Meeting Report On 3-5 Year DSpace Vision
Winchester, MA Early in May DuraSpace convened a group of DSpace stakeholders to continue discussions that began last fall about the future of DSpace. The DSpace meetings were hosted by University of Illinois Chicago. These meetings addressed the need for additional investment of both financial and developer resources in DSpace. Grant support is no longer available to support the project and current funds do not cover the ongoing maintenance, improvement and management of DSpace. DuraSpace makes up the shortfall in costs for maintaining DSpace.
Participants were asked to draft a vision statement for DSpace, and a high-level roadmap for the next 3-5 years of software development. The group discussed improvements and what types of services and functionality users are looking for to meet current and emerging needs. The aim was to develop a document that would inspire DSpace enthusiasts to take an active role in bringing a new and improved DSpace into the future.
In two half days of discussion the group reached interim consensus on these points:
• DSpace was architected ten years ago to fulfill the vision of what an institutional repository was then thought to include.
• Users have been extending DSpace ever since to meet requirements that it was never designed for, often as one-off, local solutions. Over time, the codebase has grown to be monolithic.
Attendees agreed that the best course moving forward may be to:
• Re-architect DSpace as a small, flexible core with a set of defined functions. For example, DSpace may strive to meet the primary Institutional Repository use cases, while also including improvements such as more flexible relationships between objects and hierarchical metadata support. Improvements could include functionality like a more flexible relationship between objects, hierarchical metadata.
• Define a set of external functions which can be deployed easily as extensions, built and shared by the community. For example, DSpace may support extensions for delivery of media (streaming), alt-metrics (downloads, tweets, etc.), and enhanced statistics. Potential extensions could include support for data sets, metadata extensibility, richer licensing, customizable/flexible user interface.
• Ensure that the DSpace core can easily interoperate with other third party services and applications that would extend DSpace functionality. For example, DSpace may integrate with third-party tools or services to provide advanced digital preservation activities. Potential interoperability could include support for delivery of media, improved statistics and alt-metrics (downloads, tweets, etc.).
• Make sure a newly re-architected DSpace can run locally or as a hosted service.
Next steps will be to:
• Invite a wide range of community stakeholders to provide input
• Refine and document the vision
• Investigate alternatives for implementation
• Define a more detailed roadmap for proceeding
• Formalize leaders from community to direct the effort (with support from DuraSpace)
• Enlist community contributors to support the new effort
Concurrent with this work there is an effort underway to create a representative governance model that is a good fit for the DSpace community. Effective governance will ensure a broad base of decision-making among stakeholders, facilitate accountability, and increase the pool of resources available to the project.
For more information on the recent DSpace Vision meeting, see https://wiki.duraspace.org/display/DSPACE/DSpace+2013+Vision+and+Roadmap+Meeting. To express an interest in being part of these discussions send an email to Tim Donohue firstname.lastname@example.org, or Jonathan Markow email@example.com. These themes will also be explored at this year's Open Repositories 2013 Conference which will be held July 8-12, Charlottetown, Prince Edward Island, Canada: http://or2013.net.