DPLA MAP, version 4.0

DPLA MAP, version 4.0
Posted by Gretchen Gueguen on March 5, 2015 in Hubs, News & Blog, Staff Posts and tagged , .

Hot on the heels of last week’s announcement of KriKri and Heidrun, we here at DPLA HQ are excited to release the newest revision of the DPLA Metadata Application Profile, version 4.0 (DPLA MAP v4.0).

What is an “application profile”? It’s a defined set of metadata properties that combines selected elements from multiple schemas, often along with locally defined ones. An application profile, therefore, allows us to take the parts of other metadata schemes best suited to our needs to build a profile that works for us. We’ve taken full advantage of this model to combine properties from DCMI, EDM, Open Annotation, and more to create the DPLA MAP v4.0. Because the majority of the elements come from standard schemas (indicated by a namespace prefix, such as “dc:date” for Dublin Core’s date element), we remain aligned with the Europeana Data Model (EDM), while having enough flexibility for our local needs.

Our new version of the DPLA MAP has lots of properties tailor-made for storing Universal Resource Identifiers (or URIs) from Linked Open Data (LOD) sources. These are other data sets and vocabularies that publish URIs tied to specific terms and concepts. We can use those URIs to point to the external LOD source and enrich our own data with theirs. In particular, we now have the ability to gather LOD about people or organizations (in the new class we’ve created for “Agents”), places (in the revision of our existing “Place” class) and concepts, topics, or subject headings (in the new “Concept” class).

At the moment DPLA’s plans for LOD include associating URIs that are already present in the records we get from our partners, as well as looking up and populating URIs for place names when we can. In the future, we plan to incorporate more linked data vocabularies such as the Library of Congress Subject Headings and Authorities. After that we can begin to consider other kinds of LOD possibilities like topic analysis or disambiguation of terms, transliteration, enrichment of existing records with more metadata from other sources (a la Wikipedia, for example), and other exciting possibilities.

Every journey begins with a first step, and our journey began with the upgrades announced in recent weeks (as described in our recent Code4Lib presentation, blog posts, and software releases). Along with these upgrades, MAP v4.0 has become our official internal metadata application profile. As of today, documentation for the new version of DPLA MAP v4.0 is available here as well as a new Introduction to the DPLA Metadata Model.