How Does The Met Maintain Half a Million Web Pages Devoted to Art?

«One in five people who come to the Museum's website visit a page devoted to a single work in The Met collection (what we call an "object page"). To name a couple of examples demonstrating the breadth of The Met's online collection, Claude Monet's Bridge over a Pond of Water Lilies happens to be the most popular object page, while a much less frequented page is devoted to an Egyptian petrified lemon from the 3rd to 12th century. Whatever the subject, each object page serves as the most comprehensive source of truth for all available information about that work, and The Met aspires to make it the most accessible and inspiring online presentation possible.»

Claude Monet (French, 1840–1926). Bridge over a Pond of Water Lilies, 1899. Oil on canvas, 36 1/2 x 29 in. (92.7 x 73.7 cm). The Metropolitan Museum of Art, New York, H. O. Havemeyer Collection, Bequest of Mrs. H. O. Havemeyer, 1929 (29.100.113)

Few visitors know that many of the 450,000-plus object pages are updated nightly with new information, scholarship, and images, all thanks to the ongoing work of hundreds of Museum staff dedicated to supporting the digitization and online representation of the Museum's collection.

To that end, it should be no surprise that object pages are an area of focus for the Digital Department's Collection Product Development team. This team is responsible for enhancing The Met collection's experience, reach, and relevance online through user-focused software development. In other words, our work entails maintaining and improving the functionality for this area of the website, as well as the technologies that power the online collection, to ensure that it is reliable and performs well.

Our team recently built an API [Application Programming Interface] for the collection, which now serves up high-resolution images, audio commentary, extensive scholarship and data, and descriptions to all object pages, and serves as a foundation for our newly adopted approach to the object pages. Up until now, every time an app or website has been built to present the collection—from the dozens of digital narrative features found in the galleries, to The Met app and the Heilbrunn Timeline of Art History—each of these discrete products has had to do some heavy lifting to intelligibly present basic information such as artist name, biography, title, date, geography, culture, medium, and dimensions, all of which are originally sourced from the Museum's collection-management system. Now, the collection's API will do the heavy lifting for all future applications, and because it is well documented and supported by frequent testing, it is stable, reliable, and easy to use.

So much art-historical content about artworks can be found within these content drawers on every object page. In this example, users can click to expand these sections on the Bridge over a Pond of Lilies object page—which include a rich catalogue entry, the work's complete exhibition history, and MetPublications discussing the work. All of this content is now served by the new collection API.

The development of an API is an important milestone for any modern collecting institution because it is an investment in a sustainable future. When the Cooper Hewitt Smithsonian Design Museum publicly released its API in 2014, then-Director of Digital and Emerging Media Seb Chan eloquently likened it to a city's "vast, labyrinthine sewer systems," a long-term infrastructural undertaking that provides a foundation for future health and growth. This is exactly how we see our API at The Met, and our team will be leveraging it as the core of the work we deliver for the collection experience on the website.

Some collecting institutions like the Cooper Hewitt, the San Francisco Museum of Modern Art, and the Harvard Art Museums, for example, have made their APIs available to the public. Software engineers who work with public collections may wonder why The Met's collection API is not available for their use. This is because The Met collection and its associated assets are quite vast, so it is a priority for the team to provide strong and stable technical support for the API for vital internal use first, before we expand access to meet additional demands. This approach enables us to monitor the performance of the API and scale responsibly.

Likewise, now that the API neatly serves content to every object page, the team will increasingly focus on improvements to the user experience. We iteratively release features every two weeks, and aim to incrementally improve the experience every time. For instance, we recently added interpretive audio messages to select object pages. Soon, you will see a new section added to the collection's landing page that will make it easier to browse and discover objects in the collection, and users from around the world will soon gain access to a set of object pages in 10 languages.

Each of these features is designed to be a small, simple release with the potential to be developed further depending on how well it performs and meets the needs of our audience. We cannot wait to share more about our work as it unfolds!

The development of The Met's collection API is made possible through the continued generous support of Bloomberg Philanthropies.