Excerts from an interview with Jaron Lanier
Jaron Lanier, one of the creators of virtual reality, is a uniquelly insightful thinker I just discovered. Although he inhabits the circles of technoutopianists like Ray Kurtzweil and Marvin Minsky, he disagrees with them radically on a number of points. He doesn't think it should be taken for granted that technological innovation is inherantly good. He is much more cognizant of the cultural and ethical dimensions of computing and communication. Yet he remains idealistic, especially regarding ways new and future technologies can be used in creative "post-symbolic" human-to-human communication. I find his ideas to be very exciting. He presents a perspective found outside the dominant dialogues on the future.
Here are his thoughts on the singularity, computers becoming more human and vice versa, and on a future form of art using virtual reality that has elements of cinema, jazz and programming.
The interview can be found here:
http://globetrotter.berkeley.edu/people5/Lanier/lanier-con0.html
Regarding the singularity
"...contrary to Moore's Law, we have software which is required to make computers do anything, and the interesting thing about software is if you write larger and larger programs to take advantage of these computers, they tend to have more and more bugs and difficulties. If you believe that the important thing about computers is how much raw power there is, then you see this curve of them getting better and better, and then at some point, maybe a couple decades from now, it kind of zooms into near infinitely vertical ascent, which is what is called the singularity, when computers become so much better, so quickly, that humans can no longer comprehend their splendor. But if you look at the problems related to software, which is sort of the mirror image, around the same time, you have software descending into an absolutely impassable morass in which no bigger program can ever be written.
So, between these two trends is the truth which is characteristic of reality in which you have conflicting trends and chaos, and just as in evolution, you don't have a smooth motion, you have punctuated equilibria. You have a very complex pattern of progress. Of course, technologists would prefer to only see the good news; who wouldn't? So, they tend to believe that they're riding this chariot to heaven on this inevitably improving computer that will become transcendent. I think it's nonsense, I really do, but there are people who believe it. In fact, I would say it's the dominant school of thought in the academic computer science world now, which I think is a shame."
On "the computer as a creative partner"
"An example I talk about a lot is in Microsoft Word, which is the almost universal means that people use to write -- what could me more important than that? There's an artificial intelligence design component in which the program tries to foresee what you want to do. It corrects your spelling and capitalization, and such things. But a great example of where it fails badly is if it believes you are starting an indented list. It'll throw you into this mode that you can never get out of, and you just essentially have to go along with it. So, you have this very strange thing where you can either bend over backwards to learn what cues it's looking for so that you can create the illusion that it's smart, that it knows you, or you can capitulate and do what it wants you to.
The point is that you're forced to accept the computer as a creative partner and to delegate to yourself a computer-like role instead of a human role. That process and the fact that people are willing to accept it -- in this case it's not a big deal, it's a very minor point, but as a precedent of what might come it's deeply disturbing to me. I feel it's very, very important to point out this notion of computer/human equivalence and to try to oppose it at every turn."
On a vernacular art form to come
"The notion is that in the twentieth century there are a number of new art forms or craft forms: cinema, jazz, and I counted programming, although I think it's the lesser of the three. The notion is [of] this post-symbolic communication in which people are within a virtual world and doing waking-state, collective, intentional dreaming as a form of communication, so you directly create something instead of talking about it as a way of sharing it. That sort of thing, if it comes about, will be seen as a descendant of these forms where jazz is the most sophisticated improvisational structural form as yet (with the possible exception of improvised conversation, I suppose).
Cinema, of course, is the most sophisticated construction of simulated experience as yet, although it might be surpassed by video games very soon. And programming, of course, is the first craft of simulation, which is extremely undeveloped and, I think, in a very poor state right now, but nonetheless has potential."
Here are his thoughts on the singularity, computers becoming more human and vice versa, and on a future form of art using virtual reality that has elements of cinema, jazz and programming.
The interview can be found here:
http://globetrotter.berkeley.edu/people5/Lanier/lanier-con0.html
Regarding the singularity
"...contrary to Moore's Law, we have software which is required to make computers do anything, and the interesting thing about software is if you write larger and larger programs to take advantage of these computers, they tend to have more and more bugs and difficulties. If you believe that the important thing about computers is how much raw power there is, then you see this curve of them getting better and better, and then at some point, maybe a couple decades from now, it kind of zooms into near infinitely vertical ascent, which is what is called the singularity, when computers become so much better, so quickly, that humans can no longer comprehend their splendor. But if you look at the problems related to software, which is sort of the mirror image, around the same time, you have software descending into an absolutely impassable morass in which no bigger program can ever be written.
So, between these two trends is the truth which is characteristic of reality in which you have conflicting trends and chaos, and just as in evolution, you don't have a smooth motion, you have punctuated equilibria. You have a very complex pattern of progress. Of course, technologists would prefer to only see the good news; who wouldn't? So, they tend to believe that they're riding this chariot to heaven on this inevitably improving computer that will become transcendent. I think it's nonsense, I really do, but there are people who believe it. In fact, I would say it's the dominant school of thought in the academic computer science world now, which I think is a shame."
On "the computer as a creative partner"
"An example I talk about a lot is in Microsoft Word, which is the almost universal means that people use to write -- what could me more important than that? There's an artificial intelligence design component in which the program tries to foresee what you want to do. It corrects your spelling and capitalization, and such things. But a great example of where it fails badly is if it believes you are starting an indented list. It'll throw you into this mode that you can never get out of, and you just essentially have to go along with it. So, you have this very strange thing where you can either bend over backwards to learn what cues it's looking for so that you can create the illusion that it's smart, that it knows you, or you can capitulate and do what it wants you to.
The point is that you're forced to accept the computer as a creative partner and to delegate to yourself a computer-like role instead of a human role. That process and the fact that people are willing to accept it -- in this case it's not a big deal, it's a very minor point, but as a precedent of what might come it's deeply disturbing to me. I feel it's very, very important to point out this notion of computer/human equivalence and to try to oppose it at every turn."
On a vernacular art form to come
"The notion is that in the twentieth century there are a number of new art forms or craft forms: cinema, jazz, and I counted programming, although I think it's the lesser of the three. The notion is [of] this post-symbolic communication in which people are within a virtual world and doing waking-state, collective, intentional dreaming as a form of communication, so you directly create something instead of talking about it as a way of sharing it. That sort of thing, if it comes about, will be seen as a descendant of these forms where jazz is the most sophisticated improvisational structural form as yet (with the possible exception of improvised conversation, I suppose).
Cinema, of course, is the most sophisticated construction of simulated experience as yet, although it might be surpassed by video games very soon. And programming, of course, is the first craft of simulation, which is extremely undeveloped and, I think, in a very poor state right now, but nonetheless has potential."