Among the collections in the Brooklyn Museum are found eleven bronze sculptures by Rodin, five seminal Judy Chicago works, two Edward Hopper iconic paintings, and now one open-to-the-public API, with no admission fee. The API allows for programmatic search of the digitized collections of the museum, which encompasses more than 10,000 individual works (more details at our new Brooklyn Museum API Profile).
The museum’s Chief of Technology Shelley Bernstein notes in the introduction that “we are an institution with a community-minded mission, we recognize that developers are a significant part of community on the web and one we’d like to work with moving forward. So, you may be wondering what kinds of applications could benefit from our data. There’s a wonderful X factor to all this – that we just don’t know what interesting something that someone will come up with – so it is exciting to wait and see.”
Enterprising developers have already built a Flex-based image browser shown above (our Brooklyn Browser Profile) and some Yahoo Pipes to perform the searches without code. Sample applications given on the site show basic searches by artist, title, or other keyword, and a timeline of objects from 6000 B.C. to the present.
The API is for non-commercial use with a limit of 3000 API calls a day. Naturally the museum must be respectful of artist copyrights, and requires proper attribution for any display of results. The API also permits only session-based caching with no retention of copies, and the images returned are no greater than 500 pixels in width or height, with many that may be smaller.
The API has a simple REST format with three methods – search, getitem (given the id returned from a search), and getimages for a particular item. The search can be filtered by date, keyword, and whether or not there are images present. In the documentation’s search example for keyword Eames, the API returns an object (the iconic Eames chair), the artists (Ray and Charles Eames, and the company that produced it), and the image. Other information available for works are the dates of creation and purchase, markings, classification, tags, descriptions and image captions. The return format can be either XML, JSON, or HTML.
The Brooklyn Museum has been active in other technical fronts, including the use of Twitter to feature tweets by contemporary artists every month, and a tagging game somewhat like Google’s, to help with identifying images. Other museums have been organizing to rationalize the data constructs appropriate for museum collections. As museums join other institutions in opening up their reservoirs of knowledge and culture, the Brooklyn Museum will be one to watch.