Sunday, October 30, 2005

Accelerando, by Charles Stross

Charlie Stross has been hanging around the extropians email lists and it shows. His latest book, Accelerando, (like his earlier book Singularity Sky) is replete with matrioshka brains, characters who live out at least portions of their lives in simulations, people who fork their identities across multiple bodies or software agents and reintegrate the separated memory streams, planet terraforming, and AIs with undiscernible motivations. He weaves it all together into a rousing tale.

The apparent popularity of the novel (the short stories that comprise the sequential vignettes of Accelerando were nominated for quite a few awards) makes me think there's a chance that all the inside references and barely expanded concepts might be familiar enough to most SF readers. To me, it read as a novelization of the best years of the extropians list. All the things that people worried about and all the possible scenarios that people worked through in great detail are visible in at least passing views in the final fleshed-out forms they reached on the list. If the concepts he relies on aren't too SL4 for most readers, it should do quite well, because it's a fun read.

The first part ("Lobster", "Troubadour", and "Tourist") follows the adventures of the peripatetic Manfred Macx as he flits around the world economy, inventing wildly and giving away his patents to deserving charities, or whatever custodian will most piss off his various creditors. In later chapters, the various members of his household and its progeny scatter to the distant corners of the universe along with the rest of expansive humanity. Macx's is not a nuclear family in the conventional sense, so the splintered factions head off in many directions and several combinations, which enables them to be present as humanity and post-humanity inhabit near-earth space, explore farther out, begin turning most of the solar system into computronium, and become less and less recognizable. Necessarily, the story mostly follows characters who have opted to remain more or less human, though, as in Marc Stiegler's Gentle Seduction, the gradual changes accumulate rapidly.

The book is nominated for the Prometheus award, but any libertarianism it displays is subtle enough that some on the nominating committee, while admitting that it's a good read, have whether it deserves a place as a finalist. The book has an air of freedom, and I'd like to argue that, though the libertarianism isn't front-and-center, the book presents a society that works without any central government, and that shows its characters and political factions getting along without resorting to governmental force.

In the first three chapters, we see a fair amount of commerce and crime, and most of the enforcement is privatized, or limited to (threatened) court enforcement of contractual terms. The subplot in which Manfred frees the music is about changing the rules for ownership of property, which some may see as akin to theft, but it seems to me that it's a sensible exploration of alternative choices. If technology makes the old schemes unenforceable, ceding increasing amounts of force to the state (or to property owners) in order to maintain rights that have always been inventions of the state in any case (patent, copyright, trademark) is probably the wrong approach from a libertarian point of view.

Chapter 4 ("Halo", ") starts out with Manfred's daughter, Amber, escaping the clutches of her mother by emigrating to the asteroid belt. She ends up the head of a voluntary justice association that people subscribe to because it's better situated to their problems. The characters refer back to this period throughout the rest of the story, and it's clear that it serves as a reminder that voluntary arrangements work better.

The story ends with a transhuman superintelligence (the cat, Aineko) needing something from Manfred, and having to negotiate with him to get it. It's clear from the power arrangements that Aineko is used to manipulating events so people want to help her, but in this case, she needs a willing assistant, so she offers Manfred something he's wanted all along: to be left alone to live his own life. It seems plenty libertarian to me.

Filed in:

Wednesday, October 26, 2005

Mandelbrot: The (mis)Behavior of Markets

Benoit Mandelbrot's latest book is on finance, which is apparently where he started. The book's argument is that markets are less well-behaved than we've been led to believe by the standard financial theories. And by this he means that the "Random Walk" approach, even as it tries to tell us that we can improve our returns by reducing our risk profile, is leaving us open to crashes that happen far more often than the standard theory says. The book presents a pretty convincing case that fractals (what else were you expecting?) are a better model for the behavior of markets than the normal distribution and its bell curve. If market behavior is fractal, then large upswings and large downturns happen much more frequently than you would expect from a normal distribution.

I expected the book to include some hint of how to structure an investment profile for those who bought the basic argument, but that was completely missing. There were hints that major investment firms can find ways to take advantage of Mandelbrot's claim that good and bad events come in bunches, but no indication of how an individual investor with a modest portfolio could do the same, or even where an individual should look for shelter from wild events once you are convinced that they are more frequent than the standard theories say. Modern Portfolio Theory, even while cautioning individual investors that the market is unpredictable, tells you how to allocate your funds to weather the storms.

To some extent, I read the theory as telling us that wild ups and downs are common enough that the gambler's ruin will overtake you much sooner than you would expect even though markets generally trend up, and you can arbitrage away a fair amount of risk by spreading out your bets. If the fractal argument is correct, you need three or four times as many (uncorrelated) stocks in your portfolio in order to average out the expected downturns that can wipe you out. Since the standard model suggests 10 to 30 stocks, the numbers quickly get prohibitive for individual investors to do anything other than invest in diversified mutual funds.

I'm happy to say that this lack of direction doesn't bug me as much this year as it would have a year or two ago. I'm moving most of my investments out of mutual funds and into investment real estate. It does take a fair amount of time to learn how to find, evaluate, and manage properties, but my current belief is that the returns will be better, and that volatility is much lower in real estate than in stocks or mutual funds. But that's not the subject of this review.

If your investments are in the stock market, even if they are in diversified mutual funds, even if they are in broad-based index funds, you should read and understand Mandelbrot's argument. The bottom line is that modern portfolio theory's recommendation to diversify and rebalance regularly is more important than ever.

Filed in:

Sunday, October 23, 2005

Spirochete Genes are Everywhere!

Edge's John Brockman likes to ask provocative questions. At the beginning of this year, he asked "What do you believe is true even though you cannot prove it?" He got many interesting answers from many interesting people, but two were good enough that I edited them down to be short enough to use in my signature line. The precis for Lynn Margulis' answer looked like this in my signature:

All sensory cells [in all animals] have in common the presence of
... cilia [with a constant] structure.  It provides a strong
argument for common ancestry.  The common ancestor ... was a
spirochete bacterium. 
  --Lynn Margulis (http://edge.org/q2005/q05_7.html#margulis) 

The blog gives me the opportunity to quote the whole thing where a few more people might read it. Margulis wrote:

Our ability to perceive signals in the environment evolved directly from our bacterial ancestors. That is, we, like all other mammals including our apish brothers detect odors, distinguish tastes, hear bird song and drum beats and we too feel the vibrations of the drums. With our eyes closed we detect the light of the rising sun. These abilities to sense our surroundings are a heritage that preceded the evolution of all primates, all vertebrate animals, indeed all animals. Such sensitivities to wafting plant scents, tasty salted mixtures, police cruiser sirens, loving touches and star light register because of our "sensory cells".

These avant guard cells of the nasal passages, the taste buds, the inner ear, the touch receptors in the skin and the retinal rods and cones all have in common the presence at their tips of projections ("cell processes") called cilia. Cilia have a recognizable fine structure. With a very high power ("electron") microscope a precise array of protein tubules, nine, exactly nine pairs of tubules are arranged in a circular array and two singlet tubules are in the center of this array. All sensory cells have this common feature whether in the light-sensitive retina of the eye or the balance-sensitive semicircular canals of the inner ear. Cross-section slices of the tails of human, mouse and even insect (fruit-fly) sperm all share this same instantly recognizable structure too. Why this peculiar pattern? No one knows for sure but it provides the evolutionist with a strong argument for common ancestry. The size (diameter) of the circle (0.25 micrometers) and of the constituent tubules (0.024 micrometers) aligned in the circle is identical in the touch receptors of the human finger and the taste buds of the elephant.

What do I feel that I know, what Oscar Wilde said (that "even true things can be proved")?

Not only that the sensory cilia derive from these exact 9-fold symmetrical structures in protists such as the "waving feet" of the paramecium or the tail of the vaginal-itch protist called Trichomonas vaginalis. Indeed, all biologists agree with the claim that sperm tails and all these forms of sensory cilia share a common ancestry.

But I go much farther. I think the the common ancestor of the cilium, but not the rest of the cell, was a free-swimming entity, a skinny snake-like bacterium that, 1500 million years ago squiggled through muds in a frantic search for food. Attracted by some smells and repelled by others the bacteria, by themselves, already enjoyed a repertoire of sensory abilities that remain with their descendants to this day. In fact, this bacterial ancestor of the cilium never went extinct, rather some of its descendants are uncomfortably close to us today. This hypothetical bacterium, ancestor to all the cilia, was no ordinary rod-shaped little dot.

No, this bacterium who still has many live relatives, entered into symbiotic partnerships with other very different kinds of bacteria. Together this two component partnership swam and stuck together both persisted. What kind of bacterium became an attached symbiont that impelled its partner forward? None other than a squirming spirochete bacterium.

The spirochete group of bacteria includes many harmless mud-dwellers but it also contains a few scary freaks: the treponeme of syphilis and the borrelias of Lyme disease. We animals got our exquisite ability to sense our surroundings—to tell light from dark, noise from silence, motion from stillness and fresh water from brackish brine—from a kind of bacterium whose relatives we despise. Cilia were once free-agents but they became an integral part of all animal cells. Even though the concept that cilia evolved from spirochetes has not been proved I think it is true. Not only is it true but, given the powerfulnew techniques of molecular biology I think the hypothesis will be conclusively proved. In the not-too-distant future people will wonder why so many scientists were so against my idea for so long!

Filed in:

Tuesday, October 18, 2005

Cell Phones and Economic Growth

There was an article in the Sunday New York Times on the rapid expansion of cell phone use in Africa. (Thanks to Norm Hardy for the pointer.) Many people have commented on the implications of cell phones vs. land lines in developing countries, but a new implication struck me on reading this one.

The general background is that in many developing countries, it's easier to install the infrastructure for cell phones than for land lines. No copper wires required, just occasional cell towers. That makes a huge difference when the consumer uptake at the beginning will be sparse, and will grow over time. Many things make installing the copper wires problematic, including political instability, lack of property rights (whose permission do you need in order to put up a line of phone poles or to dig a trench?), spotty infrastructure (roads, for instance) as well as the overall expense.

Norm commented on the possible applications for micro-currencies (think DSR) based on the fact that the phones have the ability to transfer prepaid air minutes phone-to-phone. But what I saw was a rapid ramp on person to person commerce that can grease the wheels to growing trade at the grass roots level. For many years, I have been contributing to Trickle Up, a charity (like Grameen Bank) that makes microfinance loans to the poorest of the poor all over the world. The promise of these organizations is that they give people a chance to start a business, which gives them money to send their children to school, or patronize other tiny local businesses, which ought to be able to kick a tiny local economy in may places.

The NYT article mentions one particular case where access to cell phones improves the prospects for very small scale entrepreneurs: fishermen who are able to choose the best market for their catch before heading to a particular port. It's easy to imagine that coordination on that scale: choosing markets, deciding what to harvest or when, taking orders for build-to-suit could help many small businesses to run more efficiently, which would make it more attractive to operate a business, and could inject more money into a local economy, providing an opportunity for organic local growth.

According to the article, cell phone use is exploding even more than the (presumably optimistic) providers' projections indicated. They calibrated their estimates according to official GDP numbers, not realizing that cell phones would make an even bigger difference in the heretofore informal economy, which the official numbers don't include.

Filed in:

Monday, October 10, 2005

Catalogues I Regularly Read

I thought I'd mention two catalogues that I actually read cover-to-cover whenever they arrive. I get plenty of catalogues that I file for when I need something (Mac stuff, cameras, fruit and nuts, hot sauces), and several that I throw away without opening. These two have a high enough hit rate that I actually page through the whole thing seeing what's interesting and new. I'm not going to mention things like Laissez Faire Books that everyone already knows about.

One is Levenger, "Tools for Serious Readers". They have furniture, innovative office supplies, stationary, carrying cases and more. I've probably purchased things from them a dozen times. Before the era of the Palm, I used their note cards and carrying case for my calendar, contact list, todos, etc. They have wonderful variations on paperclips, papercutters, and notepads that are the perfect answer to many minor problems. Their bags are well-designed (need a computer carrying case?), they have great side tables for books, as well as lap desks to make working in an easy chair more comfortable. They spend too many pages on fancy fountain pens, but their reading lights can't be matched in any of the lighting stores I've looked in over the years.

The other catalogue I read is Daedalus Books. They sell remainders, and seem to have good taste. I regularly find interesting history, science, occasionally science fiction, and general non-fiction. For many years, I carefully read all the pages of children's books looking for good new stuff to read to the nieces and nephews, but they're mostly past that age at this point. A few years ago, Ted Kaehler mentioned that he'd been reading a lot about the history of espionage in WW II, and I found that Daedalus had one or two of those most months, so I've been reading them as well.

When you know what you're looking for, searching on-line works well enough, but when you want someone to recommend things that you hadn't realized you needed, there's still no substitute for a paper catalogue. I can do a much more thorough job of perusing the pages of Daedalus than any book store or library I've ever been in, and I don't know of any stores that can change their inventory every month, or have as high a hit rate as Levenger.

Sunday, October 02, 2005

Joss Whedon's Serenity

We went to see Serenity last night with some friends. Half of the group had seen all the episodes of the TV show from the DVDs, and half hadn't seen any. (I'm in the latter group.) All of us were familiar with Joss Whedon's work from Buffy and Angel. Everyone enjoyed the movie, so if anyone was worried that the movie might not have been able to reach those who hadn't seen the TV show, this may be a relevant datapoint. Of course, all of us are libertarian or somewhat libertarian, so that may skew the sample some.

What a good movie! I thought the story was very well told. Joss included enough information to catch us all up with the characters and the milieu, and the action came fast and furious. The characters were well motivated; we were quickly shown who had a crush on whom, which characters had long standing friendships, and so on. The conflict was interesting, the antagonists were powerful, but beatable, and there were plenty of mysteries to solve.

I'm not going to try to do an in-depth review of the themes and subtexts. There are people who are far more into this than I, and I'm sure there were subtleties in the interactions between characters that weren't apparent from the movie. But it was still fun to watch, and the themes that Julian Sanchez Reason described were apparent even though I wasn't already immersed in the canon. River Tam voices the laissez-faire creed in the movie's preface: "People don't like to be meddled with." It's clear that that's the conflict right from the start, and keeping out of the way of a meddlesome government is Mal's goal. When he finds out that he's gotten hold of something with deeper implications, he struggles with it, but finds that the only way to live with himself is to do what he can to ensure that the alliance won't continue meddling.

Filed in:

Saturday, October 01, 2005

Eric Baum: What is Thought?

There was some substance to Eric Baum's What is Thought?, but it didn't add much to my understanding of the title question. Baum's has been working in AI, writing software systems, some of them reasonably successful. He argues that intelligence can be recognized in any system that acts effectively based on a compact representation of the world. The evolution of more successful creatures was a continuing process of discovery of better representations of the nature of the world. He takes it as a given that the mind is a program. (Which is part of why he didn't add much to my understanding of what thought consists of.)

His main argument seems to be that thought and the mind are the result of evolution producing better and better representations. This is neither surprising nor new. He makes little attempt to explain how the mind works beyond the fact that it uses analogies. Apparently he thinks this is fundamental and important. His focus is on economy of representation, and he apparently believes that reusing representations is the source of power. I'm not sure why this doesn't seem more powerful to me, since I often say that the source of the power of OOP is in the re-use that polymorphism gives you. I think the difference is that he didn't show any mechanisms by which the mind could be re-using modules. He shows how modular tools can be re-used in other architectures, but he doesn't say what the architecture of the mind is.

The valuable contribution of the book is in its description of the Hayek system that Baum developed with colleagues at NEC Research. This was a general learning system based on evolution and agoric feedback. The design of this system made possible the evolution of separate agents that could work together to reach a goal. I haven't heard of other systems that were successful with this combination of goals, though there have been other attempts. The aspects of the approach that seem crucial to the success were that one agent at a time was allowed to make changes, and the agent was chosen at any point by an auction among the agents. The agent who wins an auction pays the winning bid to the previous agent, and afterward gets paid the price bid by the next winner. The agents are charged rent when they're not running, and for cpu when they are running. Randomly mutated copies of the more successful agents are added to the population over time.

It's not obvious how you get this economy started, but once it's going, each agent competes to raise the value of the current state of the world so its successors will bid more than the previous state was worth. As long as there's an agent who can produce a final state of the world that is valuable according to some external metric, everyone can earn a living wage along the way. The problems he applied this ecology to included a simple blocks world problem, Rubik's Cube, and the traveling salesman problem.

According to someone familiar with the literature, the only other systems that tried to use markets in a similar way made the mistake of choosing as the next active agent by a lottery proportional to agents' bids rather than simply choosing the highest bidder. This reduces the link between agents that can cooperatively solve the problem, enabling less effective agents to intervene and destroy any order that has been built up. The clean approach in Hayek seems much more likely to work for long chains of agents, and to give evolution a better opportunity to discover shortcuts and more efficient approaches.

If you have a system that works, but not well, modified agents can attempt to shortcut the working path. If they reliably do better, the bypassed agents can wither away. If the new agent is unreliable, both paths can stick around long enough to compete, and possibly combine with an intermediate version that has a better model of when the different plans are applicable.

His discussion of the evolution of language in chapter 13 was particularly unsatisfying. He pointed out that language can't start growing beyond the level of innate language (which seems to exist in many creatures, from honeybees to vervets) until you get both the ability to learn to produce new words and the ability to learn new words. He ends up not explaining how evolution surpassed this hurdle; instead treating it as the reason why it was so rare for evolution to produce the ability.

Filed in: