Museum Data Exchange – Report Executive Summary

The final report of the Museum Data Exchange grant will be released on the OCLC Research website later this month. As a first impression of key outcomes, I’ve posted the executive summary below. Stay tuned!


The Museum Data Exchange, funded by the Andrew W. Mellon Foundation, brought together a group of nine museums and OCLC Research to create tools for data sharing, build a research aggregation and analyze the aggregation. The project established infrastructure for standards-based metadata exchange for the museum community and modeled data sharing behavior among participating institutions.

The tools created by the project allow museums to share standards-based data using the Open Archives Initiative Protocol for Metadata Harvesting (OAI-PMH).

  • COBOAT allows museums to extract Categories for the Description of Works of Art (CDWA) Lite XML out of collections management systems
  • OAICatMuseum 1.0 makes the data harvestable via OAI-PMH
  • COBOAT’s default configuration targets Gallery Systems’ TMS, but can be adjusted to work with other vendor-based or homegrown database systems.

    Both tools are a free download from here.
    Configuration files adapting COBOAT to different systems can be shared here.

    Data Harvesting and Analysis
    Harvesting data from nine museums, the project brought together 887,572 records in a non-public research aggregation, which participants had access to via a simple search interface. The analysis showed the following:

  • for CDWA Lite required and highly recommended data elements, 7 out of 17 elements are used in 90% of the contributed records
  • the match rate against applicable Getty vocabularies for objectWorkType, nameCreator and roleCreator is approximately 40%
  • the top 100 objectWorkType and nameCreator values represent 99% and 49% of all aggregation records respectively.
  • Significant improvements in the aggregation could be achieved by revisiting data mappings to allow for a more complete representation of the underlying museum data. Focusing on the top 100 most highly occurring values for key elements will impact a high number of corresponding records, and would be low-hanging fruit for data clean-up activities.

    For further analysis, the research aggregation will be available for 3rd party researchers under the terms of the original agreements with participating museums.

    In its relatively short life span to date, the project’s suite of tools has catalyzed several data sharing activities among project participants and other museums:

  • The Minneapolis Institute of Arts uses the tools in a production environment to contribute data to ArtsConnected, an aggregation for K-12 educators
  • The Yale University Art Museum and the Yale Center for British Art use the tools to share data with a campus-wide cross-search, and contribute to a central digital asset management system
  • The Harvard Art Museum and the Princeton University Art Museum are actively exploring OAI harvesting with ARTstor. (Three additional participants have signaled that this would be a likely use for their OAI infrastructure as well.)
  • Participating vendors contributed to the museum community’s ability to share:

  • Gallery Systems extended COBOAT for EmbARK, demonstrating the extensibility of the MDE approach
  • Selago Design created custom CDWA Lite functionality for MIMSY XG, freely available to customers as part of their OAI tools
  • An increasing number of projects and systems using CDWA Lite / OAI-PMH as a component (for example OMEKA, Steve: The museum social tagging project, CONA™) can be seen as a leading indicator for the future need of data sharing tools like the ones created as part of the Museum Data Exchange. When there are applications for sharing data which directly support the museum mission, more data is shared, and museum policies evolve. Conversely, when more data is shared, more such compelling applications emerge.