Interview-Vernor Vinge pg3
From FutureNovo - Anticipating things to come
August 10, 2008
Richard Yonck, FutureNovo
FN: I think Fairmont High's school motto, "Working hard not to become obsolete", is both fitting and hilarious. (Vinge laughs.) It's not exactly subtle, but it very succinctly states the situation we're all going to be dealing with in years to come. Or at least many of us. As more and more careers become obsolete, do you have any thoughts about how we might or should deal with the social and economic repercussions?
Vinge: Alas, no solutions. But even if a person is not obsessive about the Singularity, viewing it as one scenario to fit events against is interesting. When unemployment gets massive enough, you no longer call it unemployment, you call it a biological sea-change. There was a time in the sixties and seventies and eighties when academics and white-collar workers could say with comfortable objectivity, "Isn't it a shame what this technology is doing to blue-collar employment? But we intellectuals are safe and we'll create social programs to cushion the stresses for the less fortunate." Then in the eighties and nineties tech stresses started biting into white-collar workers. And nowadays, there are very few jobs untouched, except for the most talented people — of whom only the most naive may still feel safe. In the writing business, we're under an awful lot of stress just because so many good writers are writing. Of course for us writers this is not directly a Singularity issue, but in almost every department of human affairs, productivity and quality increases are great news for consumers but these very same consumers are running a Red Queen's race in their own particular production domains. It's the two sides of the same coin. One side is the best it's ever been for humans and the other is "Oh my god, how am I going to compete?"
Predicting beyond the Singularity is intrinsically more difficult than the prognostication of futurologies past. For instance, Bill Joy may be right to say that the future doesn't need us. On the other hand, it's entirely possible that not only does the future need us, it needs us a lot. In nature, one recurring theme is that while extinctions are ubiquitous, earlier forms and solutions persist in support of grander schemes. Also, and this is a point I might have more trouble documenting, over and over again you get a situation where relatively inefficient ways of doing things are appropriate because the environment can't support anything more. For instance there is the rise of aerobic metabolisms. There is the move of life onto land. In cases like these, the earlier, so-called primitive life forms make a new environment that can support something more advanced. If all the bacteria were to die tonight, humanity — as well as most plant and animal life — would die very soon. To me, this is an example of life depending on earlier life. We could look at our use of new substrates in a similar way: the silicon, the germanium, gallium-arsenide, fiber optics, and whatever there's going to be in the future.
FN: They're in support of whatever is to come.
Vinge: These are structures that could not have emerged in the pre-human ecology. Nor could they directly redevelop after a technological fall. Science fiction in the 1950s and 1960s had the notion that there could be silicon-based life forms — not computers, but biological life forms — somewhere in the universe. But carbon, oxygen, hydrogen and friends are the only way we've ever seen to do this. It's likely that what arises naturally, at least in the environments that we have access to, isn't much more broadly based than that. On the other hand, now we have a suspicion that when you start making computational devices, there can be much more effective alternative substrates for thought. Such critters just can't do it themselves to start with. So in this imagining, machine life forms are really just another step in the evolution of life.
You're familiar with the term "intelligence amplification", using the computers to enhance human intelligence?
FN: Sure.
Vinge: IA is a plausible — and for many people including me, an attractive — alternative to the alienness of Digital Gaia or network mind or standalone AI. And yet, one could regard the whole human adventure with artifacts as an enhancement of life. It really is just biology beyond our biology. It's biology that will probably depend on that earlier biology.
Charles Stross has just come out with a book called Saturn's Children.
FN: Good, I haven't read that one yet.
Vinge: It's a great book. At one point he has a robot, the protagonist, talking about humans — who are no longer around, by the way. Neither are eukaryotic life forms! The protagonist remarks of classical biology that it seems like a great waste to have built into every organism the entire factory system for building that organism. Actually, that's not precisely true, since we depend on infrastructure in the bacterial world and the plant world and stuff like that. But this character has a point, because in her world, the different components that go into making people are from all over the world and the solar system. It's essentially economics that brings them together in the right place at the right time in order to make new people. These machines have a style of reproduction that is very different from life before, and it's very efficient. But it couldn't emerge spontaneously. I figure that even wildly alien futures would need backups!
FN: Over eons of evolution, humans have learned to anticipate future events better than any other species. As a corollary of this, we tend to view significant change with suspicion, even fear. Obviously, this has provided us with considerable advantage, but as technological and social change accelerate, how do you think we can balance our inherent caution with the need to adapt quickly?
Vinge: That's a core survival question. Getting the proper balance is one of the most important things we must get right. I don't have the answer, but I think the new populism discussed before gives us the best hope: Humans communicating and collaborating, empowered by technology, is the single most powerful tool for surviving the problems civilization faces — and at the same time putting in place the steadily improving smarts needed to navigate a happy outcome for Singularity trends. Fortunately, there are clear trends in this direction. The positive effect of even basic cell-phone access in the third world (trumping much of all prior foreign aid) is a spectacular example. Wikipedia, which has rightly been debated and criticized, has shown us some of the power of network mediated cooperation. We should recognize the Internet as a platform for prototyping the tools for empowerment on the future Internet.
FN: What current or near-term technology really excites you today?
Vinge: Networked embedded microprocessors excite me, both happily and nervously. I wish there was more effort spent on failure mode planning. We're getting to the point where we could have wide-area failures of Almost Everything if there was a failure in the embedded micros. We know there are wide-area threats to embedded systems, where they all can be brought down in a few seconds. In the modern world, we have lots of things to worry about, but this one should rate higher priority.
The promise of space still excites me. The failure of the space program — and by that, I mean cheap access to space — should be a case study for anybody making extravagantly optimistic projections about tech progress. Getting self-sufficient human civilization into space is an important survival issue — of critical importance if the Singularity doesn't happen, and important insurance in any case (see http://www-rohan.sdsu.edu/faculty/vinge/longnow). I think the majority of people realize that this is very, very important and desirable. That fact has made it possible to get large sums of taxpayer money for space development. But now after more than fifty years, our space programs are still based on an infrastructure that costs something like five thousand dollars a pound to low earth orbit. Some space traffic makes good sense at five thousand dollars a pound, and I'm not objecting to that. But people who talk about major manned space programs on such a price basis are selling a doubly gold-plated product. It's gold-plated first because it costs so much money to put cargo in space. And second, every device you put in space has to be enormously — and expensively — reliable since recovery and/or repair are normally impractical. Extending civilization beyond Earth is important to the long-term survival of life — of all forms! Cheap access to space, far below five thousand dollars a pound, is essential to that extension. Spaceflight is something I grew up with and still avidly support, but at this point any major humans-to-space initiative should have much cheaper launch technology as a prerequisite.
