April 05, 2004

prada epicenter revisited

We're always at least one step behind the times here at fredshouse, which is one reason why we're so endearing. But a February article in Business 2.0 (grumble, subscription req'd) chronicling "Prada's High-Tech Misstep" doesn't seem to have been widely discussed, and since I just saw it (via Thackara's newsletter) it's news to me.

You'll no doubt remember the Rem Koolhaas designed Prada Epicenter, a $40M Manhattan store that opened in Dec 2001 to much fanfare. With wireless custom PDAs, RFID tags in all the product linked to real time inventory and CRM databases, "smart" dressing rooms with Privalite liquid crystal glass walls, and clothes closets with touchscreens showing size/color/matching options along with videos of the selected items being catwalked, it was a ubicomp retail fantasy brought to life.

Fast forward two years later, and all is apparently not well at the epicenter:

Most of the flashy technology today sits idle, abandoned by employees who never quite embraced computing chic and are now too overwhelmed by large crowds to coolly assist shoppers with handhelds. On top of that, many gadgets, such as automated dressing-room doors and touchscreens, are either malfunctioning or ignored. The multimillion-dollar technology spend is starting to look more like tech for tech's sake than an enhancement of the shopping experience, and already the store's failings have prompted Prada to reevaluate its epicenter strategy.

Regarding the custom handhelds that the employees were to carry with them:

During multiple visits this winter, only once was a PDA spied in public -- lying unused on a shelf -- and on weekends, one employee noted, "we put them away, so the tourists don't play with them."

As for the RFID-reading clothes closets:

On several recent occasions, the RFID "closet" failed to recognize the Texas Instruments-made tags, and the touchscreen was either blank or broadcasting random video loops. During another visit, the system recognized the clothes -- and promptly crashed.

I'm not surprised that Prada's experience has been less than delightful. We did some large-scale demo systems and real-world user experiments in cooltown, and we discovered some serious challenges in deploying even relatively simple ubiquitous computing technologies.

One, what seems straightforward to build and run in the lab, is an order of magnitude harder to make work in the world. Using wireless LAN? Count on interference. Using infrared? Count on sunlight and heat sources. Using RFID? Count on damaged tags and misreads. Using PDAs? Count on dead batteries, lost styli, frequent crashes, both the soft and hard (floor) kind. Oh, and everything will be obsolete or broken in a year or two, so count on plenty of ongoing support to keep things fresh and fun.

Two, the real problem is those pesky humans that use the system. You try to make serious technology for serious business, and people just want to play with it, see what it does and what they can make it do that it wasn't designed to do. The novelty or "wow factor" of nomadic and ubicomp technologies is a significant issue for evaluating and deploying these kinds of systems. We saw this in our work with the Exploratorium, for example.

Three, the real problem is those pesky humans that use the system. The potential uses and benefits of ubicomp often seem "obvious"; most of us in the field have spun variations of the same futuristic scenarios, to the point where it seems like a familiar and tired genre of joke. "You walk into the [conference room, living room, museum gallery, hospital ward], the contextual intention system recognizes you by your [beacon, tag, badge, face, gait], and the [lights, music, temperature, privacy settings, security permissions] adjust smoothly to your preferences. Your new location is announced to the [room, building, global buddy list service, homeland security department], and your [videoconference, favorite tv show, appointment calendar, breakfast order] is automatically started." And so on. Of course, what real people need/want in any given situation is FAR from obvious. Even the simplest stuff is non-obvious: If the task you are supporting in your ubi-design requires people to use their hands, don't be surprised if they don't know what to do with that handheld computer you gave them! And if people are supposed to step on a floor pedal to turn the dressing room walls opaque (huh? I suppose that must have made sense to some designer?), don't be surprised when the rest of the store gets an unintended eyeful of prime Prada-buying primadonna.

Ubicomp is hard, understanding people, context, and the world is hard, getting computers to handle everyday situations is hard, and expectations are set way too high. I used to say ubicomp was a ten-year problem; now I'm starting to think that it's really a hundred-year problem. Prada took an audacious, fun step forward, but the journey has only just begun.

Posted by Gene at April 5, 2004 10:22 AM | TrackBack
Post a comment

Remember personal info?