Thursday, June 25, 2009

Technology for Experience's Sake: Guest Post by Bruce Wyman

In this guest post, Bruce Wyman, Director of Technology at the Denver Art Museum, shares his process for developing interactive technologies to extend familiar experiences in art museums. You may remember the Denver Art Museum from this post about their newest (highly interactive) exhibit space, Side Trip.

It's a running joke at the museum that I'm frequently the person in the room that doesn't advocate for the very things that I'm responsible for (technology). It's not that I don't firmly believe in the power and potential of what technology has to offer, it's just that so often it's a red herring when we're designing experiences for our visitors. I want the technology to disappear. I want the visitor to have an amazing experience in general at the museum, and not leave thinking some piece of technology was the thing that stood out. More often than not, if the technology is memorable, it's usually in a negative way -- something didn't work as expected.

So, let's make it easy on ourselves and start off by largely ignoring the technologies that we could be using. Frankly, visitors frequently don't care about the technology and I agree with them. Give them something rewarding, some meaty bit of fun and engagement and concentrate on designing what that experience could and should be. Once you get a good sense of that, the technology begins to fall into place and you stumble across new kinds of experiences that have the power to delight the visitor and probably more efficiently serve your original goals.

Of the different sorts of things that we've design at the Denver Art Museum during the last five years, it's unusual that I have a particular technology in mind at the outset (I'm sure someone could easily call me out on that, but let's pretend together, shall we?). Our standard practice is to deliberately ignore the possible implementation and tease out the details of what will make the experience compelling for the visitor. If we're going to show a video, how is that different? What will make the video compelling? Is there a particular *kind* of interaction that's important to satisfy the experience? etc.

At some point, as you begin to think about how visitors might interact with whatever your experience could be, you start to draw on real world analogies and natural patterns of behavior and interaction begin to emerge. The real world doesn't always have the *best* interactions, but it is filled with interactions that people already know and understand. The critical behavior in making the shift to designing in concert with these interactions is to get in the habit of just watching people all around you and how they engage with the rest of the world.

Let's consider a quick possible scenario. I work at an art museum and one of my long term desires is to know what works a visitor finds interesting without having to deliberately make them punch a button or use some piece of technology that interferes with a visitor's rapt
attention with our artworks. In this theoretical situation, my simple interaction goal is to concentrate on the *capture* of information rather than delivering something back to the visitor. I'm simply trying to find a clean way to judge visitor interest.

When I'm a visitor, if I'm interested in sharing something with a friend, I might point at it. If I'm walking around in an art gallery, I'll pause in front of work that interests me. If I'm really interested, I'll lean in or get closer to the ubiquitous tiny label nearby to read a spot of info. Those are all things people already know how to do and have been doing for thousands of years (in the latter example, the thousands of years that art museums have existed, certainly). So, without even having considered the eventual technology application, I've determined that I'm interested in tracking a visitor's location, proximity to known objects, and time spent in a location. I can then begin to consider ways to do that -- vision tracking, possibly sensing a handheld the visitor may be using to gather information, or RFID tracking a membership card on their person. Given privacy concerns, I'm not sure that I actually want to do any of those, but the important part of the process is that I'm defining an eventual solution by the visitors and their interactions, not from the starting point of specific technologies we want to implement.

The Select-A-Chat in our Western Galleries had an interesting evolution. The final implementation is essentially a small theater space in our galleries. As you enter the area, there's a wall projection resting on our jauntily angled interior walls. Next to you is a comfortable sofa and a coffee table in front of it. The top of the table has a graphic depicting a number of artworks from the nearby galleries with different interview questions superimposed. To select a particular topic, there's a coaster-sized metal 'X' that the visitor places on the table. Simply, the whole interaction is described in one sentence: "X marks the spot." If you want a more detailed tour of the Select-a-Chat, check out this video.

The goal of the experience was to interview selected artists and give visitors some insight to their efforts and process. We wanted artists that felt very human and dispelled the idea that the creative process was a magical one but rather took real effort with some days being great, others not so much. The real strength of the overall interactive is the videos themselves, so we recognized early on that we wanted a relaxed environment where the process of choosing the videos largely disappeared for the visitor.

With these ideas in mind, it became substantially easier to brainstorm different approaches to how we might achieve a simplified end result. In an early iteration, we imagined something very direct -- giving voice to the artist -- in which the table would be more akin to an old telephone operator switch board. The different artists would have headphone jacks for mouths and you'd plug directly into the display. Moving on from that, but still with the same idea, we imagined a large set of fabricated lips as being the object moved around on the table. We quickly moved on when we started to imagine the mockups of photoshopped artist faces without mouths and were left with a decidedly unsettling image. The metal 'X' was a response in trying to step back to what the actual mechanism was, but by that point, we had a good idea that the general interaction was right.

Our visitor testing satisfaction surveys support our belief that the experience works well. People find the experience easy to use and at the same time, we've added a little bit of magic to the interface. The visitor *can't* do anything wrong -- if the 'X' is on the table, the experience works. It's easy to figure out and if you watch someone else do it, you learn what to do in an instant.

On a slightly different note, while we spent a lot of time thinking about the experience before getting to the technology, having a solid understanding of our goal let us think through a few permutations of how to implement the technology. At one point, we'd abstracted the interaction design so far that we considered using the building's new security system. The same system is in use at airports around the world and is particularly good at detecting potential terrorist threats -- the system is able to detect a piece of unattended luggage left in a terminal after a certain amount of time. When that happens, the system triggers a video feed in a central control room. Our problem was the same; an object left in a location (an 'X' on the table) triggers a video feed (different artist videos). We ultimately developed an alternative solution when it became apparent the cameras wouldn't have the resolution or view that we required.

Our visitors have responded well to this approach in our technology design. Not only with the Select-A-Chat, but our visitor surveys in general indicate that the Denver Art Museum's interactive components are easy to use, effective, and enjoyable. We don't always get it right the first time, and we've had to accept that, but we do believe in an iterative approach based on what we learn over time. Even better, from an internal point of view, we accept technology as part of the visitor experience and not as a competitive element that happens at the exclusion of other parts of the visitor experience.

2 comments, add yours!:

dgilman said...

When you speak about natural patterns of behavior and interaction, I think of one of the scariest words I know in Experience Design - Intuitive.

Intuitive is a scary word because, of course, it means something different to every person.

There's an installation where a bunch of objects are on a table - pick up the object, and the screen next to it tells a story about the object. Problem is, while it was intuitive to the designers that people would pick up the objects, they designed such a beautiful display for them, no one wanted to disturb it.

So the 'X Marks the Spot' is really quite brilliant - anyone who's seen a treasure map or a Road Runner cartoon knows what's going on.

For your challenge of passively noting visitor interest in your artworks, I have to say that the eye-tracking software is quite good, and can give you breakdowns of race and gender, as well as dwell time. If that's scary, maybe just proximity sensors that are finely tuned, or even do IR and use heat signatures and some software processing.

In any case, nice post. Thanks for sharing your thoughts.

'The greatest challenges we face are not problems to be solved, but dilemmas to be managed.'

Maria Mortati said...

Bruce, that's a great post. Thank you for sharing the case study and reasoning with us. I'm about to embark on a similar path with a project, and this write up will help reinforce the values of observation, focusing on the key experience, and iteration.