Rethink the Link: How to accomplish Linked Data on a budget
Wednesday, April 5th: 1:00pm - 2:30pm
David NewburyHead of Software @ Getty
Over the past five years, Getty has built out a comprehensive suite of linked data applications and infrastructure—the tools needed to try out the technologies that we\’ve all been writing about over the past decade and see if they work in practice.
Spoiler: they do.
However, some of the parts that we thought most important turned out to be inconsequential, and other parts were critical in a way that was unexpected as we began. In particular, the benefits of linking over semantics, the power of reconciliation, the reuse of off-the-shelf tools, and the importance of local expertise over global knowledge.
These will be framed within a discussion of the six levels of linking within Linked Data:
1. Authority: Providing a consistent way to identify both entities and the institution providing information
2. Reconciliation: using authorities and thesauri to disambiguate between similar real world entities
3. Bi-directional linking: making and publishing connections between systems and institutions
4. Aggregation: Enhancing discovery by providing search and access to information across collections
5. Interoperability: Developing interfaces that present information from many sources in a single way
6. Reuse: Allowing one institution to import information from another while maintaining provenance
In this paper, I will describe both the lessons learned in trying to implement each of these six levels, using Getty\’s archives, collections, and vocabularies as our case study, and I will point to practical, low-cost, low-effort tools and techniques that can be implement much of what we learned at other institutions, particularly ones that can\’t muster the scale that Getty can.