Nomisma extended to link to the Getty ULAN, British Museum thesauri

This morning, the Getty Museum announced the release of the Union List of Artist Names (ULAN) thesaurus as linked open data. Following updates made to the Getty lookup mechanism in Kerameikos.org's XForms-based editing interface, I have ported these updates into Nomisma's editor. It is now possible to quickly and easily link people and organizations in Nomisma to matching concepts in the ULAN, much like what we have already implemented for linking mints and regions to the TGN and denominations, materials, manufacture methods, and object types to the AAT.

Furthermore, I took the time to finally fully implement the British Museum lookup mechanism in Nomisma. The British Museum thesauri cover many of the same broad categories as the Getty, but the main difference between the two systems is that the BM thesauri reflect what they have in their own database, and the Getty thesauri are aimed at representing concepts across all of art history. The Nomisma editing interface now enables quick and easy linking to denominations, ethnic identities, manufacture methods, materials, mints, object types, people, and regions in the BM thesauri.

See the following examples:

http://nomisma.org/id/drachma
http://nomisma.org/id/alexander_iii
http://nomisma.org/id/antiocheia_syria

When creating or editing ids manually, we'll be able to easily add BM and Getty URIs into the system. Ideally we will want to do a mass update of all of our ids that can be mapped to concepts in both thesauri. Creating a concordance between Nomisma and Getty concepts will hopefully facilitate large-scale aggregation in the future.

Major photographic updates to OCRE

Much progress has been made recently in photographically covering the American Numismatic Society collection for the Roman imperial department. By using some NEH funding to hire a photographer to focus mainly on Roman imperial coins in a more high-speed workflow, photos for more than 7,000 Roman imperial coins have been captured over the last six months or so.

These images have finally pushed through the main image processing workflow and are now available online. The Nomisma.org SPARQL endpoint has been updated with the latest dump from Mantis to reflect these additions. Furthermore, more than 1,100 new links to OCRE have been made from the ANS collection from RIC Volumes I to IV. Nearly 15,300 coins from the ANS are now linked in OCRE, up from just over 14,000 previously. Furthermore, the photographic coverage has been extended from 3,499 coins in our collection to about 12,000.

Lastly, we have pushed the first portion of RIC Volume V into OCRE. These are the coins from during Valerian's life, from his sole reign to joint reign with his son, Gallienus. We have not yet linked physical specimens from the ANS or other museum collections into these RIC V URIs yet, but look for this to be done in the next week or two. For now, you can take a look at the types at http://numismatics.org/ocre/results?q=recordId:ric.5*.

From 0 to 60 on SPARQL queries in 50 minutes

On Wednesday, May 13, at 10 AM EDT, I will be giving a free webinar sponsored by the Dublin Core Metadata Initiative and the Association for Information Science & Technology.

This webinar provides an introduction to SPARQL, a query language for RDF. Users will gain hands on experience crafting queries, starting simply, but evolving in complexity. These queries will focus on coinage data in the SPARQL endpoint hosted by http://nomisma.org: numismatic concepts defined in a SKOS-based thesaurus and physical specimens from three major museum collections (American Numismatic Society, British Museum, and Münzkabinett of the Staatliche Museen zu Berlin) linked to these concepts. Results generated from these queries in the form of CSV may be imported directly into Google Fusion Tables for immediate visualization in the form of charts and maps.

This webinar was first presented as a training session in the LODLAM Training Day at SemTech2014. I will cover the basics of an RDF triple, but the presentation assumes some baseline understanding of linked data. The SPARQL queries will begin simply and build in complexity, culminating in spatial queries.

Sign up at http://dublincore.org/resources/training/#webinars