DPLAfest 2016 has ended
Taking place in the heart of Washington, DC, DPLAfest 2016 (April 14-15) will bring together hundreds from DPLA’s large and growing community for interactive workshops, hackathons and other collaborative activities, engaging discussions with community leaders and practitioners, fun events, and more. DPLAfest 2016 will appeal to anyone interested in libraries, technology, ebooks, education, creative reuse of cultural materials, law, open access, and genealogy/family research.

Area institutions serving as co-hosts include the National Archives and Records Administration, the Library of Congress, and the Smithsonian Institution.

To view photographs, recordings, and social media from DPLAfest 2016, visit https://dp.la/info/get-involved/dplafest/april-2016/media/.
Back To Schedule
Friday, April 15 • 9:30am - 10:15am
Testing a Linked Data Fragments Server using DPLA Data & PCDM, IIIF, and Interoperability

Sign up or log in to save this to your schedule, view media, leave feedback and see who's attending!

This session will contain two presentations:

Testing a Linked Data Fragments Server using DPLA Data

Based on Ruben Verborgh Linked Data Fragments server, this session is about the latest experiments on implementing a Linked Data Fragments server as an alternative to fully supported SPARQL endpoint. This semantic web system research uses Redis as the backend datastore cache with a server prototype in Python and an initial production server being developed in the Go language. This talk will illustrate the general approach, first demonstrated in Jeremy Nelson's Pycon Japan talk in October 2015 (http://intro2libsys.info/pycon-jp-2015/) , of incrementally improving and testing a linked data fragments server that provides RDF triples in JSON when queried by web clients. This talk will show how DPLA's JSON data is transformed into raw Redis protocol for extremely fast importing of data at scale into a Redis cluster. The talk will then show how by leveraging the flexibility, speed, and caching functionality of DPLA RDF data, Redis enables a linked data fragments caching and distributed server topology that scales while offering faster and more stable retrieval of RDF triples from a sample of DPLA's RDF data. This talk finishes by comparing the performance of a new Go language-based production server versus the Python prototype and future plans for this promising technology.

PCDM, IIIF, and Interoperability

The Portland Common Data Model (PCDM) provides a flexible vocabulary for describing repository content, and aims to allow tools and applications to work with data that have varying metadata standards and levels of complexity. The International Image Interoperability Framework (IIIF) provides standard APIs for describing and delivering images, allowing implementers to develop feature-rich applications while still working with the technology of their choice, and enabling users to integrate content from multiple repositories into a single viewer. Beyond their focus on interoperability, these projects also share a pragmatic approach centered on use cases and community engagement. This talk will describe PCDM and IIIF, and draw lessons about how building community and staying grounded in real world use cases help create standards and tools that work. 

avatar for Jeremy Nelson

Jeremy Nelson

Metadata and Systems Librarian, Colorado College
Jeremy Nelson is the Metadata and Systems Librarian at Colorado College, a four-year private liberal arts college in Colorado Springs. In addition to working 8 hours a week on the library's research help desk, providing information literacy instruction to undergraduates, and supervising the library's systems and cataloging departments, Nelson is the... Read More →

Friday April 15, 2016 9:30am - 10:15am EDT
National Archives and Records Administration: Room G-25 700 Pennsylvania Ave NW Washington, DC 20408