|
||
|
||
|
|
|
|
|
|
|||
|
|
|
|
|
|||
|
|
|
|
This Garter Fellows interview was conducted face to face with Thad Starner on 29 January 2008 at The Graphics, Visualization and Usability (GVU) Center – College of Computing, Georgia Institute of Technology.
David: As a member of the advisory board to the Georgia Tech College of Computing, I am in contact with leading academic researchers delving into the topic of advanced computing. One of these researchers – a pioneer in wearable computing – is Thad Starner. Thad and I met a few years ago, and it was clear to me that this was someone who Gartner clients would want to meet. Thad began wearing a computer in 1993 while a student at MIT and translated that hands-on experience into a Ph.D. at the MIT Media Lab. He co-founded the IEEE International Symposium on Wearable Computers (ISWC), co-founded the MIT Wearable Computing Project and drove seminal research in the field. He came to Georgia Tech in 1999, where he founded the Contextual Computing Group. To do justice to this interview, I asked Jackie Fenn – Gartner's emerging trends guru and wearable computing researcher – to join me for an in-depth look at what Thad has learned from his years of wearing a computer and living the wired life. Jackie: I understand you've been wearing your computer for 15 years now. Could you tell us what made you start, and then what's kept you going with it for so long? Thad: My interest in wearable computing was purely self-motivated. I was going to MIT as an undergraduate and spending – I guess back then it was only $20,000 a year, and now it's much more – quite a lot of money on my education … and I was discovering that I wasn't remembering any of it. Either I could pay attention in class to what the lecturer was saying and get some intuition about what was going on, or I could spend all my effort on writing notes. But I could not do both. So if I spent time paying attention to the lecturer, I would remember it for about two hours and feel like I actually understood the material. But after two hours, I couldn't remember a thing. If I spent time writing it down, I got no intuition. I got the material down off the whiteboard, but it meant nothing to me. And my handwriting was so bad that I even lost that after two weeks. I couldn't read my own notes. So I was thinking, "Why am I spending all this money on my education when I'm not getting anything from it?" Of course, a lot of education is problem solving. But I really saw the desire to remember these lectures and actually use my information in my daily life as a scientist. The other thing I discovered was that most of my education was not coming from classes. It was coming from face-to-face conversations with the researchers, the grad students and professors that I met in the hallway, or working on my projects. So I was convinced to design a system so that I could actually pay attention in class and take good notes, as well as take notes during these important face-to-face conversations. So whenever somebody says something interesting, I can get it down unobtrusively, quickly, and actually then reuse those notes later on in further conversations. That's why I made a wearable computer. I started out with a laptop. A laptop of course doesn't work when you're walking around, and it also doesn't work well in a classroom. It turned out that having the laptop screen close to me meant that I was changing the physical focus of my eyes from very near (the laptop display) to very far for the whiteboard. That process took too much time. And even though I could type much faster than I could write, the change of physical focus slowed me down too much. I found a laptop couldn't do it. So I went to a heads-up display. I could put the focus on the whiteboard, and suddenly I could keep up. And I found not only could I keep up and take good notes, especially with this one-handed Twiddler keyboard I have, but it actually focused my mental attention, so that during a lecture I got a whole lot more out of it than ever before. And that's why if you've ever seen me in a meeting or a lecture, you'll see me taking notes furiously on my wearable computer not just so I get it, but also because it keeps my attention focused. I can put in new ideas and new offshoots and things I want to ask later into the stream of data that I'm capturing. Jackie: Right from the start, you took this approach of intermediating, so you would take notes rather than just capture everything on video and audio and then try to find it later. Was that a deliberate decision? Thad: Yes, because capturing everything in video and audio is not very useful. It turns out it's not searchable. For example, looking at this interview audio recording right now, if you try to find any given phrase in the audio track, you'd have a very hard time doing it, even with decent speech-to-text translation. More importantly, a lot of conversation is not redigestible. If you actually take a transcript of a conversation and look at it on paper, you'll see that a lot of it makes no sense anymore. I found that it was much better to take my personal notes on the subject and capture the things I did not know. And generally, I can do that in around five words if it's a new concept. And that was a whole lot more useful to me later on than actually capturing audio and video. Jackie: For the benefit of the readers, what do you look like now? What's your gear? Thad: Currently, what I have is a heads-up display. It's a small thing that clips onto my eyeglasses and sits in front of my left eye. It shows me a 640x480 computer screen. I'm typing into an Emacs text buffer, but it's a full graphical environment. It's hooked up to an OQO Model 1 ultramobile PC sitting in a shoulder pack that I have. It's a full computer. It's got a gigahertz processor, half a gigabyte of RAM, 30-gig of hard disk, USB2, 1394, Wi-Fi and all that stuff built into it. I also have a one-handed keyboard called a Twiddler. It has 12 keys on the front and six keys on the top. One of those keys is actually a joystick. It allows me to type up to 130 words per minute, though I maintain about 70 when I'm writing papers, that sort of thing. And in a pocket I have a mobile phone that gives me cellular Internet access when I want it. And this is the system I wear basically every day. Right now, if you looked at my screen, you would see that I have all the previous conversations I had with David here – one on 25 January 2006, one on 25 January 2007, and one 29 January 2008, today. So every time I've met with Gartner, I've kept notes on the conversation. David: When I walked up to Thad a while back – he can tell you the exact date as he just did – we shook hands and said hello, and then he started typing on the Twiddler while he was looking at me. The note-taking began and I clearly felt at a disadvantage, not having a similar electronic record of our past meetings. Thad, you mentioned you started out with a laptop and then you went to a heads-up display. What's the most profound difference – whether it's software, hardware or a combination – that you have today that you didn't have back then? Thad: It's probably the services actually. Back when I started, we had our own instant messaging system up at MIT called Zephyr. And instant messaging was not known to the rest of the world at that time. I was doing mobile e-mail, and I had remote desktop using X Windows. I'd get AltaVista on my eyeball, which was nice. Now it's Google. But a lot of these things that were really revolutionary to people back in the early 1990s are now commonplace, and I can use them just like everybody else does. It's just that my interface is a little bit slicker. So suddenly, I think the rest of the world is realizing the power of mobile computing, especially the mobile phones. And it means the services I can tap into are a whole lot more powerful. David: Consider the iPod generation. These guys have no problem incorporating computer output into their ears. What's the tipping point that's going to push people to accept a computer screen in their visual field? Thad: I think some of that's price point. Right now, the display I'm wearing is $2,000. There is a lot of work now on video iPods. If you go to Apple's main page right now, at the store they have the crystal glasses by Myvu. And they're two-eyed displays designed for seeing video on an iPod. And they're cheap – they're like $200. But they're two-eyed, and that gets in your way. I think there are three paths you can go down for a tipping point in displays:
David: There does seem to be a difference between wearing a one-eyed display, which is designed to provide video into your normal vision stream, versus wearing a two-eyed system, which basically says you're ducking out for entertainment. Thad: Right. Jackie: Can you talk a little bit about the kind of social effects of this? How do you see people reacting differently to you? Are there times when you do feel empowered, that you have an advantage over others? And then how do other people react talking to you, living with you, working with you, when they don't have that same ability? Thad: I think one of the things that I have is fast access to the interface. It takes you forever to get to your iPhone or your RIM BlackBerry or anything like that. In a workday, the average conversation is 38 seconds or less, which means that if it takes 20 seconds to get out your RIM BlackBerry and type a note or send off an e-mail, you won't do it. That's half the conversation right there. So what happens in faculty meetings is the faculty will often ask me to send them notes or an e-mail reminding them of something we just decided to do. So I have become the secretary [laughter]. So that's sort of by default, because people know that I can get to their computers faster than they can get to their computers, through e-mail or whatever. David: So you become a broker. Thad: I become a broker. It also means that people will sometimes use me to run ideas past, because I have my life for the past 15 years on here. I've talked to a lot of people over the years. I was talking to Larry Page back in 1998 about page rank and how that's going to change mobile search. I talked to a guy who made a company called Cyrano, which worked on electronic noses. People run ideas past me just because they know that I have accumulated a weight of knowledge. I'm talking to world experts on different topics, and sometimes, if they happen to hit a topic that I'm not expert on, but I happened to interview an expert at some point, I can pull that up in the conversation pretty seamlessly and actually provide them the contact and the information. People use me also for brainstorming because of this fast access to data. I was at a talk for a very famous computer graphics professor at a conference. And this was while I was still a student. And at the end of his talk, the professor comes down from the podium, walks back to the back of the auditorium where I am, and asks me for the notes on his talk. First of all, I was thinking, "How do you know me? How do you know who I am?" Second of all – the point I vocalized – was, "Why do you want my notes?" And the answer was, "Because I want to see what came across in my talk." And my response was, "I don't write down everything. If there's something I've seen before or heard before I don't write it down. I just write down the new stuff." And the professor said, "Yes, that's exactly what I want. I want to see what you found important. And that way, I can change my talk in the future to improve it." This professor did this and so did Randy Pausch out of CMU. Every time that he saw me in the audience, he would ask me for the notes from his talks, and that's how he got feedback and improved his own speaking. So that was kind of a surprise. David: You're playing a mirror role here; others are looking at you as an avatar for their own performance or for their own communication. That's an interesting idea, but shouldn't somebody sitting and taking notes on a piece of paper have the same capability? Thad: You would think so, but they know that my notes are instantly e-mailable, and they also know they can read my handwriting, because it's typing. Also, I get much more down than anybody writing on pen and paper, because you can write, at maximum, about 25 words per minute. I can type 70. So I am much more complete in my notes. I also type LaTeX math notation. So that means if you have something highly technical, you will get book-quality-formatted notes out of me. There are a couple of other things that are more trivial from the social aspect. First of all, it's an ice-breaker. I can get in at any party [laughter]. David: There's the real "killer app" for all this wearable computing! Thad: Everybody's curious about it. It turns out it's a good way to get a date [laughter]. But there's also some other weird social effects. For example, somebody was walking behind me one day and asked me for the time. I turned and said "3:53" and then continued walking. The person got really upset, caught up with me and said angrily, "How do you know?" They thought that I was blowing them off, that I made up the time since I didn't look at my wrist. I said, "This is a computer display and the time is right in front of my eyeball." And so that was kind of a weird interaction. But the instant access to information is kind of surprising to people. Another thing that also happens when I have somebody that's a VIP visiting the lab is that the people wearing wearable computers will be having this side conversation in text while we feel out the interests of our visitor. We might say, "Okay, he's interested in this work so you take him for lunch, you'll have more time then. I'll take him after lunch because I have a meeting." So there's this negotiation going on in the background while you're having a vocal conversation. Sometimes I even do that during talks. I will be on a panel session and have my students listening in, so to speak, and providing me background references, information while I'm up there pretending to know it all. David: This happens all the time with instant messaging and e-mail while on a conference call, but here you're standing live in front of someone and carrying on almost this ESP-like communication where you've got an edge on them in many respects. Thad: Well that is really power. Google on the eyeballs is really powerful. You learn to begin constructing your sentences so that you can actually fill in the proper information at the end. I'll give you quick example. I was in a class, and the professor teaching said, "Okay, what was the importance of deixis?" And being a smart aleck, I raised my hand and got called, and I said, "Well the importance of deixis was – uh, uh ? I'm sorry. I just hit the wrong key." The rest of the class then broke up laughing, because it was very clear that I had expected to be able to find the notes on that lecture quickly. The only problem is I hit the wrong key in Emacs (my text editor) and got the wrong screen and then couldn't complete it. But then somebody else, a senior professor who was taking the class with me, leaned over and said, "Oh, now I get it. You do that all the time, don't you?" And I said, "Well, yes." He said, "Okay, now I want one." The fact that I could construct my sentences and then fill in the data at the end of the sentence, that I was that slick with it, that's what really compelled that particular person to want one of these devices. David: Many of your examples are interactions with academics. What about the average person at Wal-Mart, or the average TSA staff member at the airport? What other kinds of problems or interesting situations do you run into with the common man? Thad: Occasionally, I'll find some enthusiast – somebody who's taking computer science courses – and then they'll hold up a line as I do a quick demo. But it's actually gotten easier with TSA. With the person on the street, in the beginning, way back when the hardware was a little bit more awkward and was beige in color, it looked like a medical device, and people thought that I was blind or something. They were hesitant to ask about it. As the PalmPilot came out and the idea of mobile computing became common and now with the iPod, the reaction on the street is pretty much instantly, "Oh, where did you get that?" I was in Switzerland recently in the Bahnhofstrasse, and a banker came up to me and said, "Where did you get that?" I said, "Do you know what it is?" He said, "Yeah, obviously. Can I see?" And he wanted it for working at the banks. And these days the interaction is, "Where'd you get it, what software does it run, and how can I use it?" A lot of times in the streets, it's the kids saying, "Can I play video games on it?" David: So it's taking on a consumer appeal. Thad: Yep. The thing that happens now with the person on the street is they always want to see the display. When they see it, they're shocked at how good it is, how clear it is. Because for some reason, people don't realize you can change the optics and make the image look six inches to infinity away from you. The idea that you can have a very clear display in front of your eyeball is kind of shocking to people. Jackie: So what changes when everybody has one? Not just a few people. Thad: Well then we all start doing ESP [laughter]. David: It sounds like it. Thad: Well it's the same thing that's happened with the RIM BlackBerry or the mobile phone. We get a whole lot more efficient. And it used to be in the beginning, when I was doing this, people used to complain, "When everybody has this, then you'll be forced to use this for your business. I don't want this machine because then I'll be working all the time." And that really depends on your personality type. You get the "crackberry addicts" who do actually feel like they have to respond to e-mail all the time and are very religious about fast response. But then you get people who use their devices more to enhance all of their life, not just their work. They actually live a more calm life instead of a more agitated life. Me for example, I actually rarely have my Internet connection on. Most of what I use my wearable computer for is face-to-face conversations, writing papers and keeping notes of my ideas. So I am actually very different from the crackberry addict. I use the wearable computer to augment my own thought, my own intellectual capabilities, as opposed to just my communication abilities. When more people get them, I think we'll be a whole lot more efficient as an economy. I think you'll see people use these devices in unexpected ways. I know some people who will use them to help with their ADD, or with their dyslexia. A lot of this wearable computing comes down to a lifestyle choice more than anything else. David: You've talked about a lot of the benefits. But what about some of the side effects? Some of the downside? Anything about this you don't particularly care for? Thad: The fact that since there are very few of us who do this particular style right now, I'm my own system administrator. Every time that there's a change or something that I want to do, either hardware or software, I have to do it. I don't have somebody to send my computer back to. That's probably my No. 1 complaint. This is a custom-made system by me, for me, and I have to maintain it. Another thing that is annoying for me is when people will start sending me things like movie files over e-mail – because when you're paying per bit for your Net connection over the cell phone, that can be a problem. There's this sort of mismatch between everybody else's workflow and mine right now. What's helped me a lot is the RIM BlackBerrys, because now a lot of people understand how the RIM BlackBerry works and the workflow's gone more toward my way of doing things. David: Any negative social aspects of wearable computing? Thad: Oh, I'm known as "that guy." And it doesn't matter who got the press recently. I mean there's quite a few of us scattered throughout the world, and everybody thinks that we're all the same person. Even if the other guy's Japanese. There's a guy in Osaka who wears this thing, and for some reason, people see the display and think we're all the same person, even though the hardware we're wearing is very different. So that's amusing, more than anything else. The other night, I was at a late-night restaurant here, and I was sitting there working on a grant proposal. And people saw it and said, "Hey, what is that? Can I see it?" And I'm deep in thought on the grant proposal, just trying to do the whole sit-down-and-have-dinner-and-work thing. So occasionally I get interrupted because I'm an oddity still. But that's like having the first iPhone or the first BlackBerry. David: Or the first laptop. What happens if your wearable computer goes down and you're unable to use it for one or two weeks? I would imagine you're just completely lost. Thad: It's painful but not as bad as you might first think. The human memory works on repetition. And generally, it takes about eight repetitions before you actually remember it. So normally, if I'm having a conversation with you, I'll hear it, and that's it. But with the wearable computer, I'll hear it, I'll type it, I'll see it on the screen – three repetitions right there. And I'll keep on seeing it as I take more notes. So I get much more repetition, so even when you take the computer away from me, I have a better memory without the computer than if I never had it in the first place. Does that make sense? David: Sure, it's about what you focus on and what you consistently repeat. Thad: Right, so I have all my notes from my weekly meetings with my grad students. If you took that away from me and asked me about my meeting with one of my students today, I could say, "Well this is what we did," even without the computer in front of me because I had the reinforcement of having typed it in and thought about it and mulled it over before coming to this meeting. Jackie: Do you have your own organization and access system for the information that you have collected over the years? Thad: It's pretty straightforward actually. There's a paper on it. It's written by Kent Lyons, and he did a study on me. He used a camera, and it would take a screen shot every time my screen changed, and log the time. And what he did is he looked at what I did with my system over the course of 55 hours of interaction spread over several weeks. He looked at how I organized information. And it turns out it's pretty straightforward. There are about five main directories. The directory for this conversation, by the way, is "demos." Every time I talk to somebody where I'm presenting information, I just call it a demo. And so I have, Demo/Gartner and I have Demo/ATT – that was the meeting before this one. But I also have a notes directory, which is just general notes on the world, how things work. That includes things like my Rolodex and my "today" file. My today file is actually my calendar. It's just plain-text calendar entries, so I can search it quickly. And I have a wearables directory, which is everything on wearables. I have a Georgia Tech directory, which is everything about my service and administration work at Georgia Tech, and I have a papers directory, which includes papers I read and papers I write. And that's pretty much it. That simple directory structure allows me to access information from the past 15 years of my life within 30 seconds, and it's surprisingly efficient. Jackie: When are you going to have an implantable? Thad: I'm working on that right now. I'm looking at the Hair Club for Men [laughter]. I actually don't see much reason for implantables in general, unless you're doing something like artificial eyes or artificial ears. Doing direct brain interfaces would be very slow, about two seconds per bit. Most things you want to do can be done with an unintrusive display or a hearing-aid-type device. We have some new research that we're working on that's really quite "mad scientist," where we're trying to pick up sign language directly from the motor cortex. If we get that to work, then I'll probably be looking at fiber-optic implants so I can actually communicate with my computer in imagined sign language. But that's 10 years down the road. David: Your group is researching American Sign Language. I met with some of your fellow researchers last time I was here – I'm sure you can tell me the exact date and time [laughter]. What is it about American Sign Language that's so attractive? Thad: Well it turns out that the deaf community is one of the early adopters of technology. So people talk about the RIM BlackBerry and how RIM made this huge impact. If you're deaf, you can't use a normal cell phone. So you use two-way paging – now its called SMS. And so when the BlackBerry came out, of course among the first adopters were the deaf. There are cochlear implants; there's a lot of stuff that the deaf use and they're quite aggressive about. Sign language was a happy accident. It was something I did my master's thesis on. I have a background in pattern recognition. And in pattern recognition, people had a lot of success in speech recognition, so we applied the same technology to computer vision to recognize hand shapes and hand motions, and worked it out very well. So we've been continuing down that path and investigating how to use this technology to help with the deaf community, especially since we have a sign language recognizer. The new crazy thing is that, since sign language is actually distributed throughout the motor cortex, much more than say speech is with the larynx and the throat and tongue muscles, we might literally be able to pull off sign language from the brain, which would be astonishing if we can do it. Jackie: So people think about movements, and that's easier to detect in the brain than the more subtle thinking about speaking? Thad: That's right. The movement for the larynx and your tongue and your jaw is in a very small section of the brain. The movement for your shoulders, elbows, and fingers is scattered much more throughout the motor cortex. And we think we can actually pull out signed phrases through FMRI (functional magnetic resonance imaging). David: To clarify, you would not actually have to go through the physical sign language movements. You could just think about the specific movements, and that would be enough to trigger the motor cortex? Thad: It turns out that imagined movements also activate those regions. Jackie: You are somewhat limited with brain-computer interfaces, but if you're dealing with sign language, then you really have the entire language at your disposal. Thad: In theory, yes. That's the question: How much can we distinguish? Currently, we're going to try our first experiments just on pairs of signs that are maximally different, like cold versus hot, because the motion should be very distinctive in the motor cortex. If that works, we're going to go to phrases, and – this is all in FMRI machines – if that works, then we'll start going to mobile interfaces like functional near-infrared imaging. Jackie: What other areas of research are you looking at right now that you're most intrigued with? Thad: Well, we have a lot of work in wristwatches right now. We believe that this is going to be a popular interface for mobile phones in the future. And so we're doing a lot of work on gesture recognition using the same research background as in sign language, but also in touch screens in the watches and capacitive sensing – just about anything you can think about with gesture. So that's one part of research that we're doing that we're quite excited about. Another thing is dual-purpose speech. The idea there is that I can both talk to my companion and to my computer at the same time. For example, I might say, "Can I meet you some time next week?" And that's a code phrase for my computer to pull up my calendar on my heads-up display so I can see what next week looks like. And of course the person I'm talking to might say, "Oh I'm busy next week, how about the week after that?" And I might say, "Okay, I'll check to see the second week in February." And my computer is tracking the conversation, and I can actually see my calendar on my heads-up display while I'm talking to the person. It makes things much more fast to access. Jackie: Where do you see yourself in 10 years? Do you still want to be seen as "that guy?" Thad: Oh, I think everybody's going to be "that guy" by that point. If I had to bet 10 years from now what I'd be doing – which is a long time for a researcher – we have a particular approach to artificial intelligence (AI) that's unique. In AI, a lot of people have been looking at programming artificial intelligence: writing the rules and the algorithms to make something that thinks. Rod Brooks up at MIT with his robot called Cog tried to make a humanoid robot that can interact with the world the same way we do, and he says a lot of what we consider intelligence is actually embodied in the device, embodied in our arms and fingers and ears and eyes. And by making a robot that he could teach in the human world, we get much more intelligence behavior. I think it's the right direction; the only problem is it's not extreme enough. The robot is bolted to a table, and nobody's trying to treat it as a human baby. It's not going to get the experience that human infants do in learning about the world. So what I'm doing is something a little bit different. We're trying to make sensors that are woven throughout my clothes, so to speak, that allow the computer to see as I see, hear as I hear, and experience the world around me as I experience it. And then this system acts as an agent and helps me with my daily routines – like if someone asks, "Can I meet you next week?" it pulls up my calendar. The computer can then look to see what I do in different situations. For example, if I have a book in front of me, it can see that my hands are actually paging the book. So it doesn't have to recognize the book; all it has to do is recognize the hand motion. Then it can slowly learn what reading is. If I meet somebody at a conference it might hear me say, "Hi David," and shake a hand. Well if it then sees me typing in somebody's name or pulling up that person's file, it might actually start understanding what introductions are. Jackie: Is that even enough, or are you going to have to start by putting those sensors into the baby blankets and having them really learn from the get-go instead of from an adult's view of the world. Thad: Well, one of my colleagues is trying that. My approach is having me be a symbiote with the computer. And it's nice from a researcher standpoint because all the little agents I can do along the way, like the dual-purpose speech work, offer something compelling that I can publish. But every step gets me closer to this goal of making something that's not just an artificial intelligence – it's me. And it can actually fill in for me. And the longer it can fill for me, and the more tasks it can fill in for me, the closer we are to our goal. David: I can see a land of "the wired" and "the nonwired." This tier structure has been going on ever since we've had computing. But I can just imagine a parent trying to raise a child and the child's got a record of everything the parent has ever said. "Well dad, you promised I could go to the zoo on –" and he gives the date and he gives the time and he gives what dad was wearing when he made the promise. And suddenly there's a power inversion, and the social fabric in the family or in business gets inverted, defined by those who have and those who don't. Thad: It's happened with every technology – electricity, radio, television, Internet, SMS, cell phones. But fortunately, our social fabric is really quite robust and people adapt to it. If you really want to see a vision of the future, read Vernor Vinge's "Rainbow's End." Vernor Vinge is both a science fiction writer and a former computer science professor. He also knows a lot of us in the wearable computer world. He's written an entire book on what happens when everybody has augmented reality and wearable computing. Or I should say, when the kids going through school do. And what sort of things that's going to enable. One of the things that I think he's right about is the idea that you hire somebody to do a job. This is going to change. What you're actually doing is you're hiring that person and their social connections in a much more intimate way than ever before. So, for example, I'm writing this talk right now for CTIA, the cellular telephone industry association. I'm talking about the next six big things in mobile interfaces. And my students are contributing to that. And it's done on such a fine level that you never know whose work is mine, and whose work is the students'. I think that in the future you'll see that wearable computing allows you to interact with people seamlessly on a second-by-second basis, and it's going to change the nature of work so much that it's really going to call into question the current IP ideas. David: And it changes the idea of collaboration too. Thad: Yes. David: Could you end with a recommendation for someone who wants to follow your footsteps in wearable computing. How to get started, what to do? Thad: First of all, get yourself associated with a research group so you can have somebody else to help maintain the hardware [laughter]. But a lot of what I have you can reproduce with consumer products, especially these days. Just adapt stuff where possible. And consider access time to be your No. 1 enemy. If it takes you longer to access than two seconds, your use will go down exponentially with the amount of increased time that it takes to get to the interface. If you can get the setup time down to two seconds or less, you will have most of the functionality I do. Jackie & David: Thanks for your time, Thad! Thad: Sure! And be sure to check out some of our papers on the Web and some of our demonstrations. Go to www.cc.gatech.edu/~thad and www.cc.gatech.edu/ccg.  |
|
|
|||||||||||||
|
|||||||||||||||||||
|
