I work with wearables pioneer Thad Starner at Georgia Tech's Contextual Computing Group. Thad has been wearing his wearable computer for four or five years, and he uses it almost literally all the time. Here is a section of his PHD thesis on wearable computing (unfortunately in raw latex, but readable) detailing some of the many ways he uses his wearable. There are quite a few other interesting papers on the page from the class on Ubiquitous and Mobile computing he taught last spring. From my experience with Thad and the CCG, I've seen several issues that will influence widespread use of wearable computers:
Power Consumption
This is one of the areas that a lot of progress has been made in. IBM's wearable gets three hours of battery life. Thad's wearable gets fifteen. Know what else? He never has his hard drives spin down or his display turn off. He accomplishes this amazing feat by using extremely low power hardware; his wearable is composed of PC-104 boards. You can find more information about the hardware at MIT's Wearable Computing Project.
Display
The biggest obstacle to widespread acceptance of wearable computers, in my opinion, is the display. They are, at the moment, extrememly expensive. Quite a bit of technical progress has been made, however. Kopin makes some tiny displays (unfortunately no wearable designs shown on their page), but the ones we use most in the CCG are the ones made by MicroOptical. This page has a photo of the clip-on glasses display prototype; we've got non-prototype models in use, and let me tell you, they are sweet. They are extremely lightweight, slim, and space-age looking. Of course, they're also about $5000 apiece, but that's why we have grants. =)
Input
Input is a bit of a problem. Nobody's developed an intuitive, easy to use input device. The Twiddler is the one most used by wearables researchers at the moment. It's a chording, one-handed keyboard with 12 keys (three rows of four) on the front for the fingers, and five keys for the thumb. It acts as both keyboard and mouse, as it has tilt sensors that let you control the pointer. The Twiddler is neat, and very useful, but it's about as hard to learn as touch-typing. MIT's wearables pages have some info on other input devices buried within them.
Interface
This is another potential obstacle to widespread use of wearable computers. Thad runs Linux (Slackware, I belive) on his wearable, and his interface to everything is: XEmacs! Yes, XEmacs, heavily modified to do everything he needs it to. One of the most revolutionary applications it uses is the Remberance Agent (PDF). This watches the files on your drive and what you're typing, and suggests a list of related files every 10 seconds or so. In this way, you can see things that might be related to what you're doing currently. For instance, if I'm typing up an article (such as this one), and I talk about Brad Rhodes, the Remberance Agent might display a filename such as "rhodes-RA.pdf", reminding me that he was the one who wrote the Remberance Agent. Or, if I'd met him at ISWC and put his name in "ISWC99-people.txt", that could come up and remind me as well.
Size
Size is one of the least concerning of any of these issues. Technology will progress, and things will get smaller. Eventually, we can expect to have powerful computers that fit in our pockets, or on our wrists (check here for a wrist-sized palm pilot). Size is currently a consideration, but it's the least of them.
True, true, BUT - I'm more interested in having an MP3 player that you don't have to boot up! Not to mention, there's problems with control (you really want an entire keyboard in your front seat?), display (how do you know what song is playing, when you've got 150?), and skipping (PC CD-Roms aren't known for their skip protection).
I've actually made one of these, though. It used a MediaGX board and fit in a briefcase. No hard drive; it created a ramdrive (lots of memory!), unzipped win95 to the ramdrive and booted off that. For display, I used the old SBTalker speech->text synth to read the name of the song that was playing. Fairly slick, but took a while to boot. I had the whole keyboard up front, but you can get a keypad-only keyboard for about $25. I solved the skipping problem by having Winamp cache the entire file. Still, I'd rather have a tiny unit. =)
Re:Organization Systems
on
Wearable PCs
·
· Score: 1
I was in one of Thad Starner's classes earlier this year, and he talked about this a little. He takes notes into seperate files for each event (class, conference, meeting, etc...). He uses a piece of software called a remembrance agent, which shows a list of files relevant to whatever he's typing in emacs at the moment. Go here for a (unfortunately partially raw-latex) description of the remembrance agent and various ways Thad uses his wearable in daily life. This is the page of readings for his class, with various other interesting wearable papers.
Slashdot Mirror
This is one of the areas that a lot of progress has been made in. IBM's wearable gets three hours of battery life. Thad's wearable gets fifteen. Know what else? He never has his hard drives spin down or his display turn off. He accomplishes this amazing feat by using extremely low power hardware; his wearable is composed of PC-104 boards. You can find more information about the hardware at MIT's Wearable Computing Project.
The biggest obstacle to widespread acceptance of wearable computers, in my opinion, is the display. They are, at the moment, extrememly expensive. Quite a bit of technical progress has been made, however. Kopin makes some tiny displays (unfortunately no wearable designs shown on their page), but the ones we use most in the CCG are the ones made by MicroOptical. This page has a photo of the clip-on glasses display prototype; we've got non-prototype models in use, and let me tell you, they are sweet. They are extremely lightweight, slim, and space-age looking. Of course, they're also about $5000 apiece, but that's why we have grants. =)
Input is a bit of a problem. Nobody's developed an intuitive, easy to use input device. The Twiddler is the one most used by wearables researchers at the moment. It's a chording, one-handed keyboard with 12 keys (three rows of four) on the front for the fingers, and five keys for the thumb. It acts as both keyboard and mouse, as it has tilt sensors that let you control the pointer. The Twiddler is neat, and very useful, but it's about as hard to learn as touch-typing. MIT's wearables pages have some info on other input devices buried within them.
This is another potential obstacle to widespread use of wearable computers. Thad runs Linux (Slackware, I belive) on his wearable, and his interface to everything is: XEmacs! Yes, XEmacs, heavily modified to do everything he needs it to. One of the most revolutionary applications it uses is the Remberance Agent (PDF). This watches the files on your drive and what you're typing, and suggests a list of related files every 10 seconds or so. In this way, you can see things that might be related to what you're doing currently. For instance, if I'm typing up an article (such as this one), and I talk about Brad Rhodes, the Remberance Agent might display a filename such as "rhodes-RA.pdf", reminding me that he was the one who wrote the Remberance Agent. Or, if I'd met him at ISWC and put his name in "ISWC99-people.txt", that could come up and remind me as well.
Size is one of the least concerning of any of these issues. Technology will progress, and things will get smaller. Eventually, we can expect to have powerful computers that fit in our pockets, or on our wrists (check here for a wrist-sized palm pilot). Size is currently a consideration, but it's the least of them.
True, true, BUT - I'm more interested in having an MP3 player that you don't have to boot up! Not to mention, there's problems with control (you really want an entire keyboard in your front seat?), display (how do you know what song is playing, when you've got 150?), and skipping (PC CD-Roms aren't known for their skip protection).
I've actually made one of these, though. It used a MediaGX board and fit in a briefcase. No hard drive; it created a ramdrive (lots of memory!), unzipped win95 to the ramdrive and booted off that. For display, I used the old SBTalker speech->text synth to read the name of the song that was playing. Fairly slick, but took a while to boot. I had the whole keyboard up front, but you can get a keypad-only keyboard for about $25. I solved the skipping problem by having Winamp cache the entire file. Still, I'd rather have a tiny unit. =)
I was in one of Thad Starner's classes earlier this year, and he talked about this a little. He takes notes into seperate files for each event (class, conference, meeting, etc...). He uses a piece of software called a remembrance agent, which shows a list of files relevant to whatever he's typing in emacs at the moment. Go here for a (unfortunately partially raw-latex) description of the remembrance agent and various ways Thad uses his wearable in daily life. This is the page of readings for his class, with various other interesting wearable papers.