Betsy Macken of The Archimedes Project interviewed by Brian Hecht on November 28, 1995 "In itself, information is neither accessible nor inaccessible." Betsy Macken is leader of The Archimedes Project of the Center for the Study for Language and Information, an independent research lab at Stanford University. The Archimedes Project takes an innovative approach to making technology accessible to people with disabilities. Tripod: The Archimedes Project seems to have a really unique mission. Can you sum it up in a nutshell? BM: Our mission is to provide access to people with disabilities to computers and to people, through computer technology. Tripod: On your homepage, you talk about long-term design of systems in a way that might almost neutralize disabilities? BM: We feel that people with disabilities have the same need as everyone else for communication. They need to be able to communicate with computers and communicate with people. And sometimes they have trouble with one or the other or both of those two functions. And so what we want to do is make sure that that barrier just doesn't exist. We're trying to work with all kinds of instances when there might be a barrier. For example, on our team, we have a man who is quadriplegic, and a woman who is deaf. And so, they have trouble talking to each other. He's learned to read some of her signs, but then he can't sign back. And she can do some lip reading, but not a lot. So when he talks to her, she can't always hear. So one of the things we're working on is a way for them to communicate. Tripod: And what's the solution to that problem? BM: For that, right now, what we have in the lab is our quadriplegic does all his work for us using his voice and his head pointer on the computer. And he can do everything. He's our Webmaster, and he searches the Web for us, and he designs our Web pages. He does everything. And he can talk to any of the computers we have in the lab because he uses this thing that we've developed called the Total Access System, designed by Neil Scott. And so, if he says to whatever computer he's working on, "Talk to Cathy," then he gets a special window. And he can talk to that window. Then he can read her signs, and she can read what he says by looking in the window. Tripod: It seems like you've designed a pretty specialized system to accomodate particular problems. Are there general approaches that people who are designing technology can take to make it accessible to people with disabilities? BM: Actually, there are two aspects to our system. We feel that we're kind of designing the most general approach. So, what you want, in the end, is that same system, but in a palmtop level. So anyone who has a communication problem can carry a system with them. Or two, if they need it. For example, Cathy who is deaf might take two. So picture two devices the size of a little Sharp Wizard that she would have in her pocket. If she goes into a drugstore and encounters someone who doesn't recognize sign language, then what they have to do is pass notes back and forth. So instead, she hands them one of her devices. And on his screen, it comes up "These two devices can talk to each other. This is part of our technology." So then she types on hers something she wants to go across to the other screen. She says on it, "Hi, I'm deaf. In order to talk to me, you need to talk into this." He talks into that, just the way right now J.B. talks to his big machine. Only now, what he says comes out on Cathy's screen. And now, they can talk to each other. We have the information flowing quick from the person who can talk into theirs, and we can do that really faster than people can type. Now from Cathy's direction, she's still needing to type them, and so that's slower. So then we have the other part of our project, which is ways of speeding up input so that you get the most thought content for the smallest movement... What we envision is, we will design our accessors to be exactly right for the person. So if they need a big screen, then it has a big screen -- maybe they need the letters to be blown up. Or maybe they need a big keyboard, or maybe they need a tiny keyboard, or maybe they need a porta-keyboard. Whatever they need, that's what their accessor has. The other piece to this system is called the Total Access Port, and that's a device about the size of a cigarette case that plugs into -- right now, an IBM PC or a Mac or a Silicon Graphics machine, or a Sun -- and it, basically, takes input from the accessor and translates it into whatever this other machine needs, to think it's getting input from its own keyboard and mouse. Tripod: So that's where the versatility comes in. You're customizing the method of input? BM: Right. We fix it so that it's right for the user. With our other piece, we handle the translation. And the really neat thing about it is that we don't have to get inside the host computer. We basically fool it. Tripod: So it's modular -- you just plug it in. BM: It thinks it's getting input the way it always does. It just doesn't realize. Tripod: Are these practical for people with disabilities, in terms of cost. It seems that you're producing some very specialized systems that need to be customized for each individual person. BM: Well the accessor itself, if they can afford a computer, they can afford an accessor. It's an add-on to their computer. Then, whatever they need in the way of their specialization kind of depends on whatever the off-the-shelf technology is. Right now, we're recommending DragonDictate and we work with that. We have a system that we strip down -- everything except what you need for voice access to make that as low cost as possible. They can get just that, and then this little accessor. And then, they're in business for any computer -- they can take it to work, home, school -- so it's very affordable. Tripod: It seems that computers are inherently better able to accept input from people with a variety of disabilities than just a physical device is -- what sort of long-term planning is going on to build systems in a way that will maximize accessibility? BM: We know that there are quite a few vendors working on these access problems. The ones we know about, except from us, are always inside their boxes. So, IBM is working on something for theirs, and Sun is for theirs, and so on. The problem with that is, if you have it that way, then that's the computer it works on, but you have to have [that same computer] everywhere you go. And it also may not work when they redesign their operating system, or a number of other things that could happen ... We take care of what it will work on with the TAP -- Total Access Port -- and we take care of matching it to the person on the other end. So [the user] puts their money into what they need to match them. Hopefully, it will be good for years and years and years, no matter if computers evolve, or whether manufacturers do or don't add things. Tripod: Here's a kind of theoretical question. In a sense, you're not just overcoming disabilities -- you're working towards overriding any kind of physical interface. Do you think technologies will make bodies and physical interaction obsolete? BM: [Laughs]. I agree with you that what we're doing is something that may appeal to people without disabilities, for example, people who prefer to dictate than type. And it also could be useful for anything that needs voice access. It would be really nice if car radios and cellular phones had it so you don't have to take your eyes off the road -- that's what we call the "30-Second Disability." We'd like ours to work, too, with microwave ovens -- anything that has a computer front end. But I think that for people without physical disabilities, that will just be a matter of preference. I mean, I still use a keyboard even though, here we are, with all this great speech equipment. But I'm old, relatively. People who use speech do actually work faster than those of us who use keyboards. Tripod: Yes, I certainly understand that. I'd like to not have to transcribe this tape ... You can view the the Archimedes Project homepage at http://kanpai.stanford.edu/arch/arch.html