Wednesday, June 25, 2008

User Experience Design Patterns from the Yahoo! Library

When you design a new interactive, talkback station, or traditional exhibit for your museum, what best practices and design requirements do you consult? Do you use in-house documents to guide your actions? Do you use classic books? Museums don't have a well-developed body of industry-wide best practices. The "every exhibit is different" argument leads us into idiosyncratic practice, which is ok in institutions that retain staff for decades, but not so useful in the contemporary world of shifting workforces. We each have our own rules of thumb and dos and don'ts. But we don't have a standard vocabulary for addressing common problems, and as we solve them, we rely on institutional memory to retain the lessons learned rather than finding a way to universalize and document them.

This week I've been looking at a more deliberate way to document best practices from the world of user experience design. Yahoo! is a very large company with a wide variety of user-facing products created by staff who in many cases have zero interaction with other staff also creating user-facing products. This creates two design problems. The first is consistency. To a user, interacting with Yahoo! maps and Yahoo! fantasy sports should feel similar. If the fantasy sports staff never interact with the maps staff, how will they align what common user actions (e.g. ratings) mean in the world of Yahoo! products? The second problem is redundancy. If the fantasy sports staff have come up with a great way to rate other members of the community, why should the restaurant review staff design their own rating system?

The solution is a design pattern library, a place where Yahoo! staff publish generalized solutions to common user experience problems. These can be as broad as "
how do you communicate change on a webpage?" or as specific as "how do you rate an object?" The patterns are arranged hierarchically, with some patterns including many sub-patterns for specific manifestations of the problem. At every level, the pattern includes a problem, an image-based example, a solution, recommendations for use, and rationale. The patterns serve both as useful how-tos and thoughtful why-shoulds. They are used internally across Yahoo! by a community of designers who rate, comment, and adapt them for use.

And now Yahoo! is making some of them public. You can access a limited design pattern library
here. Of particular interest to museum folks are the "social" patterns, which include best practices for feedback/review architecture and reputation indices. The review architecture is useful when considering how to design talkback stations, the reputation patterns valuable in the application of game mechanics to museums.

In 2006, I
wrote about how game mechanics can improve the stickiness of a museum experience. Now, these Yahoo! reputation patterns explore the impact a variety of game devices (points, leaderboards, competitiveness) have on social communities. How should you identify different constituencies in the community? How can you reward participation, and what kind of participation should you acknowledge? In simple language, clear examples, and helpful bullet points, the Yahoo! design patterns help us tackle some of these questions.

They're also enormously useful from an industry development standpoint. You can read more about Yahoo!'s process for creating their internal pattern library in this paper. That library, unlike the patterns published publicly, is very much a living, shifting set of best practices. It gives Yahoo! staff a common vocabulary and supports a culture of institutional sharing and reflection. And while exploring best practices for collectible achievements is fascinating, there's a whole world of other design practices I'd like to see chronicled in this fashion. ASTC gave us ExhibitFiles, a place where museum exhibit designers share case studies and reviews of individual exhibition projects. But maybe we need to create design pattern library for exhibits as well, a place where people can share solutions and recommendations for problems across exhibitions ranging from wayfinding to personalization.

How do you share and learn about best practices for general museum-related problems and solutions? What formats and content would be most useful for you?

2 comments, add yours!:

dave said...

As always a really insightful post. I completely agree with you, as museum exhibit developers we are all guilty of not publishing or sharing our thoughts enough, as a consequence a lot of solutions get re-invented - even the bad ones. I have been having some discussions with my team at the Science Museum in London as well as some individuals in one or two other museums with a view to setting up a wiki to do exactly this. I would really value any further thoughts you have on this.

Marti Louw said...

There may be funding to support this kind of "Design Patterns" initiative from the NSF. It would be nice to see it come from within design/maker community.

A good reference based on Christopher Alexander's idea of design patterns for the emerging field of HRI is an interesting starting place. See "Design patterns for sociality in human-robot interaction"

Also to fund this kind of initiative see the new NSF solicitation with language that seems to directly encourage the development of "tools and methods" to support designers of informal learning experiences ...

"Design of Informal STEM Education Tools and Methods: The Creative IT program encourages PIs to consider interdisciplinary R&D focused on the design and evaluation of new tools and methods that enhance the creativity of design professionals in the informal STEM education field, such as exhibit designers. The objective is to produce design tools and methods whose application can potentially improve the impact on the public's engagement with and understanding of STEM. [excerpt]"

Solicitation at:

The cognizant program managers are Arlene de Strulle ( and Al DeSena (

If you have a good proposal idea, do contact the program managers to float the idea and get feedback. ISE program officers are much more open to direct communication with constituents than other divisions of NSF.

Marti Louw
Researcher, Designer

Innovative educational approaches: Creativity can be a focus for learning environments, such as studio learning, problem-based learning, and informal learning environments, that reward creative thinking. The development and evaluation of learning that uses innovative computational environments to encourage creativity can lead to new ways of teaching knowledge and skills-based subjects.