An interview with…Mark Schindler

Mark Schindler, Chief Technology & Innovation Officer of Visual i|o, offers his thoughts and experiences in today interview about infovis.

1 - When, how and why have you decided to start up a company focusing on infovis ?

Information visualization has always fascinated me. As a child I was interested in all kinds of maps and diagrams and could pore over them for hours. I’d get out colors and create my own maps. About 15 years ago, when I was studying for my master’s degree in architecture I realized that representation of information was more interesting to me than the design of buildings. My M.Arch. thesis project examined ways that information—via sound, light and electronic media– can be encoded in our built environment.

In the mid-1990s I was a partner in a company that designed and built custom databases to aggregate operational data for large companies. Although a lot of effort has gone into building better tools for aggregating, storing and architecting data over the last 15 years, we saw that there is still a big gap in the presentation of data in ways that make it easier for people to solve problems and make decisions. So we relaunched our company as Visual i/o in 2001 with a focus squarely on data visualization for decision-making.

2 - What are your education backgrounds and previous experiences before creating VISUAL-IO?

Before studying architecture I had gotten a BA in Chemistry. My co-founder, Angela Shen-Hsieh, is also trained as an architect. People think architecture is about buildings, but in a way it’s really about graphical representation. Architects don’t build buildings—they make drawings of buildings. Architects slice through buildings using graphical techniques like sections or plans. We think of what we do as applying similar techniques to datasets. So the kinds of questions that intrigue us are–how do you draw a section through a dataset, or a situation that plays out over time?

People in our company have backgrounds in diverse fields—music, finance, law, the visual arts. Ours is a team process, and it can take a lot of different perspectives to get information to communicate expressively.

3 - Which is the best results your customers got using your infovis services and/or products ?

When a customer told us—we had all the data before but it was difficult to see how it all fit together– we realized that much of what we can help with is simplification. It still surprises us, but the highest value often seems to come from helping our customers gather disparate information together in a single meaningful picture of a complex situation.

Decision-making is ultimately a human process. Experienced and expert managers rely on their gut, and often describe decision-making in terms of pattern-matching. We feel you have to respect and understand how people—both expert and not—make decisions, in order to supply information in a way that’s relevant to them. How do you draw a meaningful qualitative picture from a mass of quantitative data? That’s a great challenge for information visualization.

4 - Which is, to you, the most interesting project you have worked on and why ?

We look for commonalities among different problem-solving challenges involving information. So for us the most interesting questions span across projects—how might different people, in seemingly disparate situations, see patterns that are meaningful to them and help them find the answer they are looking for? How might people see business and situation perspectives in common ways, and take advantage of a common symbolic vocabulary?

We think the solutions lie in a greater expressiveness of graphical representation and greater fluidity of interaction, not just in finding the right kind of chart type.

5 - On what themes are you working now ?

One theme is a problem we all experience–how do you get from 200,000 Google results to the 10 that might be interesting to you? A lot of people are working on better algorithms. The solution, though, may be less about the algorithm and more about how people cognitively process all the information available.

6 - How do you think infovis solutions market will evolve in your country for next years ?

We in the US are, I think, a bit behind much of the rest of the world in terms of our acceptance of information in non-verbal form. Our street signage, for example, is verbose and less graphical than in Europe.

At the moment, info viz solutions tend to be defined in the market by charting and dashboarding componentry. But I think people are already seeing the limitations of simplistic gauges and stoplights and wondering what’s next.

7 - If you should suggest an education programme to a student interested in InfoVis, which University Course do you suggest him/her ?

Traditional curricula in data visualization have focused on techniques of transforming data into graphical form. But effective information visualization is not just about encoding information, but encoding it in ways that can be readily decoded by humans.

· Cognition and decision theory, for how people process information and put it to use.