Washington D.C., Nov. 6, 2007 — The National Science Digital Library (http://nsdl.org) Core Integration team at Cornell University led by Dean Krafft presented an overview of NCore, an open source suite of technologies and standards that power the National Science Digital Library’s (NSDL) core infrastructure, at NSDL’s Annual Meeting held in Washington D.C. November 6-8, 2007. View the presentation.

NCore technologies and standards replace NSDL’s original metadata records-based data paradigm allowing for greater flexibility in collaborating and creating context around library resources. In the initiative’s first phase, completed in January 2007, NSDL Core Integration successfully implemented existing Library services on top of NCore—a change that was largely invisible to users. Now, NSDL is leveraging what the Fedora-based NCore platform has to offer by introducing next-generation library services and collaborative tools.

MANAGING AND MAINTAINING DIGITAL LIBRARIES WITH NCORE

NCore is a general platform for building semantic and virtual digital libraries united by a common data model and interoperable applications. DLESE, the Digital Library for Earth Science Education (http://dlese.org) ported their existing infrastructure to NSDL using NCore. By overlaying their data model on top of the NCore model, developers enabled specialized DLESE services to continue without loss of functionality, and to co-exist alongside of, but independent from, NSDL.

The NCore API was used to build the DLESE index by reading DLESE metadata from the Fedora-based NSDL Data Repository (NDR), using the NDR API, and then building the index. The key technical point in this step was to extract the required metadata from NDR objects.

The DLESE Search Service is run from the Index, and is used to power the DLESE library search and browse functionality. Use of the NSDL Data Repository as the primary metadata repository required no modifications to DLESE Search Service. The indexes and search service remain the same, which means the DLESE Library interface and all other interfaces that use the NCore service continue to operate without modification.

INSIDE NCORE

The NCore platform consists of a central repository built on top of Fedora, a data model, an API, and a number of fundamental services such as full-text search or OAI-PMH. Innovative NSDL services and tools that empower users as content creators are now built on, or transitioning to, the NCore platform. These include: the Expert Voices blogging system (http://expertvoices.nsdl.org/);the NSDL Wiki (http://wiki.nsdl.org); the NSDL OAI-PMH metadata ingest aggregation system; the OAI-PMH service for distributing public NSDL metadata; the NSDL Collection System (NCS), derived from the DLESE Collection system (DCS); the NSDL Search service, and the OnRamp content management and distribution system (http://onramp.nsdl.org).

Because NCore is a general Fedora-based open source platform useful beyond NSDL, Core Integration developers at Cornell University have made the repository and API code components of NCore available for download at the NCore project on Sourceforge. Over the next six months, NSDL will release the code for major tools and services that comprise the full NCore suite on SourceForge.

The NCore project aims to support NSDL’s continued growth as a semantic and virtual digital library providing open source educational cyberinfrastructure services and tools in support of teaching and learning, while meeting the needs of new, existing, and evolving digital libraries. For more information contact the NCore team at Cornell.

This material is based upon work supported by the National Science Foundation under Grants No. 0227648, 0227656, and 0227888. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the authors and do not necessarily reflect the views of the National Science Foundation.