University of Pittsburgh | Pitt Home | Find People | Contact Us  


PittMag





Photograph by Lisa Kyle



Steve Roth believes he has found a way to capture and share thought through an all-encompassing software program. It’s a breakthrough that could usurp the Web and computing, as we know it. Already, the U.S. Army has embraced the technological innovation by the Pitt alumnus.

Visual Effects


Cindy Gill


Steve Roth
In Baghdad, a city under siege from deadly street fighting—not to mention desert heat, sandstorms, and scorpions—a Pittsburgh psychologist wants to prevent a tragedy. Steve Roth (FAS ’81, ’76) edges closer to the nearest laptop. He’s a long way from his bright, lively office on Pittsburgh’s South Side, where he is CEO of MAYA Viz, a young software development firm. He arrived in Baghdad only a few days ago, but he has no time to worry about anything but his company’s fledgling software. If it works, lives will be saved.

His brown eyes peer through wire-rim glasses as he watches the action on the nearest computer screen. He is fit, but his close-cropped beard and civilian clothes set him apart from others in this central-command complex in the U.S. Army’s Camp Victory. The others are 1st Cavalry soldiers, responsible for security in this simmering city. Everyone’s attention is riveted to the glowing computer screens scattered on tables around the room.

Each set of screens shows the same full-color map, with blue squares moving along a black line of roadway. The blue squares show the actual movements of transmitter-tagged army vehicles as they travel through Baghdad. Those watching the screens just saw a few trucks in a larger convoy make a wrong turn. The software shows they’re heading toward a neighborhood with intense gunfire. Suddenly, the army’s transmitter signal stops. The blue squares no longer accurately show the trucks’ location. Radio contact can’t locate them. The screens glow as the soldiers in the room watch intently and listen on their networked headsets for reports from the field. Where are they?

Roth never expected to be in the middle of a Middle East war. More than 20 years ago, the native New Yorker was a Pitt graduate student, studying cognitive psychology. He was drawn to Pitt’s Learning Research and Development Center (LRDC), a place known for its research on teaching, learning, and computer applications in education. Such a blend, he reasoned, would give him unique insight into how people think and learn.

While completing his PhD in psychology at Pitt and conducting research at LRDC, Roth worked with academics from many disciplines to develop software that improved reading skills in school children. That was a powerful lesson for him about the benefits of combining psychology, computer technology, and the expertise of others.

Near the end of his doctoral work, Roth was overwhelmed with informa-tion, but that led to an idea. It came to him while looking at the stacks of paper on his desk overflowing with raw materials for his thesis: facts, statistics, suppositions, and a handful of hypotheses about how people learn. The jumbled stacks sat, wobbly, on his desktop, and many of the sheets had penciled scribblings with arrows to highlight text or numbers. What I really need, he thought, is some way to save my thoughts and organize the data.

This idea came at a time when most computers were about the size of UPS trucks, yet their memory capacities barely rivaled today’s pocket calculators. “Nothing was easy then,” recalls Alan Lesgold, dean of Pitt’s School of Education and a dissertation advisor to Roth, “but while some people remained fixated as prisoners of these limitations, Steve was imaginative and ready to use and stretch available tools.”

The more Roth worked on his thesis—all the while generating data, churning through theories, and asking for input—the more he longed for a way to simplify sharing information electronically. That longing became as important as his thesis. “I was just someone who wanted to make it easier to interact with data,” he says.

When he earned his PhD in 1981, Roth crossed Forbes Avenue to work at neighboring Carnegie Mellon University, where he sought insights from experts in artificial intelligence—a discipline that attempts to build machines mimicking human thought. He joined the robotics faculty and founded a research group called SAGE or System for Automated Graphics and Explanation. SAGE’s focus was to create software programs that could make it easier for people to understand complex scenarios by generating visual representations of data.

“In my training as a psychologist at Pitt, I got a tremendous appreciation for data,” says Roth. Much of his appreciation came from the University’s multidisciplinary approach, which enabled him to do his graduate work in LRDC as well as the Department of Psychology. That freedom enabled him to exchange data with researchers in a variety of fields, including sociology, anthropology, computer science, and education. These exchanges and the ideas they generated helped shape his research and, just as importantly, his career goals. “I believed a deep understanding could come of the world if you collected the right data and you knew how to organize it, understand it, and share it.” He decided to take his concepts and apply them more broadly. “For me, visualization was a way of furthering that, a way of understanding data to gain insight,” he says.

Roth became part of an expanding field focused on the development of visualization software. It’s simply easier for humans to grasp information in the form of pictures and picture-symbols. A National Science Foundation report includes a chapter called Visualization, A Way To See the Unseen. This approach, it says, emerged to cope with the massive amounts of scientific data that had been pouring out of computers since the 1960s. “We became very good at flipping through stacks of computer printouts,” according to the chapter, “but we realized that, at some point, people needed to see their solutions in order to make sense of them.”

Now, it’s routine for researchers to use visualization software programs. At its simplest, for instance, such programs might use the size of a circle on a map to convey population densities, or different colors might be used to separate groups of data with different characteristics. “As much as possible,” says Roth, “you want to see something that visually is analogous to the concept you’re trying to express.” His work in the 1980s focused on these and more advanced kinds of data-visualization concepts, always trying “to make it easier for people to interact with data.”

In 1990, Roth met Peter Lucas, another Pittsburgh psychologist. The two were perfect counterparts, and they began to collaborate on research. Lucas was full of provocative late-night ideas. He envisioned, for instance, a universal database of information—a place where anyone could browse and select nuggets of information from a vast pool of all known facts. Roth was adept at taking abstract notions and transforming them into realistic outcomes. “[Roth] thinks like a scientist,” says Pitt’s Lesgold, “but he can also take ideas from the world of science and put them to practical use in exciting ways.”

In 1991, Lucas joined forces with several restless academic colleagues to create a company called MAYA Design. The company’s motto is “taming complexity,” and the firm’s name gives a nod to Raymond Loewy, a 20th-century trendsetter often called the father of industrial design. The Studebaker Avanti and the yellow Shell logo are just two of Lowey’s many design creations. Loewy’s credo was to design products for consumers that were MAYA, meaning Most Advanced Yet Acceptable.

Roth didn’t make the jump to MAYA, but he continued his visualization software work and his conversations with Lucas. Then, in 1998, he decided to move beyond the talk. He wanted to put their ideas into action. He was ready to build a software program that would help people visualize complex information, organize it, analyze it, play with it. So Roth cofounded MAYA Viz, a sister firm to MAYA Design.

CoMotion is the software package that evolved from MAYA Viz. It creates an electronic workspace that lets users gather and analyze enormous amounts of information from their desktops or laptops. But the information isn’t confined to a document or file. Instead, the program compresses chunks of information into compact visual symbols. Users handle pieces of text, symbols on a map, charts and graphs, picture icons—all kinds of information in all kinds of forms—which they gather onto a personal pasteboard. The whole composable process is more like an artist applying colors and concepts to canvas than it is a typical computing environment.

“You can put pieces together to build what you want,” says Roth. For example, warehouse icons can be plucked from a list and dropped into a frame that immediately gives bar charts, comparing supply inventories from each warehouse. Likewise, the bar charts can be dropped onto a map that immediately shows the locations of those warehouses. Icons can be selectively dragged and dropped to give deeper and deeper levels and types of information, allowing users to compare, contrast, and analyze scenarios at a user’s whim.

Users can even draw broad lines of “virtual ink” on their workspaces, as if writing on a blackboard. Whatever any user composes on-screen can also appear on all of the other networked computer screens, resulting in a creative flow of information, which others can contribute to in real-time. Collaborating users share more than just images. They share the actual data that is displayed.

The software was developed over several years and ready for initial use in 2002. It has handled a massive transportation operation and is being used to explore the best treatment strategies for hundreds of patients with different doctors in different locations. It has helped financial analysts examine investment outcomes and strategies, and energy companies find the most efficient distribution routes. Its uses are as vast as information itself.

From Roth’s perspective, the software is a way to think aloud visually and to observe others’ thinking—a digitized version of doodling on a napkin to share a concept in the midst of conversation, but on a far grander scale. “We’ve captured thought and exposed it visually,” says Roth. “It must be like what people experienced when the written alphabet was invented. Before that time, you communicated by talking, by word-of-mouth. Then, suddenly, you could put your thoughts on parchment, and they could be replicated and travel someplace else. We believe what we’re doing is as monumental as that.”

In Baghdad, the 1st Cavalry Division has been using MAYA Viz’s CoMotion software since April 2004, when the division was deployed to Iraq. MAYA Viz’s hands-on staff arrived last April as well to monitor the software, which had been customized for the army’s use and named Command Post of the Future (CPOF). They hovered over the prototype CPOF, wrangling with it around the clock, in the heat, in sandstorms, and in spite of some nearby mortar attacks and gunfire skirmishes.

Roth traveled to Iraq last July. On the day those trucks made the wrong turn, he could only watch CPOF’s computer-screen map and hope the software system could locate them. Soon, a report came from the field about a large cloud of dust near a neighborhood in turmoil. A CPOF user quickly drew an arrow on screen, pointing to the location of the dust cloud, and then drew a circle around the area. Soldiers in an aviation battalion equipped with CPOF saw the drawn circle and an order was given to dispatch Apache helicopters. The helicopters spotted the location of the wayward trucks, and army troops were quickly sent there, which averted a pos-sible tragedy.

CPOF’s help in finding the lost trucks did not go unnoticed; it was just one of many situations where MAYA’s software proved its mettle. “There’s no doubt that we’ve saved soldiers’ lives with CPOF,” says the division commander, Major General Peter Chiarelli. “And this is just the start. In my 32 years in the army, I’ve never seen a system that will have a greater impact on our army and our entire joint force.”

CPOF continues to be used 24/7 in operations among eight 1st Cavalry brigades, encompassing 30,000 troops. Global Infotek, ISX, Oculus, and Polexis help contribute to MAYA Viz software’s nonstop operation.

Brigade commanders and other division soldiers are connected through CPOF. All of the soldiers view the same electronic workspace, with the same map, from one of the 55 screens on lap-tops and desktops at the command center or in the field. They drop and drag information for analysis and discussion, rehearse maneuvers virtually, and test “what if” scenarios online to plan ideal responses.

CPOF has become the division’s primary method of communication and command. Training time has been drastically reduced, too, since the 1st Cavalry’s replacements, the 3rd Infantry Division, were able to follow the action in Baghdad from their stateside computers in Fort Stewart, Ga. The U.S. Army is so impressed by CPOF that other divisions will soon have it, and other branches of the military are interested as well.

Now, Roth and his MAYA team are working to apply CoMotion software for everyday use by everyone. Using MAYA’s innovation, it would be possible to pull individual chunks of information—names and data about cities, people, and travel locations for instance—from any digital document on the Web or elsewhere. These information chunks could be plucked from their document-based environments and placed onto CoMotion’s canvas/pasteboard for deeper analysis or comparison based on the user’s needs. It would be like taking information from the Web and making it interactive without the need for software programs like Excel, WORD, or PowerPoint.

It’s a bold vision, but Roth is confident the success demonstrated by CPOF will be matched in all of the software’s future applications. “Instead of 55 users,” he says, “why not two billion?”

Cindy Gill is a senior editor of Pitt Magazine.


Links to external Web sites are offered for informational purposes only and the information there is not guaranteed or endorsed by the University of Pittsburgh or its affiliates.



 Top of Page Home | Pitt Home | Past Issues | Contact Us