Visualizing Collections Using Generous Interfaces

A few years ago Mitchell Whitelaw, a professor in the School of Art and Design at the Australian National University, coined the term “generous interfaces” to describe next-generation, data-driven browse interfaces for digital cultural collections. Generous interfaces create greater engagement with audiences and go far beyond the still pervasive search interfaces most collection websites are based on.

Faces of Sweden, a timeline-centric exploration interface created by the Swedish National Heritage Board

Whitelaw describes a scenario where you visit a museum for the first time and only have limited knowledge of the collection. The attendant at the front desk asks you to enter a query on a piece of paper, and then retrieves matching objects in sets of 10 at a time for you to view. This experience is basically analogous to a search interface. Search works well when you know what you are looking for and just need help finding it. On the other hand, search is an extremely poor experience for discovery and exploration — a “stingy interface” to use Whitelaw’s term. What if instead of the above scenario, you could explore rooms filled with objects on display, organized by topic and period, with engaging captions and docent tours? That would constitute a more “generous” approach, and lead anyone to engage on a much deeper level with a collection.

Example of a generous interface created for the Prints and Printmaking Collection of the National Gallery of Australia by Mitchell Whitelaw and Ben Ennis Butler

For a variety of reasons, museum and library digital collections are still mostly based on search interfaces that require prior knowledge and motivation on behalf of audiences. This is one of the reasons that most digital cultural collections receive very little engagement. Even the more visual digital collections websites that feature large images, interpretive descriptions, video and other media don’t usually do a good job of orienting the viewer, forcing them instead to browse through hundreds of records, arbitrarily looking for something they might be interested in. How is it that collections with so many layers of knowledge and rich, compelling stories can’t even compete with mundane e-commerce websites in terms of engagement and usability, let alone drive people to explore and discover knowledge of their own accord?

While most would agree that they lead to deeper engagement, one of the problems with generous interfaces is that in the past they have been time consuming and expensive to build. Interfaces needed to be custom built so that they would be specifically suited for a particular collection. As David Haskiya of the Swedish National Heritage Board writes in An Evaluation of Generous Interfaces, “at the heart of the very concept is that [generous interfaces] should be intimately connected to the very nature of the collection they’re exploring. This means that creating generous interfaces is more bespoke than search. Bespoke development requires money, or an in-house design and development team and that is something that most cultural organisations don’t have. Or rather, they prioritise elsewhere and frequently decide to live with the search interface for their collections that comes with the off-the-shelf collection management system they have purchased.” He goes on to say that software products could provide configurable and skinnable generous interfaces, lowering the upfront investment and making generous experiences more obtainable for organizations of all sizes.

Haskiya outlines several principles for generous interfaces: Show first, don’t ask (provide rich overviews without search); provide samples and clues (using collection content); show relationships (between collection features); and provide rich primary content (deliver on the promise).

This Gist visualization of the Cleveland Museum of Art collection includes large, engaging thumbnails and filters to narrow down the view by time range, culture, department or type

This view groups objects into categories — decades, or categories like culture, department and type

Generous interfaces often include data visualizations that help expose the structure of digital collections. For example, a stacked bar chart might show the total number of artworks per decade, segmented further by genre, and clicking on any of the facets would then reveal the artworks in question for the selected slice. A pie chart could present a simple gender breakdown, or the cumulative distribution of media types across a collection. A map might show the country of origin for artists in the collection.

An interactive pie chart shows the departments at the Cleveland Museum of Art — click on a wedge to see all of the objects in that department

A stacked bar chart for a Gist visualization of the Cooper Hewitt collection shows the distribution of subcategories filtered by “poster” over time — the bar chart segments are clickable to reveal the objects in that particular selection

A map view for a Gist visualization of the Museum of Modern Art collection, showing the country of origin for artists represented in the collection.

The concept of generous interfaces is at the heart of Gist—one of the first software platforms providing generous interfaces for digital collections. A product by Schema, a research and design firm, Gist turns structured data into explorable visualizations for creating deeper engagement with audiences. Whether you have a museum collection, a library special collection, or an internal catalog of products, people, customers, or locations, Gist provides context for digital collections so that people can derive meaning and insight from them.

In addition to the off-the-shelf visualization formats Gist provides, it also allows for full customization, making it possible to capture the unique aspects of any digital collection.

Custom Gist visualization for Knoll’s company timeline, showing connections between people, products, and events

Clicking on an object shows connections to other objects

The “Connections” view shows relationships between all objects in the archive

Gist is already in use by a number of organizations. We invite you to take a look and give us your feedback. And if you think you have a usecase for Gist, reach out to us to schedule a demo—we’d be happy to hear from you.

Christian Marc Schmidt is Founder and Principal of Schema, a research and design firm based in Seattle.