David Baar on new display technology that addresses the screen real estate problem
Dr. David Baar is the chief technical officer and founder of IDELIX Software Inc. He has been active in both computer science and physics research over the past two decades, but somehow manages to sustain parallel interests in pursuits such as rock climbing, windsurfing and photography. IDELIX Software, www.idelix.com, is a Vancouver-based software company focused on improving the presentation of data and images on computing devices through innovations in computer graphics. Baar and his team have created an innovation in information visualization called pliable display technology (PDT). PDT addresses the challenge of viewing high-density data sets and images on a limited size screen display by providing a new way of magnifying data. It magnifies a subset of the data while retaining connectivity to the rest of the information -- IDELIX calls this detail-in-context viewing.
UBIQUITY: Where did you get the name IDELIX?
BAAR: The original name of the company was Advanced Numerical Methods. The marketing folks quickly convinced me that we needed a shorter name. We looked for something memorable that had a tie into our core technology. The name IDELIX comes from the Greek or Latin root delos, which means clear vision.
UBIQUITY: Let's start by talking about how IDELIX fits into the world. What's its place in the sun?
BAAR: The core members of our technical team are experts in certain niche areas in computer graphics and information visualization. We provide a technology toolkit, the Pliable Display Technology SDK, that addresses certain usability and comprehension issues that most computer users face, associated with presentation of large amounts of information on limited screen space. The toolkit is licensed to major software OEMs so that they can include PDT in their products. That's really where we fit in the world. Truth be told, we started out as a bunch of rock climbers. That's what got this experience started and it's been quite an adventure. Quite surprisingly there is a strong core of computer graphics and computer science people in rock climbing.
UBIQUITY: Which is the cause and which is the effect?
BAAR: I haven't figured that out. There are elements of both risk-taking and problem-solving in rock climbing. I think it is more the problem solving. Anyway, that's how we got started. Through ongoing research and development, primarily over the past three years, we've developed what we call pliable display technology, which provides a solution to a prevalent problem in visualizing and presenting information.
UBIQUITY: Describe the problem.
BAAR: It's a fascinating and compounding problem right now because on the one hand you have higher resolution imagery, digital cameras with increasing resolution, access to more information because of the Web, and better applications for generating information. Conversely you have users trying to view and edit this information on increasingly small screens, such as the popular handheld devices. Even if you look at a screen on a laptop or desktop machine, generally there is more information available than you have pixels to show it on. That is the heart of what we term the "screen real estate problem."
UBIQUITY: Take a particular example and work it through.
BAAR: Say, for instance, you are using Microsoft MapPoint or MapQuest online to get directions to navigate to a destination. Usually, just getting to the location of interest involves a lot of zoom-in steps to show the detail of your intended location. The whole time that you are doing those zoom-in steps, you are pushing information off the screen. In fact, it is not necessary to do that if an application incorporates pliable display technology. Imagine one of your lenses at your destination and another lens on your current location. You can see the next intersection and know which turn to take and you can also see the detail around your destination. In that way we can avoid a lot of zoom-in and zoom-out steps and finally use the map for its intended purpose, which is to show you how to get to your destination without you getting lost in the process.
UBIQUITY: Two words that we use a lot in publication are ubiquity and innovation. Talk about what you have developed in terms of those two words.
BAAR: In terms of ubiquity, we certainly feel that this problem is something that every user of a computing device with a screen experiences at one point or another. If you are working in image composition or image editing, you generally want to view things in more detail than you can see at the base level of magnification. This technology is applicable to people such Photoshop users, people looking at maps, people looking at all kinds of data on all kinds of computing devices. Most people are now working with handhelds or with hybrid handheld phones with extremely limited screen sizes. We feel that we have an improvement to the usability of various kinds of devices for many different applications. We really see this as solving a compounding problem that most computer users experience. So that's the ubiquity side.
UBIQUITY: And on the innovation side?
BAAR: On the innovation side, we have 11 patents pending on a whole framework for a powerful geometry engine for doing detail-and-context views of information. Our technology can be integrated into software applications. It has powerful detail-and-context technology with several proprietary pieces to it. In addition to in-context viewing technology we include what we call undisplace. When working in a pliable display technology lens, undisplace provides reverse mapping back to the original data coordinates. It applies to any of the lens shapes that we can create with the technology. It allows users not only to view the data but also to edit the data in the lens, to do things like markup and so on. All of those are done by the host application, and our undisplace technology enables that to happen.
UBIQUITY: How does IDELIX fit within a field of competitors?
BAAR: Most of our competitors are the older zoom-and-pan inset technologies that are prevalent right now. It's not direct competition. It's more like substitute competition. While there has been a whole body research on this technology over the past decade starting with early work at places like Xerox PARC, there are no other companies right now selling a detail-and-context software development kit (SDK). So first, we have to get people past the zoom-and-inset tools they are using now.
UBIQUITY: Is your location in Canada an asset?
BAAR: Compared with the US, labor costs are slightly lower. Another nice thing for us -- being a bunch of rock climbers -- is our Vancouver location. So far it has not been difficult for us to do business in the US. A lot of our business is done in the Washington, D.C. and Virginia areas, and in Silicon Valley.
UBIQUITY: What in particular in the Washington, D.C. area?
BAAR: Our flagship customer right now is Boeing Autometric in the DC area. We also work in defense, a lot of which originates in the Washington, D.C. area.
UBIQUITY: Tell us a little bit about you?
BAAR: I'm a physicist by training. My original research was in material science and superconductivity. My technical career started in '82, combining physics and numerical modeling plus certain things like remote sensing and photogrammetry, done through various technical jobs along the way, plus a long-standing interest in computer graphics. The common theme that led to IDELIX was my doing heavy duty numerical modeling of complex experiments. I needed unique information visualizations and data representations to present my results. In searching for that I started interacting with a number of people in the computer graphics and infovis worlds -- some of whom happened to be rock climbers. That's what really got IDELIX -- then known as Advanced Numerical Methods -- started as an information visualization company.
UBIQUITY: What schools did you attend?
BAAR: I did undergraduate and master's degrees in engineering physics. Next I took a slight turn into more of a pure science vein and did a Ph.D. in physics and superconductivity. All of that was done at Queens University. I followed up with post-doctoral work at three institutes: the Superconductivity Research Lab in Tokyo, which is Japan's flagship effort in superconductivity, the University of British Columbia and Stanford University. It was a lot of fun and I traveled a lot in the early '90s across those three institutes.
UBIQUITY: During the time that you've been at this, have you seen many changes in the industry?
BAAR: Yes. I think one of the primary ones is the fairly obvious advance in hardware, whether it be a PC or handheld device or whatever. It's possible to do more now in computer graphics, especially with 3D technologies, that either wasn't possible a decade ago, or would have required ridiculous amounts of money for the average user. So access to powerful hardware at low cost is a big difference.
UBIQUITY: What hardware does your product typically run on?
BAAR: Our stuff runs on everything from handhelds such as the iPAQ all the way up to high-end desktop workstations. Our SDK is targeted for platforms in both the Windows and Unix worlds (including Linux), and very soon Mac OS as well. We do not depend on specialized high-end hardware for general-purpose applications. We spend a lot of time working on performance optimization, so we can get the technology to work on low-end hardware such as an iPAQ. We can also take advantage of high-end hardware if the application demands it. A short answer would be that we span a lot of hardware.
UBIQUITY: Are most of the people at IDELIX computer scientists?
BAAR: Within our technical team, our people generally have expertise in information visualization and classic computer graphics. We are also very strong in the general area of "hard problem solving." Alongside people with classic computer science backgrounds, we have people with physics, and in one case math, backgrounds who are very good at computing, although not necessarily trained in computer science.
UBIQUITY: Is there a new Holy Grail that you're trying to capture?
BAAR: In terms of a business Holy Grail, one of the key objectives for us is acceptance of the technology. We're providing a technology piece, which in order to be useful, must be integrated into other applications. So part of what we're seeking is widespread acceptance of detail-and-context technology, and specifically our own pliable display technology. We're pushing it further and further into low-power devices, so we're always pushing harder on performance. We're also pushing it into Web-based applications. We're currently doing a project for the Canadian government that will result in detail-on-demand context viewing of information over the Internet. On the business side, the Holy Grail for us is ubiquity. We want to be on every user's handheld, on every application. On the technical side, we're pushing beyond what we do now in two dimensions, and over the next year will be working towards solving related problems in the 3-D world as well.
UBIQUITY: Going to handheld for a minute. What is a good example of an application for a handheld?
BAAR: One area that many people ask us about is queuing of maps on handhelds. Say you use something such as PocketStreets or some of the conventional applications in the geospacial and GIS worlds, such as ArcPad from ES Orion. They're powerful applications but when you try to use them on a limited-size screen you generally must push information off the screen to get at what you want. You lose the intended purpose of a map. We see a lot of applicability in handhelds. This would also be a valuable addition if you were reading documents on a handheld, for example looking at PDFs.
UBIQUITY: Tell me how you've been received by academia. Are you in touch still with academia or have you broken away?
BAAR: We're still very much in touch with academia. Two of our key technical advisors started out from academia, one of whom is still there. We have strong academic ties, in particular with researchers at the Canadian institutes of University of Calgary and Simon Fraser University.
UBIQUITY: Do you make use of interns?
BAAR: Definitely. We make good use of the co-op programs at both Simon Fraser University and UBC. We also consider candidates from other universities. In fact, some of our key hires have been out of internship programs.
UBIQUITY: What's your over-arching prediction or vision of the future? Do you have anything that you think is destined to happen?
BAAR: One of the things I look at is the overall progression of computing. Some of the major advances in the future are going to be with regard to making computers easier to use, improved access to information, and so on. The computing industry isn't all that old. We as humans aren't evolving at a rate that lets us keep pace with all of what's possible in computing. I think there should be a lot more human factors engineering going into computing. Improved usability will drive the next generation of applications.
UBIQUITY: Does anyone on your staff specialize in usability?
BAAR: We have three people formally trained in usability studies and one person designated as the usability champion. While the whole team is concerned about usability, we have a person specifically designated to make sure that everything that we release has a heavy emphasis on usability.
UBIQUITY: Is there anything else that you'd like to talk about?
BAAR: I'd like to add a view of where we're going in the future, which is certainly in the 3-D world. This is something that we're actively working on right now. The problem you face in three dimensions is different from the screen real estate problem. The problem is occlusion, in other words, things blocking your line of sight. Right now we're very actively moving into three dimensions and developing ways of addressing the occlusion problem.
[For demos of PDT, see http://www.idelix.com/pdtdemos.shtml]
Plan your visit to the SIGGRAPH 2002 exhibit hall using Pliable Display Technology: http://www.idelix.com/pdt_demos_sigg2002.shtml