Interview: Vernor Vinge

Science fiction writer Vernor Vinge is best-known for novels like A Fire Upon the Deep and Rainbows End, as well as the concept of The Singularity -- the idea that, in the next couple of decades, humans will become or create a super-human intelligence. What is less well-known is that Vinge has been a free software supporter since the earliest days of the Free Software Foundation (FSF). He has served several times on the jury for the FSF Awards and spoke at an FSF-sponsored event held last month in San Diego to coincide with the LISA conference. As someone who deals regularly with large scale speculations, Vinge places free software in a larger historical context. He even speculates that free software may be one of the factors that will shortly bring about the Singularity.

Part of Vinge's interest in free software is personal. A mathematician and computer scientist, he quickly found that the rise of proprietary software greatly increased the difficulties of teaching.

"When I looked at contracts and user-agreements," he recalls, "the legalese was extraordinarily intimidating, not just because it was complicated, but because it actually seemed to restrict things to the point where it was really difficult to imagine how a student could follow the agreement and still do a project. So the openness that was in the GNU General Public License (GPL) was really very, very welcome." Vinge soon got into the habit of giving students "a little spiel about the GPL" and encouraging them to license their projects under the GPL.

"If they did that," he says, "that would mean I would be able to use their stuff in later projects with other students. And a very large percentage of students in most classes though it was a cool enough idea that they actually did use [the GPL] in their projects."

The historical trend to cooperative infrastructure

However, as important as free software may have been to Vinge in his teaching, what seems to interest him the most is placing free software in a broader historical context. Early on, Vinge came to view free software -- and, later on the Internet and social networking applications that it was instrumental in creating -- as part of a historical trend towards creating an increasingly elaborate "infrastructure of trust and cooperation" that increases the rate of technological advance.

Vinge says: "There are business inventions of the last 2000 years like the widespread use of loans and credit, the use of insurance, the use of limited liability corporations, all of which involve at least at the beginning, a leap of trust." To Vinge, free software, the Internet and social networking are simply the latest extensions to the infrastructure created from such institutions. What these institutions all have in common is that they allow people to interact in more creative and productive ways.

More specifically, he sees free software as the natural and more logical extension of the insight that had produced the shareware culture a few years before the start of the GNU Project and the FSF. With the emergence of the personal computer, entrepreneurs were finding that "the barriers to entry were so low that you didn't need a lot of the overhead that was involved in commercial stuff, and you might just be able to get away with trusting people to pay you. There was much blind feeling around the concept of producing stuff in some sort of context that was different from cars."

According to Vinge, what the GPL and the software and institutions that have grown up around it have produced is "a platform for experimenting with social invention. In the 20th and 19th century, if you wanted to experiment with a new infrastructure for people to interact in, in most cases, like with the railroads, you needed enormous effort. And now -- we can actually do social experiments -- cooperative experiments -- much more cheaply, and you can design ways for people to interact based on just the software guiding what the interactions are like."

Vinge acknowledges that the consequences have not always been beneficial. "One thing the last ten years have proved is that we seem to be very bad at thinking how stuff can be abused," he says, no doubt thinking of such phenomenon as crackers and online predators. "Any time you can make something a hundred or a thousand times cheaper than it was before, there are probably side-effects. But there's a tendency when something works really, really well to push it hard and deliberately avoid thinking about side-effects."

Still, the main change has been beneficial overall in Vinge's view. In particular, he says: "One nice thing is that the price of failure is a lot lower than what you might imagine in the 19th century. Say someone spent ten million 1850 dollars, to make steam-powered dirigibles. Now, it doesn't work, and you've just spent a lot of money, and you don't have anything except a lot of ruined effort. Now, there's still ruined effort if something doesn't work out, but you can retarget or repurpose much more easily, and you can justify taking much larger leaps of faith than you could in 1850." The result is that more experimentation, and more and quicker development becomes possible.

In this view, free software represents the currently most-advanced realization of the possibilities inherent in computer technology. "It's an interesting, science-fictiony, parallel-world story to imagine what would have happened if Richard Stallman hadn't come along with the GPL," says Vinge. "Without Richard Stallman's insight, I think we would have eventually got something like what we got with free software, but it would have been a very interesting muddle. [The process] could have gone for years, and it could easily have gone on so many years that it impacted the era in which really large stuff can be built in the free model. So, overall, I think we would have got something, but, even now, the low overhead involved and even the insight that comes from the GPL would not be with us."

In other words, the GPL and modern computer structures are all "in the tradition of the last few centuries. They're taking the traditions that we saw with the industrial revolution and adding several layers of magnitude to that flexibility."

Bringing on The Singularity

Although speculation is part of Vinge's stock in trade as an SF novelist, he is cautious about predicting the future. "I always rush to say, 'Terrible things could happen!'" he says. "A giant meteor could hit the earth, or a civil war could happen."

However, caution aside, Vinge does concede that "we have the tools to keep running along the same lines for some time. And, in the absence of disaster, it quickly runs to the point where you're talking about stuff that's of the same significance as the rise of the human race within the animal kingdom." In other words, the Singularity arrives.

Vinge does not offer a map of exactly how free software and its infrastructure will lead to the Singularity. But, given the probable inability of humans to understand super-human intelligence, he should not be expected to do so. "It's easy to imagine," he says, "but you run out of adjectives and high-sounding words that could mean anything to someone like us." All that can really be said is that, as the latest manifestation of the historical trend to increasingly complex cooperative infrastructures, free software plays a large role in creating a future in which the Singularity becomes increasingly inevitable.

"I think that's going to happen in the relatively near historical future," says Vinge. "And these sorts of trends are all consistent with that possibility."

Meanwhile, Vinge is personally content with the improvements that have come to free software in the last couple of years. He is particularly pleased that you can download and install a stable and easy to use operating system in an afternoon. "If you look back over the last ten years, you see how easy it's become to do things," he says. "It's silly to put number to this, but it's ten or a hundred times easier now. I can remember spending days getting PPP to work. And now, you just plug this cable into that socket, and it works. I feel much more able to do what I have to do without having to worry very much, without having Catch-22s nibble me to death. Things have really come together in a coherent and useful way."