapt their approach when needed. In this talk, we will explore some of the challenges alongside positive findings from the user testing. We will highlight general patterns in how participants adapted to obstacles, their efficiency in completing assigned tasks, and the variability in performance among users with differing technical and subject expertise. This discussion will also explore potential strategies for addressing limitations in the default DSpace interface while leveraging its intuitive features to enhance the overall search and browse experience.

Going GLAM: The State Library of Massachuetts' Migration to DSpace-GLAM

Presenter: Emily Crawford (State Library of Massachusetts)

The presentation provides an overview of the migration process from a DSpace Express instance to DSpace GLAM instance and describes what it was like for staff at the State Library of Massachusetts to transition to the new system. In 2025 the migration took place over 6 months from our former digital repository instance to our new one that is a further evolution of DSpace-CRIS.

We successfully migrated over to a new repository in under a year with many lessons learned along the way. The transition to DSpace-GLAM marked a radical change in how we approached our repository structure and metadata. Such a change in how we thought about how our repository is structured and how we apply metadata resulted in workflows being updated. As part of the migration process staff took part in in-house and vendor-provided training which contributed to the success of adopting DSpace-GLAM as our official repository. 

From Chaos to Connections: Leveraging Entity Model Adoption as a Catalyst for Change

Presenter: Lydia Genry (Naval Postgraduate School)

In April 2023, the Naval Postgraduate School’s Dudley Knox Library and Atmire worked to migrate the institutional repository, Calhoun, to DSpace 7, transitioning from a complex web of communities and collections to an entity-based structure. While the technical migration succeeded, the aftermath presented unforeseen challenges. Just before go-live, the primary repository manager retired, creating a gap in institutional knowledge. 

Post-migration, several things happened all at once. Major staffing changes further destabilized the library. Increasing organizational demands revealed the need for innovations to leverage the new entity structure. Groupings that made sense in a “community” were no longer accurate when mapped to a specific affiliate department or organization. It became apparent that metadata had been handled inconsistently, and processes for ingesting content had resulted in previously undetected duplication. Old manual practices of harvesting content became unsustainable, which led to coverage gaps that became clear when the content was organized through entities. Workflows were unclear and ineffective, leading to staff and researcher frustration. 

This presentation aims to share lessons learned from undertaking a repository reorganization project in a resource-constrained environment, and how the new entity relationships revealed the need for a multipronged approach to effectively leverage the new data structure. Project components include the development of a comprehensive timeline of departments and organizations associated with the Naval Postgraduate School since 1952, updated workflows and templates, metadata remediation, and library system integrations to ensure the flow of data into the library’s discovery layer. Leveraging enterprise software and open source webapps, the team has been able to automate and distribute many of the tasks, establishing a sustainable framework crucial for succession planning and long-term repository health.  As a result of these efforts, our collaborative team won the FEDLINK (Library of Congress) Project of the Year Award.

Break / Networking

Lightning Talks

Stats Query by Community: a python script you can put your own variables in and get a nice Excel report from - Claire Blechman (Harvard Library)

DSpace's stats are crucial for reporting on the repository's usage, but they are not always easy to compile. This lightning talk will present a custom Python script I commissioned to pull community-level download stats from the DSpace API and automatically populate an Excel sheet that tabulates the results. Running these queries by hand (one for each community and month) then combining the data from all 7 continents took a full day's work, and this script lets us pull our biannual reporting in seconds. For those of us who do not code I will also including a practical guide to adapting that script to your own reporting needs.

More than a Feeling: Statistical reports that help you know how your repository is used - Nicholas Woodward (Texas Digital Library)

Obtaining DSpace repository usage metrics that are accurate, useful, and easily interpreted has been an ongoing challenge for many institutions. While DSpace provides some statistics in the user interface and includes integrations with Google Analytics and more recently Matomo, Texas Digital Library (TDL) members wanted repository views and downloads statistics in a more easily digestible, customizable and shareable form. With this charge in mind, TDL embarked on a project to develop lightweight, customizable Python scripts to generate spreadsheets of usage statistics and email them to the repository administrators on a monthly basis. In this lightning talk, I will describe the history of developing our statistical reports code, what statistics are generated, and how they are used by our members. I will also outline the challenges presented by large-scale bot traffic and the strategies we at TDL have used to mitigate their effects on views and downloads metrics in our 22 DSpace repositories.

DSpace 10.0 and 11.0: Pathway towards the merger with DSpace-CRIS

Presenter: Tim Donohue (Lyrasis)

DSpace 10.0 is due to be released in June 2026. This major release is the first (of two) to include features from DSpace-CRIS in order to achieve a merger of both products. 

This presentation will discuss the new features and improvements arriving to the DSpace platform that arrive in 10.0, including those coming from DSpace-CRIS. These new features include an audit trail, easier item editing, ability to customize URL paths for items, basic content management functionality, and many more. Time permitting, we could even demo a few!

I'll also discuss the (in progress) roadmap for 11.0 (due in May/June 2027), including how we plan to complete the merger of DSpace-CRIS into DSpace. We already know which DSpace-CRIS features will be coming in 11.0, so I can give a brief preview of those upcoming features.

Finally, I'll wrap up with details on how you can support DSpace (in general) and specifically 11.0 efforts.

Lunch (on your own)

Tour of Harvard University Archives' Collections

The Harvard University Archives documents nearly four centuries of institutional and intellectual life at Harvard, holding records and publications from more than 800 administrative, academic, and research units. These collections trace Harvard’s evolution from a small colonial college to a modern research university of international scope, encompassing all facets of university life—from governance, teaching, and research to residential life, dining, building projects, and fundraising. They also record the University’s participation in local, national, and global events, offering a rich lens on the emergence of higher education in the United States, the development of academic disciplines, and shifting pedagogical trends and financial models.

Beyond official records, the Archives preserves theses, dissertations, prize papers; the personal archives of Harvard faculty; and materials related to student and alumni/ae experiences that add critical depth. These materials go beyond official institutional records, and include photographs, diaries, correspondence, art, audio and video recordings, ephemera, webpages, and more, capturing lived experiences of campus life, social and intellectual networks, and responses to major historical events and changing social norms. Faculty papers and other personal collections further illuminate the professional, intellectual, and often family lives of individuals associated with Harvard, documenting not only scholarly work across nearly every discipline but also public leadership, creative pursuits, and social and political activism. 

Together, the Harvard University Archives' collections offer a nuanced record of institutional history, intellectual life, and social change within and beyond Harvard's gates.

To a Cloud-Native DSpace: Lessons from Migrating JScholarship to AWS Fargate

Presenter: Steven J. Miklovic (Johns Hopkins)

JScholarship, the Johns Hopkins institutional repository, recently migrated from a traditional VM-based DSpace deployment to a fully containerized, cloud-native architecture running on AWS ECS Fargate. This presentation walks through the journey: what motivated the move, how we designed the infrastructure, and what we learned along the way.

The talk covers the practical realities of running a production-grade DSpace stack — API backend, Angular frontend, a 3-node SolrCloud cluster backed by a 3-node Zookeeper ensemble, and PostgreSQL. We'll discuss how we maintained dev/prod parity; how we handle configuration and secret injection across environments; and how we structured our OpenTofu Infrastructure As Code into reusable, shareable modules.

I will also touch on operational patterns that emerged from this architecture: EventBridge-scheduled maintenance jobs that replace traditional cron, a CI/CD pipeline that coordinates deployments across four GitHub repositories using commit-SHA-based image tagging, CloudWatch Synthetics canaries that monitor Solr collection health, and tuning WAF rules for repository-specific traffic patterns, including SWORD deposits using AWS and Cloudflare.

The goal is to share both the successes and the surprises; measuring EFS latency with Solr/Zookeeper ensemble, health check grace periods that need to be longer than you think, bimonthly Fargate patching.

This session will be relevant to institutions evaluating cloud hosting options for DSpace, teams considering containerization of their repository infrastructure, and developers interested in infrastructure-as-code approaches for DSpace deployments. Beyond the public sharing of the IAC modules, the patterns and lessons are adaptable to any cloud provider.

 I will close with briefly touching upon the ways agentic tooling was used in this project.

Keeping links alive: Improving bitstream URL persistence in DSpace

Presenter: Ignace Deroost (Atmire)

Persistent access to digital content is one of the core responsibilities of an institutional repository. While DSpace provides strong support for preservation and item versioning, maintaining stable access to files themselves has historically been more difficult, especially when files need to be updated or replaced over time.

In standard DSpace behavior, replacing a file creates a completely new bitstream object with a new identifier. Although the repository item itself remains available, direct links to the original file will no longer point to the updated content. This becomes even more problematic in versioned workflows, where multiple versions of an item may contain related files that DSpace treats as entirely separate objects.

As repositories increasingly manage evolving content these limitations create challenges for long-term file persistence and reliable access.

This presentation discusses a set of recent and future contributions that improve how DSpace handles file replacement and version continuity. The implemented solution allows DSpace to explicitly track when one file replaces another across item versions. Rather than treating replacement files as unrelated objects, the repository can now preserve continuity between older and newer versions of the same file.

As a result, references to older files can transparently resolve to valid replacement files, improving long-term accessibility and reducing fragmentation in repository content. The contribution helps align DSpace behavior more closely with the real-world needs of repositories that manage continuously evolving scholarly materials.

The presentation will explain the underlying challenges, describe the technical approach that was implemented, and demonstrate how the solution behaves in practical repository workflows.

This presentation is intended for both technical and non-technical audiences. While we will address some technical aspects we will ensure that non-technical attendees can understand the technicalities in a conceptual manner. 

Break / Networking

Collective Action for Accessibility: Improving Content Accessibility in Texas DSpace Repositories

Presenter: Elliot Williams (University of Texas San Antonio)

The updated Americans with Disabilities Act (ADA) Title II rule, which requires expanded compliance with web accessibility standards, has extensive impacts for libraries hosting digital content in DSpace repositories. This presentation will report on the work of a Texas Digital Library (TDL) working group charged with helping repository managers prepare for compliance with the new rule.

The TDL DSpace Users Group is made up of repository managers who manage DSpace repositories hosted by TDL, as well as institutions who host their own repositories. In Fall 2025, the Users Group convened the Repository Content Accessibility Working Group to address compliance of the content contained in TDL DSpace repositories, and charged it with developing a collective plan of action for TDL members. The working group focused specifically on the content in repositories, rather than the repository interface itself.

To meet that charge, the working group developed a content accessibility toolkit for repository managers, conducted a survey of current accessibility practice among TDL members, and drafted recommendations to help guide repository managers as they work to improve the accessibility of their repositories.

This presentation will provide an overview of the Repository Content Accessibility Working Group; share the outcomes of the group’s work, including the accessibility toolkit and survey of accessibility practices; and report on planned future work.

Update from DCAT

Presenter: Kent Gerber (Chair - DSpace Community Advisory Team)

Update from DSpace Leadership Group

Presenter: Erik Moore (Chair - DSpace Leadership Group)

Breakfast (provided)

Breakout Groups

From Problem to Proof: AI-Assisted Verification for DSpace

Presenter: Jozef Mišutka (Dataquest)

This session presents a practical model for moving from an unclear repository problem to reproducible evidence and accountable approval. Using AI-assisted, semi-automated verification, repository staff can help determine whether a problem is still present, whether a fix actually works, and whether important behavior survived an upgrade or other change.

Using one practical DSpace example, we will show how AI can clarify the expected outcome, guide both technical and non-technical staff through reproducing the issue, support repeatable browser and system checks, and compare the observed result with what was requested. A successful check alone is not enough. Can it also demonstrate the original failure? Can repository staff repeat the verification later without relying entirely on a developer? Can a manager understand what was tested, what passed, and what remains uncertain? We will also show where AI accelerates verification, where it can produce convincing but incorrect conclusions, and why repository-specific knowledge and human approval remain essential.

Putting ScholarWorks on the Map: Testing a Pipeline from DSpace to GeoBlacklight

Presenter: Erin Jerome (UMass-Amherst) 

Part of the University of Massachusetts Amherst Libraries’ justification for migrating its institutional repository, ScholarWorks@UMassAmherst, from Digital Commons to DSpace was that DSpace could support increased integration between the various library-administered systems that our patrons use to discover content. This presentation focuses on one specific integration project: the creation of a possible pipeline between the data repository, housed in ScholarWorks, and our geoportal platform, UMAP GeoData. Powered by the open-source GeoBlacklight software, UMAP GeoData acts as a discovery layer that points patrons to items within the Libraries’ collections that have a spatial component, such as GIS datasets, maps, and datasets with latitude and longitude metadata values. This presentation will discuss the cleanup, modification, and standardization of metadata required to make spatial data submissions more obvious to repository administrators, as well as the changes we made to the data repository in order to improve the self-submission user experience.  We will also explore our proposed workflow for automating the process of generating and ingesting metadata into the geoportal, converting Dublin Core to the OGM Aardvark schema. The workflow that we have envisioned will enable users to deposit spatial assets into ScholarWorks and have them seamlessly appear in a geoportal search.  We are optimistic that this integration will improve our usage statistics and the discoverability of our spatial data. 

Managing Multiple Batch Submission Pipelines

Presenter: Sadie Roosa (MIT)

We have many different ways new items get delivered to be submitted into the repository, most of which can be processed in batches. However, not all of the batches are the same. We have batches that come in with files and metadata already in a spreadsheet. We have files of journal articles that come in and just have DOIs. We have files of digitized reports and theses that come in and already have MARC catalog records. With all of these different situations, over the years we had built up several different tools to handle each one. Over the past few years, MIT Libraries has worked towards a single set of tools that can handle all of these different pipelines, making it more flexible and sustainable. This presentation will give an overview of how the tools work and describe the variety of submission pipelines our new tools can process.

Closing & Appreciation

Lunch (on your own)

For those waiting for flights, you are welcome to remain and use the Lamont Forum Room until 5:00 pm.