Swimming The Streams

Author of Mirror Worlds, which anticipated the World Wide Web, founder of Mirror Worlds corporation, which may well reshape it, David Gelernter is a demiurge-a creator of worlds, not merely their reflector.
A computer scientist, he has wrought new languages (Linda), realms (tuple-space) and a radical new challenge to the dictatorship of the desktop: the lifestream. Marshalling the immense advances of computer storage, his Scopeware enables its users swiftly and elegantly to move between past, present and future-summoning all the files, e-mails, documents, and images of a past epoch or the plans and promises of a future time-in a translucent three-dimensional array. Gelernter's startling vision does for time what instant global telecommunications and jet travel have done for space.
The consummate nerd and author of several dense and original computer science texts, he evoked an attack from the Unabomber and emerged bent but unbowed as a yet more active and eloquent voice in the public square. But as author of six mainstream books, he also commands the mind of a fiercely personal thinker and creator, artist and author, art critic and short story writer. Like his software, his mind can recreate vividly a lost era, as in 1939, his book on the World's Fair, grasp an elusive value, as in his book on Machine Beauty, or pen a bold discourse on sex roles in his incendiary autobiography, Drawing Life: Surviving the Unabomber. He is one of the great men of our era and an incandescent talker. The American Spectator spoke with him in his office at Yale. Listen carefully.

THE AMERICAN SPECTATOR [henceforth "S"]: As a conservative and a technologist at a liberal arts college like Yale, you're a kind of double pariah?

DAVID GELERNTER [henceforth "G"]: It's a strange thing about technology. In the early 90s, it was hard to get anybody to pay any attention to this stuff. In the late 90s, you could barely get undergraduates to stop to put on their pants before forming new companies and becoming billionaires. Now the technology market is in a slump and so the university is much less interested in technology. Education is the most fad-ridden area in the world. But the general trend is for technologists to be less outcasts than they were before.

G: Until September, I would have said there has been no real change. If anything, a slight deterioration in the position of conservatives, at least at Yale, which was riding high, partly because of its relationship to Mr. and Mrs. Clinton and the mainstreaming of 60s liberalism, 60s leftism, which had the effect of making orthodox, garden-variety leftism more respectable. The White House swings a lot of weight, obviously.
Bush also is a Yalie. This university has in many respects been enormously condescending to Bush, as it was to his father. But there's been a change since September. I've heard more than one person comment, "The smartest people I know are all conservatives." This is a community that admires IQ, and when push comes to shove a lot of people here would admit that they admire Bill Buckley and Paul Wolfowitz. You'd have to torture them for three days to get them to say it publicly. I think the right wing has a monopoly on IQ in this country.

G: I hate to put it in such bald terms. But right-wingers are just smarter than left-wingers. A lot of people didn't feel that they could say it. But since September, it has become slightly easier to admit that you have your doubts about some aspects of the liberal agenda.

S: Okay, but on the other hand, being a conservative technologist is not an easy row to hoe either?

G: The two things are really independent. I mean, I gained enormously from the insularity and parochialism of the technology world, where people have no idea what my political views are.
Flipping around the other way-in some conservative circles, an excessive interest in technology can be as suspicious.
That's probably true, but I don't need to get ahead in conservative circles. And as far as my publications are concerned, if you were going to accuse them of bias, they would tend to be anti-technology.

G: The topics that I write about are art, culture, and history to some extent, and religion. I worked obsessively for the last year on a piece about the aesthetics of Judaism. People sometimes say, "Why don't you ever talk about computer art or technology in art?" I just don't give a damn about computer art. I've never seen any good computer art. So people who read what I write know I don't have a technologized world-view. Maybe some people who don't read what I write and who sort of vaguely know who I am, might be suspicious of me. But I'm not running for anything, so it doesn't really matter.

S: Technology is often associated with the notion of progress. And progress in some conservative circles is anathema. The world is seen as being in the state of decline, with technology to blame: television, Internet, pornography-this notion that somehow technology has contributed to our fallen state.

G: As indeed it has. My book 1939 was mainly about that topic. Intellectual life is generally suppressed on the left-by disdain for scholarship, hatred of art, disdain for good writing, disdain for the intellectual tradition. And so conservatism is forced to be a broad intellectual tent. Certainly there are hardcore traditionalists in the conservative community. And they're correct in the following important sense: The United States was a better and happier nation when I was a child in the 1950s and the early 1960s than it is today. It was more at peace with itself. It was fulfilling its obligations to its children, meeting its moral responsibilities better. In this sense, the United States has experienced a catastrophic deterioration. Parents today must raise children in an environment they know is not as good as the environment that they had when they were children.
But blaming technology is ridiculous. It's a moral transition, an inevitable result of turning this country over to the intellectuals, which is what we did beginning with the universities in the late 1950s and 1960s. Intellectuals are known for their obtuseness, and have been known for centuries to be out of touch with reality. It's got nothing to do with technology. On the other hand, it's also the case that the only progressives today are conservatives, who are optimists, and to be progressive one must be an optimist. To be liberal is status quo reactionary. A significant constituency among conservatives today supports progress, economic growth and technology.
The fact that George Gilder-a first-rate analyst and philosopher of technology-comes from the right is not a weird anomaly. So I don't think I'm that much of an outcast.

G: This pencil is value free. It doesn't make me do good things or bad things. It's a tool. Over the course of human history technology has had no net effect on moral standards. It's made us richer on the whole, but not happier. We do technology not because we believe it will change the world, but for the same reason we eat and drink and breathe, because that's what human beings do. And it doesn't excuse us from our moral responsibilities.

G: If we're doing our jobs, we shape technology. It is our responsibility. Usually we're too passive and don't bother. America has never been in a more passive relationship in the world of technology, and in the world of the intellect in general.

G: Uninterested, disengaged, believing one's self to be powerless, incompetent. They all have the latest computers and they know how many megahertz the processors cycle. But the level of interest in how the machines work and what they ought to be doing is extraordinarily low. Yes, they're complex. But this country used to be composed of engineers. Every 15-year-old boy could tune up a car, change a sparkplug, have some sort of relationship to the technology of the day.

G: It's much simpler to build a Web site than change a sparkplug. Because to change a sparkplug, you need some understanding of what an engine looks like and how it works, where the different parts go.
When my kids were a little younger-they're 11 and 14-we had a lot of fun with electronics kits. You can teach a boy how to solder and what a resistor is and what a capacitor is, just as you could do in 1935. The circuit that you solder together is vastly more complicated-it's got a chip with 10,000 or a million transistors. It's not just an amplifier. It's a burglar alarm, a cell phone, a fire detector. But the basic process hasn't changed that much, we don't put the emphasis on it that we used to. Engineering used to be a highly honorable activity. It's despised now, because intellectuals are more or less in charge of culture and intellectuals have always hated this sort of stuff. They don't want to build things.

G: The PC is an embarrassment to the industry today. Predictions that it will wither away are attractive but hard to believe, because processors are so cheap. PCs will become more ubiquitous, in fact. The issue is what to do with them. Today, the NFL simulator games my young boys play are vastly more sophisticated than the operating system-they have software two or three generations beyond anything I ever do. The fact that it's the 10-year-old boys who are getting the power out of these machines, while adults struggle with 1976 software, is bizarre.
This is an unstable situation. The Desktop was a brilliant idea 25 years ago. But it was designed for a world in which computers were rare and expensive, disks were small or nonexistent, memory was rare and expensive, and processors were slow. And there was very little e-mail, no Internet and no Web. Most computer users and programmers were commercial data processors. But today industry is desperately trying to figure out what the hell to do with this thing.

G: The fault is not Microsoft's. Microsoft did what it needed to do-it provided a stable platform on which we could build my first-generation "lifestream" alternative. The new generation of operating systems doesn't kick out Microsoft. It needs Microsoft. It builds on top of Microsoft, and the stable platform provided by Microsoft is a huge technological advantage.

G: I'm not a lawyer. It's all too easy for people in technology or writers to pontificate about legal questions they know nothing about. Edward Rothstein published an impressive piece in Commentary last summer, which I had my class this term read sentence by sentence. He makes a compelling case that the government, in prosecuting Microsoft, didn't know what the hell it was talking about. Microsoft was not above reproach and committed unpleasant acts.
But the core of the government's case was built on misconceptions: the idea that Microsoft saw Netscape as a threat because Netscape would be a platform to replace Windows. That's silly.
You don't have to love Microsoft to say that Microsoft didn't impose itself on the world. Microsoft provided a service that people chose to pay money for. They did have a choice. They will have more choices in the future. The fact that they have so little choice today is not because of Microsoft. It's because the whole industry has been looking at the wrong problem. For ten years people have been obsessed with the Internet and the Web. I understand that obsession. In my 1991 book, Mirror Worlds, I was early to obsess about the global network. It is exciting.
But as early as 1990 we were already long overdue for a new operating system. A new approach to core information management. Instead of thinking about that problem, the industry, the academic field, the press decided that the Web was the only interesting topic in technology.

S: And yet Microsoft hasn't done much to help us escape it? It's got $36 billion in cash and a 95 percent market share and a complete knowledge of the technology. Where is Microsoft?

G: It is exactly where a large company has to be. A large powerful company is reactionary by definition. If you had that much cash in the bank and that many rich stockholders, you'd be crazy to do anything different. Microsoft would be irresponsible if it tried to lead a technology revolution.

S: It's a fine line though between being conservative and being an obstacle to change, isn't it? For instance, they've dropped JAVA support from the new XP. That could be seen as being more than just reactionary.

G: Yeah, but there's no way they can prevent me from writing JAVA code if I want to. The JAVA community is thriving and developing its tools, making its choices, buying its platforms. This platform running UNIX, running LINUX, is not a Microsoft platform. And anybody in the world has the option of doing exactly what I'm doing here. Microsoft isn't going to force me to run Windows or confiscate JAVA.
The industry is making excuses for itself in blaming Microsoft. Apple hasn't had a thought since 1984; nor has the rest of the industry. It's much easier to blame than to, say, look at what their own labs are producing-essentially nothing.

G: Only in the sense that, once a successful new type of operating system emerges, the live operating system that shows you information as it emerges, then I'm sure Microsoft will build a great version. But don't count on Microsoft leading this transition.

G: The browser and the Web site are obsolete; just a chaotic bunch of links stuck together.

G: Microsoft was absolutely right: It is a trivial piece of software. It's going to disappear just as the device drivers for the printers will disappear.

G: I'm not blaming anyone. Hypertext is very useful in its place. But it is nowhere near a full solution.

S: You've been speculating for many years what it would mean to put one's entire life on the Net. Mirror Worlds seems to be taking a major step in this direction.

G: That's exactly our goal-to make the Net and the computer disappear. We need to get over being so transfixed by the power of the stuff that we warp our lives around it, instead of shaping it to our own needs. What will matter will be information structures. Not the computer, not the operating system. Information has to come in a collection that makes sense, that I know how to navigate. We have a proposed solution.

G: We want to make sense of all the electronic stuff that enters your life and mine: all the drafts and memos that I write; each e-mail message I get; each digital photo I put online; and each fax and voice mail message that comes into my life. When such a new item arrives, I don't want to have to treat it as a file clerk with a file cabinet. I don't want to give it a name, put it in a directory, worry about where to put it and where to search for it when I need it. We want each of these objects to be appended to the end of a growing time-ordered stream.
I'm fundamentally disorganized. Instead of taking the e-mail and copying it into an address book, I just figure, "Well, it's in the e-mail." And of course, it's usually archived or lost or I can never find the data.
We're all getting snowed under by it. I've got to have e-mail messages be first class, to stand shoulder-to-shoulder alongside files, so when I need everything related to the Joe Schwartz meeting I'm having later, I can have a complete collection of Schwartz-related e-mail, appointments, files and memos, whatever.
So whenever a new e-mail shows up or I write a new draft of a memo, it gets popped on the front. When I put in a photo, it gets popped in front, so the stream keeps growing. Each e-mail message is a separate item on the stream, and must be instantaneously searchable and readily browsable. We never ask, "Where is it?" It is always "on the stream." How will we find it? It's up to the software to help me. It's not up to me to figure out, to remember where I put it.

G: Under the name Scopeware, it is in the hands of beta customers. We have been enthused about their response.

G: It would be crazy for me to make commercial predictions about the company. I don't know if our version will succeed. I don't know if the company will succeed. I know that something like this technology will be on 95 percent of the world's computers in ten years. It works. It does what it needs to do. It's coming.

G: We use three techniques that reinforce each other. The stream is ordered in time the same way my life is. It's also searchable and browsable.
If I get an image of a document, such as a fax, I've got to be able to search on any word in a document, not just key words. But I don't want to be forced to use search to pinpoint a document. If I'm looking for something Schwartz once wrote about zeppelins, I don't want to type a long, detailed query that will give me the one thing I want. And it's got to be trivial for me to browse that stuff without taking the time to open each document.

G: Onscreen a stream appears in receding perspective. It looks like a parade of documents. When you sweep a document with a cursor, a summary of the document instantaneously pops up in what we call a "browse window." So just by sweeping the cursor over twenty or thirty documents in a stream onscreen, I can glance at each one in seconds. When I was designing this originally, it was really time order plus search. But when you add browse into the mix, you take a tremendous burden off of search. You don't have to target the search to a bull's eye. You get a handful of stuff, and two seconds later you've found this memo from May of '98 or that spreadsheet from last week.
As new things arrive, they get popped on the front. And the stream has a future as well as a past. So there's a "now" line where things occur, which drifts naturally into the past. And there's a "future" where your plans are-a calendar with your travel or appointments. As the future flows into the past, it becomes an archive.

G: Eventually, it will be peer-to-peer, by which we mean not on a server, but distributed around the Net. We think it should be a stream, but whatever structure it assumes, it's got to be accessible from anywhere.
When I want to watch CSPAN, I don't care if it's on a Hitachi or Toshiba TV. Your information has to be available on any computer at home, at work, on your laptop, on your cell phone, on your Palm. Even also from the computer at the supermarket, on the street and at the airport. You need to be able to walk up to any computer and tune in your whole cyber life.

S: What does this actually look like? My desktop shows a certain gradual evolution in a sense that I'll create a new folder, add more documents, and create more and more of an impenetrable mess.

G: I know. You respond by just throwing things away. But whenever I turn on lifestreams, I see the twenty or thirty e-mails, memos or postings that most recently arrived. I can search back and find earlier things. It goes all the way back to my birth certificate.
The parade also continues behind me in the future-my schedule. And you're backing into the future-you don't know what's going to happen. Whenever I need anything, whenever I need any kind of document, I ask, "Where is it?" and the answer is always the same, "It's on the stream."

S: With the desktop metaphor, we do remember this sinking feeling of, "Have you seen our desk?"

G: Have you seen mine? A computer screen has none of the flexibility of a real desktop, with other spaces around it, bookcases, the floor. It's tiny and you can't take something off the screen and park it somewhere else, on the screen next door. You can put it in folders, but the whole point is to have things where you can see them.
So even in the radically different and more limited computer world of the 1970s, the desktop was a stretch. But today it's just wildly obsolete.

G: We live in time. We think of our lives in temporal terms. And the stream represents our attempt to mirror as organically as possible the natural human flow through life.
A computer used to be just a file cabinet. It just was too weak to do anything other than accept files passively and say, "Okay, tell me the name. I'll put the name on it." We can do better today.

G: No. Because the sub-streams are created automatically and they persist. If you create a "Gilder" stream, the machine sweeps through everything, however many hundreds of thousands, millions, billions, however many documents are in the whole system, and automatically puts the right ones in the Gilder stream. Putting them in the Gilder stream doesn't remove them from any other stream. So, if I had something that talks about Gilder's writing about Joy and I had a "Bill Joy" sub-stream, it would already be on the Joy sub-stream. It could be in however many other sub-streams it wants to be, and it continues to exist forever until I get rid of it.

S: On the hardware side, what enables this to become a reality and what else do we need to make it an easy and practical reality?

G: It's clear when people look at their Palm device or they're looking at their cell phone, there's no way that the desktop model will work on the tiny screen.
Stream is the obvious answer. You can get a good view of a stream through a peephole and then can tune in the stream using a Palm, even using a "WAP" web-phone. It's a thing we didn't predict. But those small screens are an important part of what is driving the commercialization of the lifestream technology.

G: People are just going to vacuum all that stuff up and put it in streams, which will continue growing, and you'll never have to make up another name, another directory. We're not proposing that you take documents and throw them out. Why should you? It's cheaper to keep them than to throw them out.

G: A stream will capture the information flow of the company. It's a concrete form of the company's life and existence. New information comes in all the time. New information gets generated and put on the screen.

G: A stream will run through your home. When you buy a new computer and plug it into the wall, you will bring up your stream. It's not acceptable to have to store certain things on certain machines. But clearly your other electronic devices, such as TV, have to be integrated, too. NBC will produce a video stream, and the TV will be a stream-viewing device.

G: Exactly. NBC publishes stuff in stream form, and the TV is a filter with the stream rushing past it and showing me moving images. If I want to go back and watch it five minutes behind real time or search back and watch something five years ago, I can. If I look into the future, I see what NBC is planning and at twelve o'clock they're going to broadcast this and one o'clock they're going to broadcast that.

G: It's not acceptable for me to walk up to a voice-mail machine and have it say, "There are twelve messages and if you want to know what they are, listen to them." I want to see them. Of the twelve messages on the screen, I can see that the first eight are irrelevant and the ninth is from somebody important.

S: So what you're saying is, "We've invented this vast ability to store information, and now it's too big to do it the computer's way. We, humans, have to do it our way, in a way that's natural to us."

G: That's exactly it. We need to invent a form, kind of like a book, which is as valuable for manufacturers, who know how to build it, as it is for authors. We don't have to decide, "Should I write a single, long sentence or stack the words this way?" Readers buy a book and know how to operate it. They know how to skim through it, how to find the pictures, how to look up something in the index. They know what the table of contents means. It's all part of a robust, flexible form that is tremendously valuable in converting the raw material of words into the reality of information that people can use.

G: I think storage becomes a more sophisticated, higher level of service. I don't want to buy disk space. I don't want to buy file space or directory space. I want to buy stream-storage capacity. If I have a new company with ten employees or twenty, I need a stream that will accumulate all the stuff that defines the company's life. The company's website will probably just be a sub-site of the stream. I want options-to go to the computer store and buy a stream box, and then plug it in, in the basement or somewhere. All the network wires will plug into that one box. All the devices, PCs, laptops-everything will be wireless.

G: You could, but there's what I call the dandelion problem. You may not want dandelions, but there really is no way to eradicate them. Boxes are what the computer industry is good at selling, and what people are used to buying. Computer processors are so cheap, disks are so cheap, that they're going into every house. There is no point in fighting it.

G: It stores my local stream. My house stream with all of my personal stuff and my wife's and my kids'. And everything that constitutes the house itself. It's also crucial that I not be limited to this stream in my basement. I want to tune in your stream also and a million other streams, remotely.

G: Why not? The box will be cheap and powerful, delivering blindingly fast performance. And lots of boxes all over the place will support the dynamic creation of new groups. That is part of the trend toward peer-to-peer computing.

S: But as a business, maybe I want someone else, a storage company, to handle this? I mean obviously we'll buy from Mirror Worlds anyway, so?

G: That would be okay, but whether Mirror Worlds is a multibillion dollar hit or not, this is a game with lots and lots of players that will support these streams.

S: I'm presuming that you've had some discussions with Microsoft about Mirror Worlds.

G: No, we don't talk a lot to Microsoft? We know Microsoft is paying attention to this project-for a long time, Microsoft's been a number-one visitor to our website. When the time is right...

G: For a while, this was a research project at Yale, then it became a commercial-development project. It was a breath of fresh air, moving into the commercial area. The caliber, the interest, the excitement, the enthusiasm, the whole energy level is much higher than it was in the academic world. The obvious explanation is that there's so much more money. But it goes beyond that, because we've reached a point where the intellectual caliber of people in technology and entrepreneurship is strikingly high, as lots of people notice.

S: You write a lot about beauty in technology. Did it affect the way you did this project?

G: I'm not in a position to allege that the stream is beautiful technology, anymore than I can say that paintings in a show are beautiful pieces. I can only say what my goals are. My goals are aesthetic. My goals in designing the stream were to invent a beautiful technology, one that was simultaneously simple and powerful. Whether I succeeded is not for me to say. I'm not an engineer by nature or by training. I'm a writer and a painter and that's the way I approach everything.

S: Well, you have written three dense textbooks of computer science, invented a software language and the Trellis information structure, and contributed to dozens of collections.

G: It's not bad for technology to have some people in it who don't think like technologists, who don't particularly enjoy playing with computers, who don't have patience for figuring out how software works. We could do with more such people. Not copies of me, but more people who are skeptical rather than blown away by computers and software.

S: Can't someone with an outside viewpoint identify a metaphor that has led us astray-metaphor revision?

G: That's the issue. The operative metaphor is the most important thing in technology. The idea of the desktop, the idea of the file tree, the idea of the Web site. These have been the shaping concepts-the metaphors shape the field. If they're right, the field can be shaped right. But they can be wrong, and then the field is shaped wrong. Coming up with the right image is vastly more important than the technical details.
So new metaphors-not new code, new specs, new business clients-need to be written.

G: The hyperlinks at the heart of the Web are ultimately bad, because breadth instead of depth is a recipe for intellectual disaster. The idea that the instant you get bored with something-click and it disappears and you're somewhere else-brings out all the worst overtones in our culture. Our shrinking attention spans, our superficiality, our unwillingness to come to grips with real issues, our insistence on bite-size chunks of everything. Wiring things together with hyperlinks is not an organizational strategy. It's a nonstrategy.

G: Delete the Web and they've still got five orders of magnitude more information than they can handle. Fifty years ago there was more information in the Times and the Herald Tribune every day than a school kid could possibly use. There was more information in a small-town public library or an elementary-school library than any school kid would ever need.
Kids aren't learning to read or write properly, they don't know any history. The idea that we need to dump more information on top of them, and make it easier for them to skate from site to site, is perverse.

S: Is part of the problem with the Net as learning device that you're not developing emotional connections to the subject matter?

G: Yes! Emotion is what allows us to draw analogies. I developed this in The Muse in the Machine. But I think it's a separate problem, really, from the practical software issues that we're faced with.

G: In practical terms, why do we need to understand how the mind works? Understanding how the mind works is its own reward. Human beings care about human beings. We want to know what it is to be human. I don't feel any need to justify that curiosity by saying once we know that, we'll be able to build this software and that software.
Sure, there will be practical advantages to understanding how the mind works. I doubt that we'll see intelligent machines that have minds in the sense of being conscious. But to build machines that can solve hard problems, we're going to have to build machines with fake emotions because we can't solve problems without bringing our emotions to bear. In The Muse in the Machine I claim that it is the emotional footprint of two very different objects that allows me to associate them and come up with a new analogy.

S: Socrates talked about learning as an erotic experience. We observe constantly in modern life how the withdrawal of passion causes the degeneration of intellectual or artistic enterprises.

G: That's exactly what the art world is suffering from-intellectualization. Passion is the only basis for breaking new ground in any area. If you approach something passionlessly, you're condemned to pace back and forth over ground that's already been broken. We as an American society have chosen to turn our cultural institutions over to intellectuals. It was a crazy thing to do. Universities used to have nothing to do with Hollywood or the art community, and little to do with publishing. But we have chosen to put all of our cultural institutions in the hands of institutions like mine. That's a crazy, crazy thing to do.

S: So we have the gap, high culture and low. High culture is what is in charge of the intellectuals and no one goes to. Low culture is what everyone goes to and is debased by.

G: Except that intellectuals have asserted their sovereignty over low or popular culture also, and they have essentially dismissed high culture. Nowadays the people who are not just indifferent to literature but who actively hate literature, are in English departments. People who are not just indifferent to art, but who hate the whole idea of art and great artists, reside in the art history departments.

G: The make-or-break factor in intellectual history and technology is not ideas. It's ideas plus someone who's interested in them, who cares about and talks about them. It's simply not enough to have ideas.
One could be bitter and say, "What good is my work in technology when what I really need is marketing skills?" But I think the absence of marketing passion and outreach is the reason that the field is so crowded and there's so much garbage in it.

S: Assume that everybody has a terabyte drive. What does that mean to you? What other things will they have to have?

G: The industry still does not understand that the problem is not how to conserve storage, but how to squander it creatively. One of the things we talk about in selling people Scopeware, the trade name for lifestreams, is the idea of communal memory and your own history: if you're using it, everything that happens to you online is saved routinely.
We still deal with many companies that say, "We've got to throw everything out. We order people to throw their e-mail out after a certain amount of time," because there are legal issues here. Corporate memory is being flushed down the tubes even though storage is dirt cheap and all sorts of drives are 98 percent empty.

G: In the wake of the terrorist disasters in September, we had several companies with stream systems up and running. When their first-line data systems were wiped out, they were able to reconstruct what they had. Just saving everything in itself is useless-the question is, did you save it in a context that makes it possible not only to find what you want, but to know what you're looking for?
Throwing our lives out, the record of our daily transactions and our daily experiences, is no longer a rational strategy.

G: Right-it'll just be vacuumed up onto the stream, whether there are 15,000 documents or 400,000 documents. It's like a tape that keeps recording and you look at the most recent bit.
E-mail shouldn't be a pain. It should make things easier, not harder. It's a tremendous advantage in a million ways. But our systems were designed in 1977. They're just not adequate to today's environment.

S: Does asymmetry worry you-the fact that bandwidth into and out of homes may be very unequal?

G: I don't buy the argument that the asymmetry is necessarily wrong. Most are consumers, not producers, of information. Everybody's home videos, local songs and whatever they publish won't be of interest to the whole world. There have always been fewer suppliers than consumers in all information exchanges. I don't think the Web is going to change that.

G: The question really is security before and after encryption. I'm content with current-generation encryption systems for storing my personal papers on a server on the Web. The information I store in my bedroom isn't perfectly secure either. Nor is the information I store in my office in a file cabinet. As long as encryption and identification algorithms provide a reasonable approximation for the level of security that I'm happy with in the paper world, then that's good enough.

S: Current trends in computer science-peer-to-peer, distributed systems-all point to decentralization as a model. Do you think it's coincidental that you could say the same thing about the market, which is a way to decentralize economic systems?

G: The forces at work are the same. To invest autocratic or tyrannical powers in a machine is a bad idea in engineering terms, just as to invest such power in a human being is a bad idea in philosophical and political terms. If we have centralized intelligence, we have a potential point of vulnerability and failure. The difference is that the computer is a neutral agent. It doesn't care how much power it has. But any computer can fail, so it makes more sense to limit the potential damage.

S: Yet we have trouble coming up with software that will reliably run even a single PC?

G: Software is no good because we have engineers but no designers. But let's be fair. The field is brand new. Personal desktop computers emerged in the 1970s and we're one generation later. One generation after 1903, when airplanes were invented, we developed the DC3. The first modern commercial airline didn't emerge until 1935. In desktop computers we've got different types of engines, but we're a long way from anything that looks like a modern airplane.

S: Does unlimited storage change an individual's relationship to time? The past used to recede. It used to just go away and you couldn't call it back.

G: That's exactly what it does. Limitless, cheap information storage does for time what the Industrial Revolution did for space. In 1800 a man's view of space was fuzzy except for his immediate locale. Most people never traveled away from the villages where they were born, and even for those with the means to travel, it was slow. And there was no photography to give people an idea of what lay outside their spot-lit space.
We have exactly that same view of time right now. We know the spotlight shines brightly on today and last week and next week. It falls off into the dark quickly. We all have memories, of course. But it's remarkable how bad most people's memories are.
For better or worse, our storage technology will expand our view of time the way the industrial revolution expanded our view of space.

S: We'll all be living our own Remembrances of Things Past, our own little Proustian lifestream?

G: Yeah, and there's more to a Proustian world than just the data. If we subtract artistic discrimination and sense, we just get information, not art. Nonetheless, we have something that people demonstratively want. I mean, they want their memories.

G: We provide that level of information to everybody, only as an individual and as a member of any group he's a member of, which has its own memory narrative in body and casual form.

S: Today there are jetsetters, people who travel around a lot and go to faraway places. They will have full bandwidth multimedia lifestreams. Whereas, other people may have no lifestream at all, because they can't afford it or they live in the wrong place. Like people who travel versus people who don't? Are we going to have people who have streams versus people who don't?

G: I think this issue is attitude and education more than the cost of the technology. Look at the penetration of the VCR-anyone who wants a VCR can have one. The question will not be can people afford them, but can they think in these terms?

S: This is a serious break with history. Until now you basically had what you could haul around in your head or in a few boxes. You're talking about a kind of electronic alter ego, scooping up everything you do. Presidents of the United States have had that for the past forty years. Their every utterance, unless they turn the damn thing off.

G: Presidents have the burdens and the advantages of being constantly shadowed and connected. This kind of technology will not be restricted to five or six immensely powerful individuals. In fact, many people who don't want it, will have it. It is a very big change in our view of time.
I wish that I could read something about what it was actually like to be in the Austrian-Hungarian Army in 1914 and 1915-one of my grandfathers was. But we know nothing. And this blank spot is in the twentieth century, which is within living memory. Clearly there are many blank spots in our communal memory. In the future, there will be fewer of them-and maybe we'll learn something.

S: How do you do archeology when you're confronted with the lifestreams of a billion people?

G: This system needs a focus knob. I can't worry about finicky searches in which I may come back with a huge number of documents. I want to sweep the cursor over and see very quickly what these things are. I want to know at a glance what is in a file or other item arriving at my computer or into my lifestream.

G: Exactly. Links press my nose in the dirt. They say, "Jump through this door." And I say, "Why should I? What's through this door?" You go through the door, you're in a completely different environment. This is an irrational technology. Our science allows us to do much better.

S: Both on the left and the right, there is the ongoing notion that science is dangerous, freedom is a luxury that we cannot afford, technology is something that has to be contained. Do you see that as a problem?

G: We will be in bad shape if we tell scientists and technologists, "You're the bosses. Just do whatever the hell you want," and let progress lead us. You see that in the cloning controversy. People are ready to abdicate their moral responsibilities when some scientist says, "I can do this." Who cares that he can do it? He can also build a hydrogen bomb and blow us all up. We have a responsibility to regulate, in the interests of our moral standards, what technologists and scientists do.

G: I respect Bill Joy. He's a very smart and fine person. But I think his Wired article, "The Future Doesn't Need Us" [April, 2000; TAS response March, 2001], was crazy. The idea that a world that has successfully lived with the threat of thermonuclear weapons for fifty years, is now going to panic because of the threat of software is insane.

S: Software is scary in a funny way to somebody who doesn't know much about it; it dematerializes.

G: Software may be scarier, but that's strictly an effect of ignorance. Solar eclipses were truly terrifying before we understood them.
We've dealt with technologies of mass murder. Software is just a footnote compared to those other things.
Our legitimate threats concern our moral standards, our inability and unwillingness to reach moral judgments. What is truly alarming is how many Americans are unable or unwilling to do that.

G: You don't, anymore than you make moral nerve gas or you make moral tanks or moral hydrogen bombs. It's our job to have the morality. The machine's job is to do what we tell it to do.
People worry that software can get out of control-even the terminology the field uses (such as viruses) supports the illusion that we're dealing with a quasi-organic substance that can get out from under us. If I have a virus on my machine, if I'm not careful, I can let it loose, and lose a lot of data. But there's the plug right over there, and if I'm really worried, I'll unplug it and that will be the end of my problem. There's no way to unplug nerve gas or anthrax or....
I don't deny there's a threat, but in the ranking of threats, a computer virus scores zero compared to the threat of nuclear, chemical or biological weapons.

G: We have been amazingly fruitful in generating new commercial institutions and amazingly dead in generating new cultural institutions. There's no intrinsic reason why a creative society shouldn't create new universities just as it creates new businesses. Not as rapidly or as casually, but why shouldn't there be a significant number of new universities and new cultural institutions? Or publishers? There are so few publishers.

G: Yeah, but there are no websites with the influence of the New Yorker, or The Wall Street Journal or the New York Times.

S: Maybe that's because the New Yorker and The Wall Street Journal are industrial-age artifacts in centralization and the future is more of a free-for-all.

G: I don't think so. The New Yorker was where the best authors published. And that was a centralization, an amazing phenomenon. Its combination of content and form was tremendously successful and we haven't seen that online. People read The American Spectator because it's a package that they like.

S: But will Internet decentralization erode the great centralized institutions? Whether they're universities or newspapers or the New Yorker magazine? Is this technology corrosive?

G: The institution will be what a university in the future will be: a group of people and framework in cyberspace that can glue them together. Not just random email or links-a coherent structure.

S: So an institution like the New York Times or Yale University can make a transition?

G: It can and I think it will. For newspapers and magazines, the transition will be particularly graceful, because they will be able to deliver stuff they can't deliver now. They'll be able to integrate themselves into my life in a way they can't now. The transition of the universities will be a lot messier and complicated because the physical structure plays an important role. We will be losing something when we go beyond it, but the country desperately needs new universities.

G: Existing universities will make the transition, but the important players will be new universities. People from different institutions will cut across the grain of existing institutions.

G: Sure. Or perhaps a depoliticized university, which teaches history for the sake of history, as opposed to leftist ideology or even conservative ideas. It is possible to teach the truth, leaving politics to the side. And I think some nonconservative scholars will be part of this effort. The press will describe it as a "conservative" university. It's hypothetical, but you can almost taste it. It's going to happen.

S: You don't sound like a conservative. Conservatives aren't supposed to be going on about new institutions.

G: Well, being a conservative, in part, means that you don't believe that your generation is the greatest thing since sliced bread. In part, it means respect for the tradition that created you. But in the larger political sense, the world of the left today, the mainstream world, is the status quo. To be a dissident today is to be a conservative. Dissident artists are conservative artists. Dissident writers are conservative writers. Liberals were once dissidents, but today they are the establishment.

G: Exactly. Anybody who looks around and says, "This isn't good enough. We can do better than this." Such a person is a conservative, because conservatives don't think this is good enough. The New York Times and Hillary Clinton are thrilled because they're running the show, and you know they don't want it to change.
I was raised in a Democratic family. I was liberal. But I was never an orthodox liberal because of my grandfather, who came to the United States as a very young child from Russia. He always reminded me that there was a lot in the doctrinaire intellectual environment that was just anti-Americanism. And that if it was okay to oppose the war in Vietnam, if you started clicking over to the other side and saying, "I wish the Viet Cong would win," you were committing an inexcusable sin. He made it impossible for me ever to really make common cause with my fellow students.
Then I had a light-switch moment in the late 1970s when I saw people in the workers' paradise of Vietnam put into sea in rowboats. To see people setting out on the ocean in rowboats and to go on as we did, business as usual, is morally incomprehensible. But there were very few leftists who said, "Maybe we were wrong about Vietnam." Not only were the leftists wrong, but the boat people ended up being savagely kicked and beaten by the left and abandoned to die. How can a morally honest person come out of the 1970s and still be a Democrat? It seemed inconceivable.
So I wasn't that liberal. I was enmeshed in the New York Times basically, which has huge weight on this campus. You can dissent from the Times, but still you read it everyday. You read it on Sunday. It's a huge part of your world view. Until you make the transition of saying, "You know, there are true stories that the New York Times does not print. And there are false things the New York Times does print." Until you make that transition away from having your whole worldview fed to you by the New York Times, it doesn't matter what you call yourself, you are a liberal, really.

G: It would be hard to describe this society as intellectual at all. But it's run by people who think of themselves as intellectuals. They were educated in certain ways and do certain kinds of work and tend to see things from a certain point of view. They are the most powerful group in society today because they run the universities, and the universities are the key players in peopling the elite.

S: How will the culture get beyond this worldview? Can technology lead the way? Can the passion you seek be oriented toward technology.

G: People don't love computers and they never will. They don't love technology. I think they love truth and beauty. They love art, which delivers truth and beauty. They love each other.
S: Arguably we are coming out of a golden age of technology: Do you worry that we may be exiting that age of heroes, that the stock market busts will somehow cause technology to go out of fashion for 50 years? And then big companies will end up dominating the landscape and not much happening?

G: Nothing the stock market does, nothing anybody does, nothing any institution is capable of doing, no army or navy and or air force or police department is capable of stopping technology. Because technology is human nature. Technology has always been the distinguishing feature of human beings. We do it because that's just what we do for a living, for a life, in our lifestreams.

Ahead of his time . . . or?

Very much ahead of his time. Gelernter is no crackpot, although if anyone had a good reason for being so, getting his right hand and the right side of his face blown off by that real crackpot Kaczynski would give him a pretty good excuse. I read his autobiography "Drawing Life" last year, and I can't recommend it highly enough - Gelernter has a lot to say about the culture of victimology we find ourselves in.

That being said, my initial reaction to streams is this - it seems to implicitly assume that we want to remember everything that flies past us. I don't think so - I think there are many things in our daily lives that individuals and organizations want to forget. Individuals may not want to keep that note from one's mistress around, permanently archived, just waiting for someone to access it. Organizations like to forget things all the time - Enron and Arthur Anderson certainly would have been happy to "forget" about their creative accounting, so as to not leave something behind for those who would send them to prison.

But the whole point of a stream is that it sits in the background, invisibly hoovering up everything, so that the stuff you want is always there, without you having to go to the trouble of saving it. So if you decide there are things you don't want to have saved, then the utility of the stream is wrecked - now you're back to having to decide about each and every thing whether or not you want to keep it, and it's no longer effortless or invisible.

Except that intellectuals have asserted their sovereignty over low or popular culture also, and they have essentially dismissed high culture. Nowadays the people who are not just indifferent to literature but who actively hate literature, are in English departments. People who are not just indifferent to art, but who hate the whole idea of art and great artists, reside in the art history departments.

Doesn't it just break your heart, Phaedrus? Thank you for this fascinating article. BBL. All my very best -- bb.

Phew! Our friend Gelernter does go on a bit! Interesting interview though.

In fairness to Bill Joy, I think it should be pointed out that the major concerns in his Wired article, Why the Future Doesn't Need Us, revolved around nanoscale molecular technology, which would operate outside of Gelernter's remedy of "pulling the plug" in a wall outlet.

I think Gelernter is right, though, to place the threat level from such technology lower in the overall scheme of things than Joy does. Joy did come off a bit hysterical and draconian in the Wired article.

Ahead of his time . . . or?

Well, perhaps behind our time? He first says, "People . . . don't love technology." And then he says, "technology is human nature. Technology has always been the distinguishing feature of human beings." What does Gelerntner mean by this? Some ancient paradox about ourselves? Or shall we press him to say that technology is not the distinguishing feature, as if it were the only distinguishing features?

With love, truth, and beauty, in the mix, perhaps we had better qualify our distinguishing feature(s) (whatever they may be) stand in some relation and derive their meaning from what is perhaps not our own sole prerogative.

Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.