Manifest Digital Human Experience Society

This is the second story in a series about the development of the Bally Results Center, the groundbreaking interactive experience design by Manifest. For the first part of the series, click here.

When Danny Davis occasionally felt discouraged while working on the Bally Results Center, he sometimes thought of Thomas Edison.

“I came across a story about how he failed more than 1,000 times to make the lightbulb,” Davis said. “But when asked about it, he said, ‘I have not failed 1,000 times. I have successfully discovered 1,000 ways not to make a lightbulb.’ That gave me peace of mind.”

The comparison is appropriate. As with much of his work, Edison’s true genius in devising the common lightbulb was not to uncover a new basic principle of electricity or even to invent an original device – people already knew they could make wires glow in the dark.

Instead, Edison’s goal was to find a lightbulb design that could be produced cheaply enough and last long enough to be implemented on a large scale. Like Davis, his invention was informed not just by a great vision, but by contemporary technologies, economic conditions, and the needs of business partners.

Key to Edison’s development were the years he spent as a telegraph operator when he was a young man, observing the progress of that world-changing machine.

Davis, on the other hand, played XBox and watched YouTube – which turned out to be the perfect preparation for the assignment.

“This project made us realize that people had experience from outside the office that we didn’t know about,” said Ryan Noel, who drove strategy on the Results Center. “I knew Danny was a gamer. I knew he was interested in lots of different types of technology. But when he showed me a prototype, I thought, ‘Wow, this guys is four steps ahead of everyone else.’ We could dream as big as we want and he’ll be able to make it happen.”

Besides enjoying gaming in his spare time, Davis had for some time collected data on interesting technologies on a website called the Interactive Design Archive.

“There are lots of awards sites out there, but those are elite clubs where only the best of the best get in,” Davis said. “The experiments are never shown. I wanted to created a space where the kid in his basement messing around could show his work.”

As a result, Davis was well aware of the ways people had used open source software to achieve a variety of results with the Kinect, the motion sensor technology developed by Microsoft for its XBox gaming system. The examples he cites as inspiring range from corporate to kindergarten: a display wall for Toyota’s corporate offices, an interactive advertisement at a Connecticut airport, a family who had figured out how to make the Kinect follow their hand motions to create a virtual puppet show in their living room.

“Seeing the stuff that was out there, I was jealous,” Davis said. “Any time I would see a new one, I wondered, ‘How can I re-create that? What is the next iteration of that thing?’”

The next thing turned out to be the Bally Results Center. The initial idea was dreamed up in a collaborative session between the leadership of Manifest and Bally dedicated to re-imagining the gym experience. The company’s Chief Information Officer, Guy Thier, brought up the idea of an interactive wall inside the gym.

Davis was impressed when he heard about the CIO’s ambition.

“One of his goals was to win a Webby,” Davis said. “When a client has a goal like that, you know they are not going to hinder your creativity.”

Noel and Davis discussed how to develop the interactive wall. Because of some tinkering with the Kinect at home as well as the videos he had seen online, Davis was confident it could be done. Still, very few have actually developed production-ready, public-space applications for the Kinect. Despite his enthusiasm for the project, Davis said it was hard to view his experiments at home as adequate preparation for the job.

“That knowledge basically kept me from panicking,” he said. “But I knew other people could do it. I knew I could figure it out.” 

That didn’t mean it would be easy. Davis said he and his colleagues needed to spend about a week simply determining what a normal workflow on the project would look like. Early on, they decided that it would be helpful to bring in Helios, an agency with particular expertise in motion sensor technology, to work on perfecting the functioning of the Kinect incorporated in the design. At Manifest, Brandy Taylor designed a look that would tie together the Results Center applications with the revamped Bally gym experience. Then Davis and his colleague Andrew DeMay would use Flash, a common programming language, to translate Taylor’s visuals onto the seven-foot-tall screen.

But first, they had to be able to see what they were doing.

“We’re working on something that is 1536 pixels by 2720 pixels,” Davis said. “The highest resolution TVs are only 1080 pixels tall – this is double the height. I kept saying, ‘ I need a bigger monitor.’ Then I got a bigger monitor and it wasn’t big enough – and right now I have a 27-inch Apple display. There were times when I literally flipped my monitor on its side and rotated the resolution, but it still fell a little bit short.”

Davis ended up devising a workaround in Flash. The files he worked with, he said, were the actual size of the Bally wall. But he added a line of code that allowed him to publish the code at a lower resolution. He also “encased” the files in a movie clip that made them easier to see. In the end, the added flexibility enabled Davis to work on a screen of any size.

“You could literally shrink it down to the size of your phone and play with the Results Center at a size of 100 pixels tall,” Davis said. “That was a big win.”

When it came to actually building the Results Center apps, a major problem Davis faced was the fact that Flash is built to respond to a mouse, while the end product would be controlled by users’ movements. Because of the examples he’d researched, Davis knew it was possible, but he was concerned that translating human motions into Flash over so many separate applications would be unwieldy.

“To have thousands of lines of code and to have to constantly switch back and forth between the two would have been a nightmare,” he said.

To avoid the problem, Davis added a cursor to the program whose ultimate purpose was to match up with the user’s hand via the Kinect. When the Kinect was not present, the cursor defaulted to the developer’s mouse. A major difficulty was solved with a single variable in an XML file.

“I could just say Kinect equals false,” Davis said. “That saved us countless hours of development.”

Davis said he could not be fully confident the project would succeed until he saw that the work he was doing on his screen meet up with the motion sensor side being developed by Helios. For weeks, he nervously anticipated the day of the integration. When it came, though, he was pleasantly surprised to find that the solution required just a few new lines of code.

Though his plan had succeeded, it wasn’t the end of the road. He and the team still had to figure out how to solve for the many bugs users experience in the first generation of a new technology.

“The mouse has been around for about three decades,” Davis said. “The Kinect is something that Microsoft has only made into a household name over the past couple of years. Now we’re waving our hands around like Jedi Knights and expecting to control an eight-foot-tall television screen.”

As Noel describes it, Davis and the team were encountering entirely new problem sets each time they made adjustments to the device. An important gesture to control the Results Center was the wave, but the team had to account for the fact that everyone waves differently. While testing one wall in the gym environment, the team realized that the vibrations from a room full of treadmills were disrupting the way the device’s camera picked up images and were forced to recalibrate it.

The gap between the amazing potential of the technology and the tedious work of debugging could be frustrating, Davis admits.

“There was a period of insanity where I felt like I had become codependent with the Bally wall,” he said. “If it wasn’t working, I would feel physically sick.”

But the process also gave Davis a greater estimation of what he was capable of.

“I know coming out of this project that I can do just about anything,” he said. “I can say, ‘I made this seven-foot wall. I’m pretty sure I can make your website.’”

There is no doubt that the Results Center is a significant accomplishment. Yet as Edison might have observed, its technological life has just begun. To the “Wizard of Menlo Park,” the market was the ultimate measure of a technology’s usefulness. And the Bally gym members currently trying out the Results Center in gyms across the nation may be a small percentage of the users whose lives could ultimately be improved by this kind of technology.

Check back tomorrow to learn about potential future applications of the technology in the Results Center.