you're reading...

Collections and Connections

Picking apart the alpha Cooper-Hewitt online collection

Cooper-Hewitt is USA’s National Design Museum in New York. They recently launched the alpha version of their new online collection site. Over the last few weeks I have found myself returning to it again and again because of a number of unusual features. In part this is because I’m a museum techie and I’m curious about the nuts and bolts that hold it together, but it is also because of the excellent use they make of their somewhat patchy data.

This post is a chance for me to jot down some of the points that make this site stand out from many other museum collection sites. Micah Walter from Cooper-Hewitt has posted on their labs blog the thinking that went into the new alpha site and Suse Cairns wrote about the alpha site in her post: “A museum collection that never ends?”.

Alpha as a wireframe

In a sector where ‘polished before it goes public’ is the default, it’s great to see an early draft of where they are headed with their online collection site. It works as a public wireframe – a rough but functional demonstration of the key flows within the site. The public get to see some of the new features nice and early and Cooper-Hewitt get responses and metrics to help decide what works and what doesn’t.

Natural language and playfulness

Museum catalogue records can make for heavy reading. Cooper-Hewitt have taken a lighter approach, introducing some playful language (‘This object is resting in our storage facility’) and combining the key fields into a more readable sentences (like in this example of a spoon).

This is similar to how Flickr presents some metadata as a complete sentence. E.g. “This photo was taken on February 26, 2011 in West Coast, NZ, using a Canon PowerShot S80” (where the date, place and camera model are metadata fields).

Twitter Cards

The site makes use of the recently added Twitter Cards, which allow tweets to show in-line citations for the content they link to. In this case if you add a link to an object record you will see the object image and key fields in the tweet, plus links through to the Cooper-Hewitt Twitter account. I liked this so much we added the meta tags into eHive, but now we have to wait for Twitter to approve us for participation in the Twitter Card programme.

Making use of the data

The site uses the data well to present a 10,000′ view of the collection. Groupings like media and periods show counts of how many objects fall into each category. The data is interlinked in so many ways that the visitor never reaches a dead-end. They click from one related object, result set or term to another, discovering previously buried objects along the way.

Many objects have no image and the object description is displayed in the image area of the page in a large font. This works surprisingly well, probably because of the strangely poetic way that many objects have been described.

When it comes to the data, the site makes a wise move by standing on the shoulders of giants. Key fields such as artist/maker names and places are now linked to external sources like Wikipedia and Freebase. Why write your own page about each country when you can build upon Wikipedia’s text.

What else do I want to see?

The data is indeed messy, so that can only get better. It would be great to see images in the results views and have a way to filter the results to just records that have images. I’d like to try tagging Flickr photos that relate to their objects, but I’d need an easier way to browse the Cooper-Hewitt images first.

Comments on objects would also be a useful later addition but could just be a distraction for Cooper-Hewitt team in the alpha phase.

All in all, it is a bold site, prototyping in public (and as of today, there’s an experimental first version of a collection API).




No comments yet.

Post a Comment

Recent Tweets (@armchair_caver)