Comparing collections practice across the pond

Claire Sussums, Assistant Director for Content at the Museum of London, recently caught up with friend and former colleague Frances Lloyd-Baynes, Head of Collections Information Management at the Minneapolis Institute of Art (Mia), to compare their experiences of more than two decades managing collections on either side of the Atlantic.

Map of the world
Embroidered map sampler, 1783, Metropolitan Museum of Art, PD.

Claire Sussums: The UK cultural heritage sector benefits hugely from the unifying force of Collections Trust’s Spectrum standard. In contrast, the US cultural heritage sector has not adopted Spectrum or an equivalent. What do museums use instead?

Frances Lloyd-Baynes: There are a number of standards used when defining data structure, for example Getty’s Categories for the Description of Works of Art and VRA CORE, and for controlling vocabulary, such as the Getty Vocabularies and Chenhall’s Nomenclature, but there is no generally agreed-upon approach to either.
US museums tend to develop procedures independently, following advice from publications by the American Alliance of Museums, such as Museum Registration Methods. Each museum also develops data standards based on their particular needs, for example using standardised vocabulary tools for some purposes and locally produced tools for the rest. This involves significant effort and can result in duplication of work and a lack of consistency across the sector. Most also seek advice from colleagues: list-servs and discussion forums serve an important role in communicating best practice.
As I have experience of implementing the Spectrum standard in the UK, I use its units of information as clear examples of best practice when developing Mia’s data standards. Given Spectrum’s increasingly international adoption, this practice also helps to align Mia’s collections content more readily with a global data set.

CS: What are some of the methods you’re using to open up Mia’s collections information to an international audience?

FL-B: Using Linked Open Data (LOD) to achieve a global connection to collections information is getting some limited attention in the US and the UK. A few US museums, including the Yale Center for British Art and the American Art Collaborative members, have created full LOD by mapping data to the CIDOC CRM. At Mia we’ve been exploring LOD and linking our content to international authorities such as the Virtual International Authority File and the Getty’s Union List of Artists Names. We’re sharing our collections on the museum’s website and our complete metadata set on GitHub for anyone to use, as well as participating in Google Arts and Culture and the Digital Public Library of America. We unfortunately do not have the resources to develop LOD at present so, in common with a number of UK museums, we are instead building the foundations for it by cleaning and standardising the museum’s data. This is a great positioning strategy for future development.

CS: When Google arrived, our profession appeared at risk as some predicted the extinction of structured data (ie defined, organised data as opposed to free text). Have you found this to be proving true?

FL-B: Thankfully the value of structured content is being rediscovered. Structured data is important for machine readability and thus the discoverability of collections. It can also impart meaning, enabling users to match ‘concepts’ not just ‘words’ in their searches. My colleagues and I are at the moment simultaneously contributing structured content to Wikidata (from Mia’s artist records) and also mining it for data missing from our records, creating a virtuous circle.

CS: How documentation specialists record collections information can also play an important role in tackling issues around diversity and inclusion. How is this being played out in the US?

FL-B: US museums are working to understand their inherent biases and diversify their organisations, as well as the perspectives they document and share. I am currently leading a working group to codify why, what and how Mia will document our artists’ gender and ethnicity. These are very personal aspects of creators’ lives, so such decisions require care and thoughtfulness, and must be part of a larger process. We must allow creators to speak for themselves, by asking artists to self-identify, and we need to suppress the urge to prioritise a complete set of data over someone’s desire to remain ‘undocumented’. The decisions we make will affect wider museum processes and procedures, for example determining which pronouns to use in a gallery label or what information to make discoverable on our website.

CS: How do you feel the landscape of collections management today compares to that of two decades ago?

FL-B: On reflection, it feels to me like many of the current challenges and opportunities for collections data, in both the UK and US, are similar to those we have faced over the past 20 years, despite the huge advances in technology: structured data remains important, terminology is key, and the role of documentation in museums–particularly given the responsibility to address equality and inclusion–is vital, and there are plenty of opportunities ahead to keep our work engaging.


Be the first to read our next new blog post by signing up to our monthly newsletter.