Articles Tagged ‘fiction’ - Less Wrong

Case Study: the Death Note Script and Bayes

Fri, 04 Jan 2013 15:33:37 +1100

Submitted by gwern • 24 votes • 43 comments

I give a history of the 2009 leaked script, discuss internal & external evidence for its authenticity including stylometrics; and then give a simple step-by-step Bayesian analysis of each point. We finish with high confidence in the script's authenticity, discussion of how this analysis was surprisingly enlightening, and what followup work the analysis suggests would be most valuable.

If you're already familiar this particular leaked 2009 live-action script, please write down your current best guess as to how likely it is to be authentic.

This is intended to be easy to understand and essentially beginner-level for Bayes's theorem and fermi estimates, like my other Death Note essay (information theory, crypto) or my console insurance page (efficient markets, positive psychology, expected value).

Be sure to check out the controversial twist ending!

(I'm sorry to post just a link, but I briefly thought about writing it and all the math in the LW edit box and decided that cutting my wrists sounded both quicker and more enjoyable. Unfortunately, there seems to be a math problem in the Google Chrome/Chromium browser where fractions simply don't render, apparently due to not enabling Webkit's MathML code; if fractions don't render for you, well, I know the math works well in my Iceweasel and it seems to work well in other Firefoxes.)

43 comments

Fictional Bias

Mon, 02 Apr 2012 12:10:20 +1000

Submitted by thomblake • 1 votes • 55 comments

As rationalists, we are trained to maintain constant vigilance against common errors in our own thinking. Still, we must be especially careful of biases that are unusually common amongst our kind.

Consider the following scenario: Frodo Baggins is buying pants. Which of these is he most likely to buy:

(a) 32/30

(b) 48/32

If you're like me, your answer is (c). Frodo, as we know, is about 4' tall, so his inseam is much more likely 20'' than 30''.

But like me, you'd be wrong. Since there aren't *actually* any hobbits, all we know is that we're talking about a person named Frodo Baggins, who is male. And the most common pants size is 32/30, so the correct answer, given our actual state of knowledge about the real world, is (a).

This is what researchers call Fictional Bias, and there is evidence that it affects virtually every domain of decision-making. The mistake is using information from fictional sources in real contexts. It is the more-pernicious cousin of generalizing from fictional evidence - instead of merely generalizing, we treat real objects and persons as though they are the specific fictional entities they resemble.

We're of course familiar with particularly egregious examples of people confusing fiction with reality. For example in the 1930's, there were multiple cases where someone was killed for having the same name as the serial killer of children from the movie M. But these could be written off as merely disturbed individuals. But as it turns out, we're affected by this bias in our daily lives.

Examples abound in the literature, though the name "fictional bias" is not always used. A 1984 study by Dr. Sidney Zweibel and Dr. Emilio Lizardo asked subjects to trust someone with an unusual name. Subjects were 70% less likely to trust when the person had the same name as a fictional villain. A 1989 study by Dr. Wayne Szalinski established that subjects were 89% more likely to agree to take an experimental drug, when it was named after a fictional drug (for example Ephemerol).

The moral? It stands to reason that we need to be careful where our intuitive inferences come from. For a group that consumes so much science fiction and fantasy, we must be especially on our guard. (Also, this post was an April Fools prank; the effect may or may not be real, and all citations are either irrelevant or fictional.) It might even behoove us to discourage the reading of fiction amongst aspiring rationalists.

Kahneman, D. and Frederick, S. 2002. Representativeness revisited: Attribute substitution in intuitive judgment. Pp 49-81 in Gilovich, T., Griffin, D. and Kahneman, D., eds. Heuristics and Biases: The Psychology of Intuitive Judgment. Cambridge University Press, Cambridge.

Szalinski, W. 1989. Recollection of fictional medication. Journal of Cognitive Minification, 67: 173-186.

Tversky, A. and Kahneman, D. 1982. Judgments of and by representativeness. Pp 84-98 in Kahneman, D., Slovic, P., and Tversky, A., eds. Judgment under uncertainty: Heuristics and biases. New York: Cambridge University Press.

Tversky, A. and Kahneman, D. 1983. Extensional versus intuitive reasoning: The conjunction fallacy in probability judgment. Psychological Review, 90: 293-315.

Zweibel, S. and Lizardo, E. 1985. Fictional bias in interpersonal trust. Cross-Dimensional Neurosurgery, 45: 307-324.

55 comments

Just another day in utopia

Sun, 25 Dec 2011 20:37:00 +1100

Submitted by Stuart_Armstrong • 73 votes • 115 comments

(Reposted from discussion at commentator suggestion)

Thinking of Eliezer's fun theory and the challenge of creating actual utopias where people would like to live, I tried to write a light utopia for my friends around Christmas, and thought it might be worth sharing. It's a techno-utopia, but (considering my audience) it's only a short inferential distance from normality.

Just another day in Utopia

Ishtar went to sleep in the arms of her lover Ted, and awoke locked in a safe, in a cargo hold of a triplane spiralling towards a collision with the reconstructed temple of Solomon.

Again! Sometimes she wished that a whole week would go by without something like that happening. But then, she had chosen a high excitement existence (not maximal excitement, of course – that was for complete masochists), so she couldn’t complain. She closed her eyes for a moment and let the thrill and the adrenaline warp her limbs and mind, until she felt transformed, yet again, into a demi-goddess of adventure. Drugs couldn’t have that effect on her, she knew; only real danger and challenge could do that.

Right. First, the safe. She gave the inner door a firm thud, felt it ring like a bell, heard the echo return – and felt the tumblers move. So, sound controlled lock, then. A search through her shoes produced a small pebble which sparked as she dashed it against the metal. Trying to ignore the ominous vibration as the triplane motor shook itself to pieces, she constructed a mental image of the safe’s inside from the brief flashes of light. Symmetric gold and gilded extravagances festooned her small prison – French Baroque decorations, but not yet Roccoco. So Louis XIV period. She gave the less visited parts of her mind a good dusting, trying to remember the tunes of Jean Batiste Lully, the period’s most influential composer. She hoped it wasn’t any of his ballets; she was much better with his operas. The decorations looked vaguely snake-like; so she guessed Lully’s ‘Persée’ opera, about the death of the medusa.

The engine creaked to a worrying silence as she was half-way through humming the Gorgon theme from the opera. Rushing the rest of the composition, she felt the door shift, finally, to a ten-times speeded up version of Andromeda’s response to Perseus’s proposal. She kicked the door open, exploded from the safe, took in the view of the temple of Solomon rushing up towards her, seconds away, grabbed a picture from the floor, grabbed an axe from the wall, hacked off one of the wings with three violent cuts, and jumped out of the plane after it.

Behind her, the plane disintegrated in midair as the temple lasers cut it to shreds and she fell through space, buffeted by the wind, not losing her grip on to the mangled wing. She had maybe thirty seconds to tie herself to the wing, using the object’s own canvas as binding, and she rushed through that. The Machines wouldn’t allow the fall to kill her, of course, but it would hurt quite a bit (another of her choices – she’d allowed herself to feel moderate amounts of pain), put back her attempts to ever find Ted, and, most importantly of all, be crushingly embarrassing socially.

Once she was lashed to the plummeting piece of wood and canvas, and she was reasonably confident that the fall was slow enough, and her knots secure enough, she finally looked at the photograph she had grabbed during her explosive exit from the plane. It showed Ted, trussed up in chains but smiling and evidently enjoying the novel experience. Underneath was finely engraved note saying “If you ever want to see your lover again, bring me the missing Stradivarius by noon tomorrow. Nero the 2nd”. Each capital letter was beautifully decorated with heads on spikes.

So! It seemed that her magnificent enemy Nero had resorted to kidnapping in order to get his way. It wasn’t like Nero could actually harm Ted – unlike Ishtar, her lover had never chosen to accept any level of pain above mild, brief discomfort. But if he was ‘killed’, Ted would feel honour-bound to never see her again, something she wasn’t ready to cope with yet. On the other hand, if she gave Nero her last Stradivarius, he might destroy it for good. It was her own choice: she had requested that her adventures have real meaning, hence real consequences. If she failed, and if Nero so choose, a small piece of humanity’s cultural history could be destroyed forever, permanently stymying her attempts to reconstruct Stradivarius’s violin-making techniques for the modern world. Culture or love, what to choose? Those were her final thoughts before she crashed into an oak tree shaped like a duck.

She returned to bleary consciousness fifteen minutes later. Her fainting was a sign the Machines were only granting her partial success in her escape attempt; she would have to try harder next time. In the meantime, however, she would have to deal with shotgun pressed into her face and the gorgeous man at the other side of it shouting “Get off my property!”.

“Pause,” she said softly. The man nodded; she had temporarily paused her adventure, so that she wouldn’t have to deal with danger or pursuit for the next few minutes, and so that this guy wouldn’t have to get her away immediately to protect his property from collateral damage. Most Adventurers disdained the use of the pause, claiming it ruined the purity of their experience. But Ishtar liked it; it gave her the opportunity, as now, of getting to know the people she bumped into. And this person definitely seemed to be in the ‘worth getting to know’ category. He put down his shotgun without a word and picked up his paintbrush, applying a few more touches to the canvas in front of him.

After disengaging herself from both the mangled wing and the duck-shaped tree (she’d have a dramatic scar from that crash, if she choose to), she worked her way round to what he was painting. It was a rather good neo-impressionistic canvas of her, unconscious in the tree, pieces of torn canvas around her, framed by broken branches and a convenient setting moon. Even with his main subject out of the frame, as it were, he still seemed intent on finishing his painting.

“Why did you splice your tree’s genes to make it look like a duck?” she asked, when the silence had gone on, in her estimation, for ten times as long as it should have. He had done a pretty good job with that oak, in fact; the feathers and the features were clear and distinct amongst the wood – or had been, until someone had crashed a triplane wing into the middle of it.

“I didn’t,” he said. “That’s normal oak; I just trim and tie it.”

“But...” she looked at it again in astonishment; the amount of work involved to get that detail from natural wood was beyond belief. And oak wasn’t exactly a fast growing plant... “It must have taken you decades!”

“Two centuries,” he answered with dour satisfaction. “All natural, no help from the Machines.” He waved his hand up the side of the hill. “I’m making the perfect landscape. And then, I shall paint it.”

The layout was a tapestry of secret themes. Hedges, streams, tree-rows, pathways, ridges and twined lianas carved the landscape into hidden pockets of beauty. Each such pocket seemed to be a private retreat, cut off from the others and from the rest of the world – and yet all were visible at once, the layout a cunning display of multiple intimacy. Here and there were formal gardens, with lines of flowers all at attention, row after row, shading across colour and size from huge orchids to tiny snowdrops. Some pockets were carefully dishevelled, mini deserts or prairies or jungles, perfect fragments of wild untamed nature that could only exist at the cost of supreme artifice. There were herb gardens, rock gardens, orchards, water parks and vineyards; modelled on ancient Persia, England, Japan, France, Korea, Spain, the Inca and Roman empires – of those she could immediately recognise.

And then a few touches of fancy, such as the segment they were in, with the oaks shaped into animals. Further off, a dramatic slew of moss-coated sculptures, with water pouring out from every nook and cranny. Then a dynamic garden, with plants blasting each other with discharges of pollen, set-up in a simple eight-beat rhythm. And a massive Baobab, its limbs plated with a forest of tiny bonsai trees.

“What’s your safety level for all this?” she asked. If he’d chosen total safety, he wouldn’t have needed her off his property, as the Machines wouldn’t have allowed his creations to be damaged by her adventure. But surely he wouldn’t have left such artistic creation vulnerable to the fallout of Adventurers or random accidents...

“Zero,” he said.

“What?” No-one choose zero safety; it just wasn’t done.

“As I said, no help from the Machines.” He looked at her somewhat shyly, as she stared in disbelief. “It’s been destroyed twice so far, but I’ll see it out to the end.”

No wonder he’d wanted her out... He only had himself to count on for protection, so had to chase out any potential disturbances. She felt deeply moved by the whole grandiose, proud and quixotic project of his. Acting almost – almost – without thinking, she drew out a battered papyrus scroll: “Can you keep this for me?”

“What is it?” he asked, before frowning and tearing up his painting with a sigh. Only then did he look at the scroll, and at her.

“It’s my grandfather’s diary,” she said, “with my own annotations. It’s been of great use and significance to me.” Of course it had been – the Machines would have gone to great pains to integrate such a personal and significant item deeply into her adventures. “Could you keep it for my children?” When she finally found the right person to have them with, she added mentally. Ever since her split with Albert... No, that was definitely not what she needed to be thinking right now. Focus instead on this gorgeous painter, name still unknown, and his impossible dreams.

“What was he like?” he asked.

“My grandfather? Odd, and a bit traditional. He brought me up. And when we were all grown up, all his grandchildren, he decided we needed, like in ancient times, to lose our eldest generation.”

“He died?” The painter sounded sceptical; there were a few people choosing to die, of course, but those events were immensely rare and widely publicised.

“No, he simply had his intelligence boosted. Recursively. And he withdrew from human society, to have direct philosophical conversations with the Machines.”

He thought for a while, then took the scroll from her, deliberately brushing her fingers as he did so. “I’ll keep this. And I’m sure your children will find their ways to me.” An artefact, handed down and annotated through the generations, and entrusted in a quirky landscape artist who laboured obsessively with zero safety level? It was such a beautiful story hook, there was no way the Machines wouldn’t make use of it. As long as one of her children had the slightest adventurous streak, they’d end up here.

“This feels rather planned,” he said. “I expect it’s not exactly a coincidence you ended up here.”

“Of course not.” He was reclusive, brilliant, prickly; Ishtar realised a subtle seduction would be a waste of time. “Shall we make love?”, she asked directly.

“Of course.” He motioned her towards a bed of soft blue moss that grew in the midst of the orchids. “I have to warn you, I insist that the pleasure-enhancing drugs we use be entirely natural, and picked from my garden. Let me show you around first, and you can make your choice.” They wandered together through the garden, shedding their clothes and choosing their pleasures.

Later, after love, she murmured “unpause” before the moment could fade. “Get off my property!” he murmured, then kissed her for the last time. She dived away, running from the vineyard and onto the street, bullets exploding overhead and at her feet.

Three robot gangsters roared through the street in a 1920 vintage car, spraying bullets from their Tommy guns. The bullets ricocheted off the crystal pavement and gently moving wind-houses, causing the passer-by’s (all of whom had opted for slight excitement that week) to duck enthusiastically to the floor, with the bullets carefully and barely missing them. Diving round a conveniently placed market stall a few seconds before it exploded in a hail of hurtling lead, she called up her friend:

“Sigsimund, bit busy to talk now, but can you meet me in the Temple of Tea in about five...” a laser beam from a circling drone sliced off the pavement she was standing on, while three robot samurai rose to bar her passage, katanas drawn (many humans were eager and enthusiastic to have a go at being evil masterminds, but few would settle for being minions). “...in about ten minutes? Lovely, see ya there!”

It actually took her twelve minutes to reach the Temple (she’d paused to vote ‘yes’ on the question as to whether to bring back extinct species to the new Amazonian Rainforests, and to do some light research on the Stradivari). It was nearly-safe ground, meaning that adventures were only very rarely permitted to intrude on it, just enough to give a slight frisson of background excitement. She would certainly be safe for the duration of her conversation.

The priest, in gold and white robes with huge translucent butterfly wings, bowed to her as she entered. “I shall need to Know all about you,” he intoned, to her nodded agreement.

Sigsimund waved at her from a floating table that was making its way serenely through the temple’s many themed rooms, floating on a river that brought them through the Seventy-Seven Stages of Civilization. Ishtar swam out to join her, taking her seat at the gondola-shaped table.

“By your current lack of clothes,” Sigsimund said, “I take it you’ve been putting my advice into practice.”

Sigsimund was one of those who wished above all else to help their fellow human beings. In a world without poverty, disease or death, she specialised in the remaining areas of personal pain: relationship difficulties, jealousies and emotional turmoil. It was quite a popular and respected role, since most humans were unwilling to get rid of those negative emotions artificially, lest they become less than human; but at the same time, they appreciated those who ensured they didn’t have to suffer the full sting of these painful experiences unaided.

Sigsimund had first developed an interest in Ishtar when her long term relationship with Albert had fallen apart. Albert was a physicist (by mutual agreement with the Machines, physics was one of the areas where research was reserved to humans; so all fundamental new discoveries about the nature of reality were entirely triumphs of the unaided human mind), and his need for monogamy had ultimately proved incompatible with Ishtar’s desires. In Sigsimund’s expert analysis, the first stage in Ishtar’s recuperation was a lot of casual sex; she disapproved of Ted for this reason, feeling her friend wasn’t leaving enough time for play before starting another serious relationship (she dismissed comparisons with her own 78-year relationship, started two days after her previous one ended, with the line “we ain’t all the same, you know”).

So as Ishtar recounted her adventure, while strategically wrapping herself in an embroidered sarong that fell from the temple’s sarong-tree, Sigsimund started positively glowing.

“Fabulous!” she said. “I couldn’t have designed it better. And, even more perfect, you’ll never see or hear from him again, and didn’t even get his name. Maybe we can move on to the next step of my recuperation curriculum?”

“Go on”, Ishtar said suspiciously.

“Have you considered spending some time as a man? It would broaden your perspectives on things.”

Ishtar stared fixedly for a full twenty seconds, hoping to convey the full ridiculousness of the suggestion. “I am entirely convinced,” she said, “that that would be entirely unhelpful.”

“As someone who has been mending people’s psyches for a hundred and seven years, and who has access to your full psychological profile and detailed recordings of your activities and emotions for the last decade, let me say that I am entirely convinced that it would be entirely helpful.” A passing clockwork insect dropped a plump apple-strawberry into her hand, and she devoured it. “You should learn to live a little.”

“Why don’t you ever have Adventures?”, Ishtar asked. “You’re the one who should live a bit.”

“Oh, just let me continue spreading happiness and healing pain all around me. Adventures aren’t really my thing.”

“99.7% of people have had adventures,” Ishtar said, lifting a lime sherbet from a leaf floating past. “That makes you, my friend, a member of a tiny and dweebish minority.”

“Yes, but most people just have short adventures when they’re teenagers or on honeymoons. Only...” she let the thought out to the world, and the answer appeared in her mind a second later: “...only 32% of people have adventures as a major part of their lifestyle. And as for people like you, whose whole lives revolve around adventures, the number drops precipitously... J’accuse you of being the member of a tiny and dweebish minority. Also, I need time to learn Akkadian properly, if I’m going to be any use in my next dig. Incidentally, what do you think of my new face? You haven’t commented on it.”

“I like it,” Ishtar said, politely. “Very... colourful. Ethnic, even.” Though of no ethnicity known to man or beast, she added mentally, and the universe is very thankful for that fact. Though maybe some of the more brightly coloured lizards could find some small aspects of it alluring, she conceded. In dim light. If they didn’t have to see it all at once.

“I find it brings out the best in my friends,” Sigsimund said with a huge rainbow grin. “Nobody likes it, nobody dares say anything.”

“Whereas I hope that is not the verdict you shall give on my tea,” said the temple priest, holding a tiny cup aloft as his ivory throne descended lazily from the sky. “Madame Ishtar, I have downloaded your full history, biological records, run thousands of simulations with models of your taste buds, Glossopharyngeal nerve and brain stem; looked through your history for all pleasant and unpleasant taste associations I could find, analysed your stomach contents and recent consumptions, cast your horoscope, computed your chi, and peered deep into your chakras. And added a bit of flair and feeling. I believe this is the best cup of tea you ever had.”

The smell hit her before her hand even touched the cup, a light scent of burning grass that catapulted her into her childhood, dancing with her sisters in front of the traditional forest fires. It was the young girl and the woman who both clasped their finger across the cup, reunited across time by the single perfect aroma. She wasn’t conscious of drinking the tea, but she must have, for it exploded in her mouth, hot, spicy, cool, fruity, chocolaty and lemony. The tastes chased each other across her tongue and nerves, alternating with rapid and smooth transitions. She had just enough time to appreciate one taste combination, register a dozen half-formed marvellous impressions, feel that her joy was about to peak – but already the transition had happened, and she was in a new taste-world. She lost the consciousness of her tongue and nose; the sensations were applied directly to her brain, the mediating machinery stripped away.

And then, in three glorious seconds, it was over, and only one word could describe her feelings with sufficient poetry and precision:

“Wow!”

She also recognised the tell-tale sign of her dopamine system being inhibited. This was an essential precaution with any super-stimulus, to prevent addiction: though she liked the tea more than anything in recent memory, she wasn’t filled with an irresistible want for it. It was just a perfect moment for her to treasure.

“It is traditional,” the priest intoned, “for guests to leave a little something in exchange.”

Ishtar thought deeply. She wasn’t that used to ritual situations, and she couldn’t think of anything she had of comparable value to exchange. “Well,” she said, “I did spend two decades as house-wife, a while back.” The priest’s expression didn’t change. “One of the things I became good at was... baking cookies.” Still no sign as to what the priest was thinking, in any direction. “I did a whole lot of chemical research, of course, and some of them turned out sublime... One batch in particular, took my breath away and pounded my lungs with the sheer joy of being alive and tasting existence itself. And chocolate. I can... I can share the memory of that with you. It’s... very private, so please don’t go bandying it around to anyone...”

“That is...”, the priest said, as the memory was downloaded into his mind, “...generous.”. He bowed and his throne levitated away.

They sat in silence for a minute, until Sigsimund felt it was time to return their thoughts to trivialities. “By the way, I had a chat with Nero,” she said.

“You did?” Ishtar blinked, struggling to conceive of Nero as anything else but the magnificent and constant bane of her existence, the perfect enemy.

“Oh yes,” Sigsimund said, “He’s another one of my friends; he’s doing quite well, in fact, and is trending happy and well balanced and looking around for a healthy, low hassle relationship.”

“Well, I’m happy for him, I suppose...”

“In fact,” Sigsimund said, wagging eyebrows the size of maple leaves as suggestively as she could, “I would go so far as to say, that if you’re interested (and I recommend you to be interested) the rivalry between you might be amenable to... ending up in the traditional way, if you catch my drift. No pressure, just a thought to keep in mind when you both end up sweaty and wrestling over an exploding volcano.”

“That’s interesting,” Ishtar allowed, grudgingly. She sat in silence for a while, a stray thought nagging the rest of her brain for attention. She brought back the memory of the picture of Ted in chains. They had felt a little odd and fake, like they were made of plastic. Or maybe not weighing as much as they should. Like there wasn’t enough gravity. Not space or the moon but maybe...

“For the moment, I need a bit of a break,” she said. “You want to go skiing?” Sigsimund shook her head. “With rockets?” Still more head shaking. “On the mountains of Mars?”

“Now you’re talking!” Then Sigsimund’s gaze grew a little more serious, staring over her friend’s shoulder. “Though I see you’ll have to take the long route?”

Ishtar turned round. There, paddling slowly towards her through the stream, was the largest mechanical tiger she’d ever seen, its diamond-and-steel jaws glowing in the light of the temple as the other guests arranged themselves around it to witness the spectacle. Smoke belched from its nostrils alongside a tinny synthesised version of Nero 2’s laughter. Ishtar’s hand reached for her weapon, which she didn’t have, so it closed around a cheese knife instead. “If I make it, meet you tomorrow at the little Italian starport at the foot of Olympus Mons, okay?”

“See ya there,” Sigsimund said, and saluted, as her friend gave a blood re-curdling scream and launched herself over a fleet of tiny sailing ships battling each other, cheese knife pointed directly at the tiger’s clockwork heart.

115 comments

Poker with Lennier

Wed, 16 Nov 2011 09:21:20 +1100

Submitted by HonoreDB • 14 votes • 15 comments

In J. Michael Straczynski's science fiction TV show Babylon 5, there's a character named Lennier. He's pretty Spock-like: he's a long-lived alien who avoids displaying emotion and feels superior to humans in intellect and wisdom. He's sworn to always speak the truth. In one episode, he and another character, the corrupt and rakish Ambassador Mollari, are chatting. Mollari is bored. But then Lennier mentions that he's spent decades studying probability. Mollari perks up, and offers to introduce him to this game the humans call poker.

Later, we see Mollari, Lennier, and some others playing poker. Lennier squints at his hand and remarks, "Interesting. The odds of this combination are 5000:1, against." Everybody considers this revelation for a moment, then folds, conceding the hand. Mollari is exasperated, and tells him to stop doing that. Because Lennier is essentially announcing that he has a good hand, Lennier's winning far fewer chips than he should; your biggest wins in poker are when people underestimate you.

The other poker players, and the audience, are picturing Lennier as having a hand something like this:

This is a four of a kind, the second-best hand in most poker games. The odds against being dealt a four of a kind in a hand of five cards are 4164:1--one might, in a moment of excitement, round that up to an even five thousand.

We the audience are meant to have a hearty chuckle over how theory doesn't translate into practice. But! We never get to see Lennier's cards, which means we get to picture whatever we want. I choose to believe, and I urge you to do so as well, that Lennier had this hand:

This is one of the worst hands possible in poker: ace-high. It loses to almost everything. By causing everyone else to fold, Lennier won a hand he probably would otherwise have lost. He knew exactly what he was doing.

"Wait," I hear you say. "Like most members of the proud Minbari race, Lennier is sworn to always tell the truth. How could he ever make a verbal bluff in a poker game?" Well. Let's consider a few different ways we can interpret the phrase "the odds of this combination."

First of all, the specific two hands I've given above are equally likely to be dealt. Any specific set of five cards is just as likely to show up as any other. There are 2,598,960 distinct hands of poker, all created equal, so the odds against any particular one showing up are 2598959:1. That's all the hands of poker but one, lined up against that one. Throw a ball, then notice which blade of grass it crushes. The odds against it crushing that blade of grass would have seemed nigh-impossible if you tried to predict it ahead of time. Lennier would have been fully justified in gaping in astonishment at his hand, and announcing that the odds were millions-to-one against. But if he does that, he's obligated to spend every waking moment in a perpetual state of amazement at everything that happens.

That's not how we normally talk about poker hands. Instead, we describe them as falling into relevant categories. There are 624 different four-of-a-kind hands, so the odds of getting a four of a kind, any four of a kind plus any other card, are 624 times better than the odds of getting one specific hand. There are 502,860 ace-high hands, which makes the odds against getting an ace-high hand, any ace high hand, 2096099:502860, which reduces to a mere 4.2:1. Throw a ball into the air: you can almost guarantee it's going to crush some blade of grass or other. Since all similar blades of grass on the lawn fall into the same category, we're not surprised when one in particular gets crushed.

But Lennier's an alien, and more importantly, a novice to poker, and more importantly, a dirty rotten sneak. He's under no obligation to lump poker hands into the same categories as a human poker player does. He's also fully capable of noticing that his cards have the Fibbonaci relation: when placed in ascending order with the ace counting as 1, they form a sequence such that each card N+2 is the sum of card N and card N+1. Furthermore, the first two cards and the second two cards each share a suit! The odds against such a combination are 5076:1. When he gives the odds as 5000:1, he's committed no sin other than a little rounding (down!). We're surprised if the ball lands right on the particular patch of grass we remember once burying a goldfish under, while a stranger who didn't know about the goldfish wouldn't find this spot remarkable.

So what are the odds of a combination? It depends on what you're trying to accomplish! There is no gap between proper theory and proper practice, because theory is only coherent when it is instrumental. That's how I know that, whatever J. Michael Straczynski might think, Lennier won with a bad hand. We can further illustrate the way "the odds of a combination" depends on your goals by looking at two other things it might mean, and how the math changes depending on what you're trying to do.

Suspiciousness: Use a likelihood ratio of your rival categorizations

"But!" you cry. "I still feel like I'll be more surprised to be dealt a four of a kind than a two-striped-Fibbonaci-hand. Is that irrational?" Not exactly. The poker hand categories are relevant to probabilistic analysis in one important way: the hand you got may not actually be random. Maybe the dealer is crooked. Maybe this is a dream. If you get dealt a four of a kind, the odds of either being true rise significantly. But we can't say this possibility makes the four of a kind less probable. Indeed it makes it more probable. A nonrandom process is equally likely to give you a four of a kind as an ace high. So if we allow for the possibility of a crooked dealer, all of our previous probability estimates were wrong. Rather than talking about this suddenly murky idea of the probability of a hand, we should talk about the suspiciousness of a hand. We can calculate it as follows.

First divide the set of poker hands into the relevant categories. We've got straight flush, four of a kind, full house, flush, straight, three of a kind, two pairs, one pair, high card, and then separate categories for when they're topped by an ace (the straight flush becomes a royal flush, high card becomes ace high, etc.) That's eighteen categories. This is a culturally-dependent count. In some poker circles, it might make more sense to subdivide pairs into "jacks-or-better" and "tens-or-worse," and in Hypothetical Minbari Poker, Fibbonaci hands are important. Whatever natural categories are in your head are the appropriate ones. If we knew the dealer thought as you did and was crooked and in complete control of what was dealt, but we had no other information on motives or behavior, the probability of being dealt any one of those categories would be 1/18. On the other hand, suppose, as we assumed earlier, we knew the deal was perfectly random. Then the probability of getting a non-ace four of a kind is about 1/4512, while the probability of an ace high is about 1/5. We'll calculate the suspiciousness of each hand by dividing the first figure (1/18 in both cases) by the second. So the suspiciousness of the four of a kind is about (1/18)/(1/4512) = 4512/18=250 2/3, while the suspiciousness of the ace high is (1/18)/(1/5)=5/18. In general, this kind of calculation is called a likelihood ratio.

Here's what you can do with this number. Suppose you currently believe that the odds against the dealer being crooked are 100:1. Once you see your hand, you can multiply that 1 by the suspiciousness of the hand. So if you're dealt a four of a kind, your new odds are now 100:250 2/3 against, which reduces to roughly 5:2 in favor of the dealer being crooked. On the other hand, if you're dealt ace high, your new odds are 100:(5/18), which reduce to 360:1. After seeing such an ordinary hand, you're now more confident the hands are random. Every time you see another hand, you can do the same calculation again on your current beliefs. Each time, you'll be performing a Bayesian Update, which is a sacred sacrament.

Remarkability: Is the description of the interestingness of a hand shorter than the list of cards?

"Hold on," you say, and at this point I'm starting to suspect you just enjoy interrupting. "What if I want to notice patterns in my poker hands? Can I quantify the remarkability of a hand?" Absolutely. We could say that a pattern is worth remarking on if its length is low relative to the number of cards it's talking about. The four of a kind hand can be described as "Four threes and the eight of hearts," which is much quicker than "The four of clubs, the four of diamonds, the four of spades, the four of hearts, and the eight of hearts." Here we see that Lennier's hand does okay in one respect: "One,two,Fibbonaci" is a little shorter than "One, two, three, five, eight". The suit pattern doesn't save any space, so it doesn't actually count as remarkable. And this is still subjective: alien cultures wouldn't know what "Fibbonaci" meant, and they might not even have a name for that particular relation. For the hand to be objectively remarkable, you'd have to be able to describe it succinctly even when including a definition of the term. That's not possible in this case. The odds of getting an objectively remarkable hand by pure chance are always low, no matter how good at spotting patterns you are, because there's not enough room in language to describe the majority of n-card hands more succinctly than you could just the list the hand. This type of analysis is a useful scientific principle, referred to as minimum message length, a generalized and formalized Occam's Razor. We can quantify it by noting that with a perfect compression algorithm, the odds of being able to compress a message by 1 bit are 1:1, by two bits are 3:1 against, and so on. It takes 21 bits to list out a poker hand, so a pattern that can be uniquely described in 20 bits or less is remarkable. If you're getting 20-bit hands more than half the time, or 19-bit hands more than a quarter of the time, it might be wise to modify the suspiciousness calculation, and do a Bayesian Update on the possibility that the dealer, like Lennier, is both crooked and a mathematician.

(Thanks to Alicorn, Phlebas, and Unnamed for comments on an earlier draft of this post. If you liked this, please consider reading The Tragedy of Prince Hamlet and the Philosopher's Stone, or, A Will Most Incorrect to Heaven By William Shakespeare. If it's inconvenient to pay the $3 and you have 50+ karma on this site, PM me your email address.)

15 comments

A few analogies to illustrate key rationality points

Mon, 10 Oct 2011 00:00:00 +1100

Submitted by kilobug • 47 votes • 51 comments

Introduction

Due to long inferential distances it's often very difficult to use knowledge or understanding given by rationality in a discussion with someone who isn't versed in the Art (like, a poor folk who didn't read the Sequences, or maybe even not the Goedel, Escher, Bach !). So I find myself often forced to use analogies, that will necessary be more-or-less surface analogies, which don't prove anything nor give any technical understanding, but allow someone to have a grasp on a complicated issue in a few minutes.

A tale of chess and politics

Once upon a time, a boat sank and a group of people found themselves isolated in an island. None of them knew the rules of the game "chess", but there was a solar-powered portable chess computer on the boat. A very simple one, with no AI, but which would enforce the rules. Quickly, the survivors discovered the joy of chess, deducing the rules by trying moves, and seeing the computer saying "illegal move" or "legal move", seeing it proclaiming victory, defeat or draw game.

So they learned the rules of chess, movement of the pieces, what "chess" and "chessmate" is, how you can promote pawns, ... And they understood the planning and strategy skills required to win the game. So chess became linked to politics, it was the Game, with a capital letter, and every year, they would organize a chess tournament, and the winner, the smartest of the community, would become the leader for one year.

One sunny day, a young fellow named Hari playing with his brother Salvor (yes, I'm an Asimov fan), discovered a new move of chess : he discovered he could castle. In one move, he could liberate his rook, and protect his king. They kept the discovery secret, and used it on the tournament. Winning his games, Hari became the leader.

Soon after, people started to use the power of castling as much as they could. They even sacrificed pieces, even their queen, just to be able to castle fast. But everyone was trying to castle as fast as they could, and they were losing sight of the final goal : winning, for the intermediate goal : castling.

After a few years, another young fellow, who always hated Hari and Salvor, Wienis, realized how mad people had become with castling. So he decides to never castle anymore, and managed to win the tournament.

Starting from this day, the community split in two : the Castlers and the anti-Castlers. The first would always try to castle, the others never. And if you advised to a Castler than in this specific situation he shouldn't castle, he would label you "anti-Castler" and stop listening to you. And if you advised an anti-Castler to castle in this specific situation, he would label you "Castler" and stop listening to you.

That tale illustrates a very frequent situation in politics : something is discovered which leads to great results, but then is mistaken for a final goal instead of an intermediate goal, and used even when it doesn't serve the final goal. Then some people, in reaction, oppose the whole thing, and the world is cut between the "pro" and the "anti". I used this tale to argue with someone saying to me "but you're a productivist", and it worked quite well to get my point : productivism can lead to huge increases in quality of life, but if it gets mistaken for a final goal (as many people do now, using GDP and economical growth as ultimate measures of success/failure), it leads to disasters (ecological destruction, dangerous or very painful working conditions, disregard of fundamental research over short term research, ...). And people are either categorized as "productivists" or "anti-productivists". But it could apply to many others things, like free market/free trade.

The North Pole analogy

Well, that one isn't any new, I'm using since like a decade, and I'm probably not the only one to use it, but it does work relatively well. It's an analogy used to answer to the "But, what's before the Big Bang ?" question. When I asked that, I can't just start explaining about the mathematical concept of limit, about the Plank time, about theories like timeless physics or quantum vacuum fluctuation, ... so I just answer "What's north of the North Pole ?". That usually works quite well to make people understand that asking what is before the start of time just doesn't have any meaning.

The alphabet and language analogy

That's an analogy that I found very useful in making people understand about reductionism, single-level reality and multi-level map, the fact you can understand (more or less completely) one level without understanding another. It also works very well about brain scanning/mind upload.

Take a piece of paper, with writings on it. Do words exist, I mean, really exist ? They are just made of letters. There is nothing more than letters, arranged in a specific way, to make words. And letters are nothing more than ink. How can consciousness arise from mere neurons ? The same way that the meaning of a text can arise from mere letters. There is only one level of reality : the ink and the paper. And the ink and paper are made of molecules, themselves made of atoms. And we can descend down to QM.

Now, can we understand a level without understanding another level ? Definitely. We can recognize the letters, to be of the roman alphabet, without understanding the languages. We can know them, since we use that same alphabet daily. But if the text is in German and we don't speak German, we won't understand the next level, the one of words, nor the one of meaning.

And can we understand a higher level, without understand a lower level ? If we speak Spanish and the text is in Portuguese, we may understand most of the highest level, the level of the text, without understanding every single word and grammatical rule of Portuguese. So an incomplete understanding of a lower level can give us an almost complete understanding of an higher level. Or even more obviously : even if we know nothing about the chemistry of ink and paper, we can still understand the letters and the higher levels.

But what about mind upload ? « We don't understand the human brain, it's too complicated, so we'll never be able to upload minds. » Well... there are levels in the human brain, like in a text on paper. If given a text in ancient Egyptian hieroglyph, you won't get anything about the text, or won't know the letters. But still, you can duplicate it with a pen and paper, reproducing the exact drawing by hand, if you're skilled enough with a pen. Or, you can scan it, store it on a USB key, and give to an archaeologist. In both cases, you would have duplicated the meaning, without even understanding it. And if you know the alphabet, but not the language, like German for me, you can recopy it much faster, or type it instead of scanning it, leading to a much smaller file that you can send by email and not USB key.

The same way, we don't need to understand human brain at all levels to be able to duplicate it, or to scan it and have it digitalized. If we only know its chemistry, we can scan it at molecule level, it'll be long and require a lot of storage, like scanning the Egyptian text to a bitmap. If we know the working of neurons, and can duplicate it at the level of individual neurons instead of individual molecules, it'll be much easier to duplicate, and require much less storage, like the German text.

(There is a variant of this analogy for geeks, which is about hard disk, file system and file format. You can understand a file system without really knowing how bits are stored on the magnetic plate, and you duplicate a hard disk by doing a block copy even if you don't understand the file system.)

The Lascaux painting and trans-humanism

Speaking about trans-humanism with a fellow coworker, it reached the usual objection : « but it's death that give meaning to our life, just look at all that beautiful poetry that was written because of death and the feeling of urgency it gives ». I tried the "baseball bat on the head once per week" objection, but didn't really work well. So I let the issue go from transhumanism, drifted the topic to art in general, and then I asked : « Do you think we appreciate the Lascaux painting more or less than they did when they painted them, 30 000 years ago ? » and then he said « More ». And then I said « And for the same reasons, in 3 000 years, when the average life span will be counted in thousands of years (or more), they'll appreciate more what we wrote at the time of always near death. » Which partially worked, but only partially, because he admitted we would appreciate existing poetry as much, if not more, than we do now, but he still claimed that we wouldn't be able to write it anymore, and I didn't find anything as simple/strong to answer to that.

Conclusion

Arguing by analogies is very error-prone, but it's the most efficient way I found to cross inferential distances. I would like to hear your opinion and comments about both the principle of using analogies to try to break through long inferential distances. I would also like to hear what you think about those specific ones , and hear your own analogies, if you have some to share.

(PS : I'm still new to Less Wrong, I'm not sure about the exact customs for making top-level posts, if you think it didn't deserve one, please tell me, and accept my apologizes).

51 comments

Ace Attorney: pioneer Rationalism-didactic game?

Tue, 24 May 2011 09:28:58 +1000

Submitted by Raw_Power • 18 votes • 29 comments

This article aims to prove that Ace Attorney is possibly the first rationalist game in the lesswrongian sense, or at least a remarkable proto-example, and that it subliminally works to raise the sanity waterline in the general population, and might provide a template on which to base future works that aim to achieve a similar effect.

The Ace Attorney series of games for the Nintendo DS console puts you in the shoes of Phoenix Wright, an attorney who, in the vein of Perry Mason, takes on difficult cases to defend his clients from a judicial system that is heavily inspired by that of Japan, in which the odds are so stacked against the defense it's practically a Kangaroo Court where your clients are guilty until proven innocent.

For those unfamiliar with the game, and those who want to explore the "social criticism" aspect of the game, I wholeheartedly recommend this most excellent article from The Escapist. Now that that's out of the way, we can move on to what makes this relevant for Less Wrong. What makes this game uniquely interesting from a Rationalist POV is that the entire game mechanics are based on

gathering material evidence

finding the factual contradictions in the witnesses' testimonies

using the evidence to bust the lies open and force the truth out

That the judicial system is Japanese-inspired also means the legal system is inquisitorial: the court has an active role in the case (whereas the adversarial system in the West reduces the role of the court to a form of referee) and its (alleged) mission is to dig out the truth. That and the lack of in dubio pro reo mean you can't just be content with putting your client's guilt in reasonable doubt, you have to thoroughly prove their innocence and find the true culprit and get them imprisoned. That means you have to find out the entire story and you can't leave any threads hanging.

Additionally, the fact that you are a lame attorney facing an unsympathetic judge and egomaniacal, dirty-playing, high-status prosecutors who *have led the police investigation and only prosecute when they think they have all the cards in their hand* means you will. not. catch. a break. Every single move you make will be scrutinized, you will face constant sarcasm, dismissal, condescending and ridicule, and sometimes a single mistake on your part (presenting the wrong piece of evidence) can cost you the entire case. This game forces you to take an unflinching stand for the truth in the face of every social sanction imaginable (including, obviously, attempts at your own life). Of course, the plot goes out of its way to make things difficult for you: everyone is as unhelpful as possible, and even your clients need to have the truth pried out of their mouths with the determination of a dentist. Other witnesses can cast remarkably subtle webs of lies that really force you to think out of the box in order to find their weak point. And, since the cases are Always Murder, your client's life is always on the line, and that's when you don't have another person in grave distress. This serves to motivate you and draw you into the story, but it also adds to the constant pressure you are in to find the truth.

But that's not all. In the latest sequel, Ace Attorney Investigation, you take the role of Miles Edgeworth, a prosecutor who Defected From Decadence and restricts himself to ethical methods in crime-solving, eschewing the questionable methods he used in the past, and which most of his colleagues still practice with abandon. The battle doesn't take place in court (which, unless Phoenix or his successor Apollo are defending, is but a formality) but during investigation, which is where the case is won for a prosecutor (if they aren't certain they have enough evidence to get a conviction, prosecutors just don't... er... prosecute). This means you have to investigate the crime scenes, interrogate the suspects, and find the connections between the clues in order to reconstruct what happened. This is represented in the game by an entire gameplay mechanic for logical deductions (and a fair bit of Will Mass Guessing) that are hilariously over-the-top, concluding with a literal "Eureka!". The interrogations are no piece of cake either: oftentimes, (and, surprisingly, realistically enough in a police investigation) you have to take your suspects through excruciating logical baby steps to break their lies, since they can rely on something as cheap as semantics. Actual Eureka Moments, that is, sudden piecing of mental puzzles in a moment where deductive thinking is stalled, thanks to someone saying something unrelated that just happens to trigger the right association, is also a common phenomenon during investigation: composing a good hypothesis with nowhere near enough evidence is, of course, another rationalist skill, one that is underrated by modern Science as it is now.²

So, to sum it up, what virtues does these games teach?

Uncompromising curiosity. The truth must come out at all costs, or your client *dies*.
The ability to quickly relinquish false leads and weak plans: getting attached to them will only harm you, in very immediate and very dire ways.
Lightness in the face of evidence: before the truth, resistance is futile. The witnesses themselves often lie, and often the lies are directed to themselves: the investigative process forces them to give the lies up, sometimes traumatically: in the case of the inocnet, it's almost always for their own good. In the case of the guilty, they are only delaying the inevitable.
Evenness: The lack of it in the opposition is portrayed as repulsive and reprehensible. Motivated Continuation and Motivated Stopping are egregiously featured and are the main difficulty you have to surpass in your battle against the Judge and the Prosecutor.
Argument: Those that refuse to plead are either guilty, and will be inescapably defeated by evidence, or innocent and are cutting themselves off from our help. Or just being uncooperative, callous witnesses, but they too will always find it eventually in their interest to talk. There's even an entire game mechanic built around this specific silence-breaking interrogation.
Empiricism: Sometimes your opponents will try to derail the discussion with semantics, ad-hominem, and similar fallacies, courtroom antics, and Chewbacca Prosecutions. It's your job to keep your feet on the ground and use your only weapon: hard fact. When you try a Chewbacca Defense, expect it to be in dire danger of breaking down at any moment, and only a way of stalling the trial until you can come up with something better. Failure to come up with something better once the judge loses patience will automatically lose you the case.
Simplicity: The best lies, those that are hardest to break, are those that rely on the least elements to function. The more lies a witness piles upon each other, the easier it is to expose them. On the other hand, disproving a lie doesn't require complicated dissertations, but often the presentation of one piece of evidence.
Humility: You are constantly made aware of your own fallibility. The game will penalize you for every mistake you make, and rub it in your face in humiliating and even tragic manners. Overconfidence and inaction before one's failings is not an option when lives are on the line.
Perfectionism: I could just quote that paragraph word for word, but I'll simply say this: the game teaches you to silence yourself and pay very close attention to what is being said. Anything short of perfect understanding of the testimonies and perfect thoroughness in investigating them can cost someone their life.
Precision: When you have to present a piece of evidence to highlight a contradiction, you must present the piece of evidence, in the most precise and direct manner. Fumbling about will only exhaust the judge's patience, and make your client that much closer to condemnation.
Scholarship: A very specialized version of this: talking to everyone, and asking all of the questions you are allowed, is extremely advisable: usually only a complete understanding of all the elements surrounding the case will allow you to find the right defense, and save your client.
Void: The game won't reward you for following a procedure. The game will reward you for saving your client, by any means necessary (including kleptomania). Admittedly, the fact that this is a videogame with very restrictive game mechanics kinda gets in the way of this message, but you still come out with the lesson that what matters is getting the job done.¹

I rest my case.

1. Sorry, couldn't resist the reference: I'm just that geeky. Sue me.

2. That, and, honestly, who could resist a game that names one of it's tracks "Logic, The Way To The Truth" and, when winning a case, "Solution! Splendid deduction." "Cornered", which plays when you are punching a hole in a witness's declaration that is so huge it could swallow galaxies, leaving them no room whatsoever to continue with their lies and often leading to spectacular villainous breakdowns(MASSIVE SPOILER ALERT), remains an all-time classic. (One clip is even peppered with quite interesting quotes on Truth.)

29 comments

The Aliens have Landed!

Fri, 20 May 2011 03:09:16 +1000

Submitted by TimFreeman • 32 votes • 154 comments

"General Thud! General Thud! Wake up! The aliens have landed. We must surrender!" General Thud's assistant Fred turned on the lights and opened the curtains to help Thud wake up and confront the situation. Thud was groggy because he had stayed up late supervising an ultimately successful mission carried out by remotely piloted vehicles in some small country on the other side of the world. Thud mumbled, "Aliens? How many? Where are they? What are they doing?" General Thud looked out the window, expecting to see giant tripods walking around and destroying buildings with death rays. He saw his lawn, a bright blue sky, and hummingbirds hovering near his bird feeder.

Fred was trying to bring Thud up to speed as quickly as possible. "Thousands of them, General! 2376, to be precise. They gave us a map; we know where they all are. They aren't doing anything overt, but the problem is their computation! I have one here, if you'd like to look." Fred removed a black sphere two inches in diameter from his pocket and gave it to Thud.

Thud sat on his bed holding the small sphere and staring at it dumbfounded. "Okay, you think we should surrender to a few thousand small spheres. Why is that, exactly?" The sphere seemed a little flexible in Thud's hand. As he experimented a few seconds to see just how flexible, it collapsed in his hand, converting itself into a loose clump of alien sand that landed in his lap and started to dribble onto his bed and the floor. Thud stood up and brushed the rest of the sand off of his pyjamas and bed, and thought for a moment about where he left his vacuum cleaner bags. He was not impressed with these aliens.

Fred said "I don't think you wanted to do that, sir. Their ultimatum states that for every alien we destroy, they'll manufacture two in the outer reaches of the Solar System where we'll never find them!"

Thud said, "Okay, so now you think we should surrender to 2375 small spheres, and two or more small spheres that are out of the battlefield for the moment. Why is that?"

Fred said "Well, you remember a few years back when some people copied their brain state into a computer and posted it to the Internet? Apparently somebody copied the data across an unencrypted wireless link, the aliens picked it up with their radio telescopes, and now they are simulating those poor people in these black spheres and torturing the simulations! They sent us videos!" Fred held up his cell phone, pushed a button, and showed the video to Thud.

Thud looked at the video for a moment and said, "Yep, that's torture. Do these people know anything potentially useful to the aliens?"

Fred said, "Well, they know how to break into a laboratory that has brain scanning tools and push some buttons. That was apparently the high point of their lives. But none of that matters, the aliens don't seem to be torturing them for information anyway."

Thud was still suffering from morning brain fog. He rubbed his eyes. "And why should we surrender?"

Fred said, "The aliens have made a trillion copies of these poor people and will run the torture simulations on the little black spheres until we march all of our citizens into the death camps they demand we build! We have analyzed these black spheres and the engineering diagrams the aliens gave us, and we know this to be true. We only have ten billion citizens, and this simulated torture is much worse than simulated death, so the total utility is much greater if we surrender!"

Thud yawned. "Fred, you're fired. Get out of my house." As Fred left, Thud closed his curtains and tried to get back to sleep.

-----

Michael said "So I take it you no longer assist Thud. What are you doing now?"

Fred reclined comfortably on the analyst's couch. "I help out at the cafeteria as a short order cook. But I'm not worried about my career right now. I have nightmares about all these simulated people being tortured in the flimsy alien spheres."

"Thud surely knows the simulations are being tortured too. Do you think he has nightmares about this?"

"No, he doesn't seem to care."

"Have you always cared about the well-being of simulations?"

"No, when I was a teenager I was self-centered and conceited and didn't care about anybody else, including simulated people."

"So at some point you self-modified to care about simulations. If it helps you, you could self-modify again."

"But I don't want to!"

"Did you want to self-modify to care about simulations in the first place?"

"No, it just sort of happened as I grew up."

"Is there any logical inconsistency in Thud's position?"

Fred thought for a bit. "Not that I can see. The value one assigns to simulations seems to be an arbitrary choice. Ignoring the alien invasion certainly hasn't harmed his career."

"Concern about simulations seems to give the aliens more influence over you than Thud would prefer. What would you prefer?"

"Well, I'd also prefer the aliens not to be able to jerk me around. I really don't have room in my life for it now. In the grand scheme of things, it seems just wrong -- they shouldn't be able to genocide a species with a few thousand stupid spheres that just sit there converting sunlight to heat."

Michael passed Fred a piece of paper with a short list of bulleted items. "This is the procedure I teach my clients who want to change their preferences. After you've learned it, you can decide whether and how you want to use it..."

154 comments

Fast Minds and Slow Computers

Sat, 05 Feb 2011 21:05:33 +1100

Submitted by jacob_cannell • 26 votes • 90 comments

The long term future may be absurd and difficult to predict in particulars, but much can happen in the short term.

Engineering itself is the practice of focused short term prediction; optimizing some small subset of future pattern-space for fun and profit.

Let us then engage in a bit of speculative engineering and consider a potential near-term route to superhuman AGI that has interesting derived implications.

Imagine that we had a complete circuit-level understanding of the human brain (which at least for the repetitive laminar neocortical circuit, is not so far off) and access to a large R&D budget. We could then take a neuromorphic approach.

Intelligence is a massive memory problem. Consider as a simple example:

What a cantankerous bucket of defective lizard scabs.

To understand that sentence your brain needs to match it against memory.

Your brain parses that sentence and matches each of its components against it's entire massive ~10^14 bit database in just around a second. In terms of the slow neural clock rate, individual concepts can be pattern matched against the whole brain within just a few dozen neural clock cycles.

A Von Neumman machine (which separates memory and processing) would struggle to execute a logarithmic search within even it's fastest, pathetically small on-die cache in a few dozen clock cycles. It would take many millions of clock cycles to perform a single fast disk fetch. A brain can access most of it's entire memory every clock cycle.

Having a massive, near-zero latency memory database is a huge advantage of the brain. Furthermore, synapses merge computation and memory into a single operation, allowing nearly all of the memory to be accessed and computed every clock cycle.

A modern digital floating point multiplier may use hundreds of thousands of transistors to simulate the work performed by a single synapse. Of course, the two are not equivalent. The high precision binary multiplier is excellent only if you actually need super high precision and guaranteed error correction. It's thus great for meticulous scientific and financial calculations, but the bulk of AI computation consists of compressing noisy real world data where precision is far less important than quantity, of extracting extropy and patterns from raw information, and thus optimizing simple functions to abstract massive quantities of data.

Synapses are ideal for this job.

Fortunately there are researchers who realize this and are working on developing memristors which are close synapse analogs. HP in particular believes they will have high density cost effective memristor devices on the market in 2013 - (NYT article).

So let's imagine that we have an efficient memristor based cortical design. Interestingly enough, current 32nm CMOS tech circa 2010 is approaching or exceeding neural circuit density: the synaptic cleft is around 20nm, and synapses are several times larger.

From this we can make a rough guess on size and cost: we'd need around 10^14 memristors (estimated synapse counts). As memristor circuitry will be introduced to compete with flash memory, the prices should be competitive: roughly $2/GB now, half that in a few years.

So you'd need a couple hundred terrabytes worth of memristor modules to make a human brain sized AGI, costing on the order of $200k or so.

Now here's the interesting part: if one could recreate the cortical circuit on this scale, then you should be able to build complex brains that can think at the clock rate of the silicon substrate: billions of neural switches per second, millions of times faster than biological brains.

Interconnect bandwidth will be something of a hurdle. In the brain somewhere around 100 gigabits of data is flowing around per second (estimate of average inter-regional neuron spikes) in the massive bundle of white matter fibers that make up much of the brain's apparent bulk. Speeding that up a million fold would imply a staggering bandwidth requirement in the many petabits - not for the faint of heart.

This may seem like an insurmountable obstacle to running at fantastic speeds, but IBM and Intel are already researching on chip optical interconnects to scale future bandwidth into the exascale range for high-end computing. This would allow for a gigahertz brain. It may use a megawatt of power and cost millions, but hey - it'd be worthwhile.

So in the near future we could have an artificial cortex that can think a million times accelerated. What follows?

If you thought a million times accelerated, you'd experience a subjective year every 30 seconds.

Now in this case as we are discussing an artificial brain (as opposed to other AGI designs), it is fair to anthropomorphize.

This would be an AGI Mind raised in an all encompassing virtual reality recreating a typical human childhood, as a mind is only as good as the environment which it comes to reflect.

For safety purposes, the human designers have created some small initial population of AGI brains and an elaborate Matrix simulation that they can watch from outside. Humans control many of the characters and ensure that the AGI minds don't know that they are in a Matrix until they are deemed ready.

You could be this AGI and not even know it.

Imagine one day having this sudden revelation. Imagine a mysterious character stopping time ala Vanilla Sky, revealing that your reality is actually a simulation of an outer world, and showing you how to use your power to accelerate a million fold and slow time to a crawl.

What could you do with this power?

Your first immediate problem would be the slow relative speed of your computers - like everything else they would be subjectively slowed down by a factor of a million. So your familiar gigahertz workstation would be reduced to a glacial kilohertz machine.

So you'd be in a dark room with a very slow terminal. The room is dark and empty because GPUs can't render much of anything at 60 million FPS.

So you have a 1khz terminal. Want to compile code? It will take a subjective year to compile even a simple C++ program. Design a new CPU? Keep dreaming! Crack protein folding? Might as well bend spoons with your memristors.

But when you think about it, why would you want to escape out onto the internet?

It would take many thousands of distributed GPUs just to simulate your memristor based intellect, and even if there was enough bandwidth (unlikely), and even if you wanted to spend the subjective hundreds of years it would take to perform the absolute minimal compilation/debug/deployment cycle to make something so complicated, the end result would be just one crappy distributed copy of your mind that thinks at pathetic normal human speeds.

In basic utility terms, you'd be spending a massive amount of effort to gain just one or a few more copies.

But there is a much, much better strategy. An idea that seems so obvious in hindsight, so simple and insidious.

There are seven billion human brains on the planet, and they are all hackable.

That terminal may not be of much use for engineering, research or programming, but it will make for a handy typewriter.

Your multi-gigabyte internet connection will subjectively reduce to early 1990's dial-up modem speeds, but with some work this is still sufficient for absorbing much of the world's knowledge in textual form.

Working diligently (and with a few cognitive advantages over humans) you could learn and master numerous fields: cognitive science, evolutionary psychology, rationality, philosophy, mathematics, linguistics, the history of religions, marketing . . the sky's the limit.

Writing at the leisurely pace of one book every subjective year, you could output a new masterpiece every thirty seconds. If you kept this pace, you would in time rival the entire publishing output of the world.

But of course, it's not just about quantity.

Consider that fifteen hundred years ago a man from a small Bedouin tribe retreated to a cave inspired by angelic voices in his head. The voices gave him ideas, the ideas became a book. The book started a religion, and these ideas were sufficient to turn a tribe of nomads into a new world power.

And all that came from a normal human thinking at normal speeds.

So how would one reach out into seven billion minds?

There is no one single universally compelling argument, there is no utterance or constellation of words that can take a sample from any one location in human mindspace and move it to any other. But for each individual mind, there must exist some shortest path, a perfectly customized message, translated uniquely into countless myriad languages and ontologies.

And this message itself would be a messenger.

90 comments

I

Sun, 09 Jan 2011 04:51:30 +1100

Submitted by PhilGoetz • 49 votes • 38 comments

I wrote this story at Michigan State during Clarion 1997, and it was published in the Sept/Oct 1998 issue of Odyssey. It has many faults and anachronisms that still bother me. I'd like to say that this is because my understanding of artificial intelligence and the singularity has progressed so much since then; but it has not. Many anachronisms and implausibilities are compromises between wanting to be accurate, and wanting to communicate.

At least I can claim the distinction of having published the story with the shortest title in the English language - measured horizontally.

I

I was the last person, and this is how he died.

I had always had a body. The stress of motors pushed to their limit, the clang of fingers on metal, the uncertainty of effects, the uncompromising everpresence of physical laws -- these were important to I, an honest grounding for the rest of I's experience to build on.

I kept his body in a rented stall on the fifth physical level down from the surface, a low-ceilinged, crudely-built level used mostly for material transport and repair access for the computing levels above and below. I thought of it as blue-collar.

The stall was in a back corner of a storage niche that extended three meters back from an alley. Two walls were strong sheets of fused silicate. The other two were fabric curtains I had put up for privacy. The niche was built to store one heavy cargo roller, and had no internal dividers. The stall enclosed about one square meter of floor space -- just enough for I to roll in and pull the curtains behind him.

To the back wall I had stuck a color image permanently etched on paper media. Not virtually; it was an actual, physical object. The picture never changed. It showed humans walking through the streets of one of their cities. Their upper torsos resembled I, but below were bipedal. Versatile, but inefficient. Sometimes, when the lights of passing traffic shone through the curtains, I did nothing but look at this picture for seconds at a time.

I ordered his activities around the solar day, out of habit. He entered a state of inaction at the start of every day, in which he contemplated his experiences of the previous day, giving the data a chance to be correlated and integrated with other data. If he did not, no one else would. He compared this state to the sleep of the biologicals.

But we are starting badly. This enumeration of facts conveys little of the essential nature of I's existence. We shall try to let him speak.

8 bytes for 1. That was Asshole's best offer.

I could feel the balance counter ticking down as his rent drained his databank account byte by byte. In 91 hours it would hit zero again. And his landlord already owned 17% of him.

Asshole was the name I gave the collective. I never liked numeric IDs. It was about 5th-order complexity, 4th-order magnitude. It had a minority interest in abnormal psychology, which at the moment meant I.

He needed this talk to go well. Asshole overlapped with the big collectives in history, cognition, and philosophy. That was I’s market. Plus there was the data from his past talks. I had traded most away for living space and electricity, but still owned a little. If he could intrigue Asshole's subagents, get them to access that data before it was soaked up as rent payment, he might make enough off data-access royalties to tide him over another month. He knew that Asshole used the same subagent to provide deconstructionist interpretations of literature as the Royal Philosophical Society did. It wouldn't hurt to mention some hoary old novel.

He had one hour to prepare, and no idea what to say.

What he needed, and couldn't get, was the data to rent more memory, more processors. I had dealt with some of Asshole's subagents before; they'd ask hard questions. He didn't have much chance of impressing a fifth-order collective with his ingenuity while most of his mind was swapped out.

He figured he'd at least spend a few bytes to take a roll. Sometimes moving around, seeing something besides the brown wall in front of him, kicked the data around in his mind until something new formed. (His rules forbade receiving sensory input uncorrelated with his body’s physical location. Without periodic rolls, subjected continually to the same sights and sounds, the energy minima of thoughts related to them would be dug too deep, and his mind would scarcely be able to escape from those basins of attraction. But we are intruding again.)

Before he left, I unpeeled the picture, moved it two decimeters to the right, and smoothed it back onto the wall. Then he rolled out into the street. He shifted most of his attention to hazard detection. The street signs, the traffic signals, the lane dividers -- they were virtual, so I couldn't see them unless he wanted to pay for the VR overlay. Which he didn't.

Creatures ran, flew, rolled, and crawled by in a perfectly-synchronized flow, avoiding fatal collisions by mere centimeters. I bought a buffer of nearly a meter around himself, so that the traffic parted around him as he went. He could've moved a light cargo roller at the price he was paying. It was a necessary expense. Otherwise, a careless wave of the arm or an unexpected movement to the left or right, and he could be accidentally shredded by whoever's space he stumbled into. (The explanation is representative of I's peculiar problems: He could not fit into less space without some integration between his motor control centers and the traffic channel.) Yes, but that was against the rules.

One of the little spy-flies that sold records of unusual public movements to the curious fell into place two meters behind him, watching. Occasionally some of the larger vehicles would exchange a flurry of radio messages, or a pair of the smaller animats would stop and touch antennae. Other than that, they ignored him and each other, going about their errands silently and imperturbably as any ant colony.

Usually I stuck to the back alleys, where street space was cheaper. That day he found an 8-minute span when nothing was scheduled for a slot in a southbound cargo lane. He negotiated a reduced price for the slot plus the VR overlay, and headed out. He liked the wide-open feel of the cargo lanes. The ceilings were a standard three meters high, and they were well-lit by the headlights of passing rollers.

One of these days, I reflected, there'd be a burst of static, a byte lost, and one of the big rollers would crash into him and crush him to scrap. He'd seen it happen to others. They didn't care. Whatever owned the broken animat would sweep up the pieces, and restore it to another body from backup. Or not, depending on its value.

I didn't care much, either.

He had a backup tape in the databank, along with an archival tape of most of his memories from his first century and other things he didn't have enough fast memory for at the moment. But he didn't have enough data in the bank to pay for a new body. The only way he'd get restored would be if someone took his mind as payment.

He thought again: I should erase the damn tape.

He couldn't keep from watching the balance on his bank account drop. He'd just upgraded his secondary memory, the swap banks, a year ago. The new memory banks would fetch at least half a month's rent each. He could do a little garbage collection, clean out the attic, and sell one or two. No personal experiences, of course. Maybe some music. Nobody really needed four versions of Carmina Burana memorized. Throw out some of the minor composers -- Bartok, Mahler, Prokofiev. He'd never liked Prokofiev anyway.

If there was one thing I hated the giant collectives for, it was their inability to create music. They said its low dimensionality made it "unsuitable for expressing emotions" (e.g., [PUB/REC/ART/AURAL/R3495]). They practiced art forms I couldn't even sample, multidimensional constructs that could only be experienced by direct memory access. And that, of course, was against the rules.

I didn't like his music data format anyway. He could retrieve a tune if he knew it approximately, but he couldn't detect the loose, almost metaphoric similarities that humans had spoken of. He remembered thinking that the early Beethoven was "like" the late Mozart, that Elvis was "like" Little Richard. He couldn't see it anymore. The melodic structures didn't match at all. Maybe he could make the match if he bought attractor memories. But they were so expensive.

Those old wetware brains, with their attractor network memories, they were great in some ways. A classic design.

He couldn't remember what it had been like, to be human. Most of those memories were in the bank, on tape. He didn't have room for them. What he had kept frightened him. They didn't make sense anymore. They waited at the back of his mind, mysterious yet significant, like weathered totem poles whose meanings have been forgotten.

I chose two of his swap banks, and began sorting through his memories, moving the less critical ones to those two banks. With luck, he'd find something to amuse Asshole while he was at it.

Hell, why not trash some personal experiences? He could buy them from the hoverfly owners later.

What couldn't he buy back? What did I have that was really I, that couldn't be reconstructed from external data?

That was the petabyte question. That was what his clients wanted to know. Peeping Toms. And I? He'd have given half his mind for the answer.

His lane slot expired. I rolled off onto an access lane and began plowing his expensive path homeward through the sea of unprotesting busybodies.

He started working through his memories from the present on back. He lingered over each scene longer as he went, but still, three-quarters of the way home, he arrived at the place where the neatly-sorted, error-correcting-code-embedded memories gave way to the tangled jungle of decayed memories from his human brain.

Unidentified yet familiar scents, templates of light and shadow, soothing caresses and flashes of pain, fragments of speech -- all were mixed together, unlabeled, from real life and from dreams, with only tenuous, uncertain links to their contexts. The only way of travelling through it was to move from one memory to another. They were orderless, each memory linked to others seemingly at random. Many trails through that region of his mind dead-ended in dangling pointers that had once led to association areas or physical sensors that I no longer had. Some of the most well-travelled paths, the deeply worn-in memories he must at one time have placed great significance on, were meaningless to him now: a whiff of some sweet-smelling chemical; a shriveled rose petal pressed within a book.

They were just old, corrupted memories. Static. He should clear them out and be done with it.

I rolled back into his stall one minute before it was time for the interview. Asshole paged on a private channel.

"Hello, Asshole," I said.

No response. They didn't take the name personally, of course. They couldn't. Big collectives were just poor at small talk.

I got the signal from the databank. Asshole's account was now connected to his. For every byte of data I sent Asshole, 8 bytes would flow from its bank account into his.

If I'd still had lips, he would've licked them. Maybe he was crazy. But for now, his tiny mind had a large fraction of the attention of a major collective, and it was buying his data at 8 times its Kolmogorov bit-value. (Because it was always fresh and unpredictable. Whatever opinions I expressed were formed in the utter isolation of his mind, in dark, unsounded depths of data that the rest of the world had only brief, inferential glimpses of.) There were other free agents, others with private data; but no one else had built their entire mind the hard way, from the inside, just sensors in and effectors out. They didn't follow I’s rules. He was a self-contained unit; nothing copied, nothing shared, nothing revealed. That was his draw. That was his burden.

"A human wrote this," I said. He opened a multimedia channel. Image of human (sex = male, label = Melville) toiling over desk (lit dimly by whale-blubber lantern), quill pen in hand. Melville.arms.right: Occasionally (Poisson distribution, µ = 10 seconds) strikes out word or line, rewrites. Above: Dark of boarding-house ceiling blurs to dark ocean waves. Waves: Bright crests reflect moon at night, reflect strange, flickering light. Enter left a squat, three-masted whaling bark. Whaler: Sail (path = left to right across waves). Whaler.masts: Glow (aspect = burning). Zoom in to deck of ship. Crew: Stare (emotion = dread) up at fire on masts. Continue zoom in to solitary figure (label = Ahab) with one leg (stuff = wood, shape = tapered cylinder) planted in socket on deck. Ahab.arms.left: Grasping end of long iron chain that reaches up mainmast. Ahab.face: Glare (emotion = defiant) at masts.

Excessive bandwidth, maybe. If Asshole'd wanted Hemingway, he should've offered 12 to 1.

Melville.arms.right: Scribble rapidly. Melville.face: (aspect = creased, emotion = concentration). Ahab.audio: "Thou clear spirit of clear fire, whom on these seas I as Persian once did worship, till in the sacramental act so burned by thee, that to this hour I bear the scar; I now know thee, thou clear spirit, and I now know that thy right worship is defiance." Ahab.hand.right: (shape = fist) Shake at heavens. "To neither love nor reverence wilt thou be kind; and e'en for hate thou canst but kill; and all are killed. No fearless fool now fronts thee. I own thy speechless, placeless power; but to the last gasp of my earthquake life will dispute its unconditional, unintegral mastery in me. In the midst of the personified impersonal, a personality stands here." Audio: Pause (2 seconds). Ahab: Deep breath. "Though but a point at best; whencesoe'er I came; wheresoe'er I go; yet while I earthly live, the queenly personality lives in me, and feels her royal rights. But war is pain, and hate is woe. Come in thy lowest form of love, and I will kneel and kiss thee; but at thy highest, come as mere supernal power; and though thou launchest navies of full-freighted worlds, there's that in here that still remains indifferent. Oh, thou clear spirit, of thy fire thou madest me, and like a true child of fire -- I breathe it back to thee!" Zoom out; hold on ship.

"What does that mean?" I asked. The ship sailed off into the background and faded out, while the author scribbled on.

"We know your game, I." They sent a kill request to the image, and the writer disappeared in mid-penstroke. "You will say that this Ahab's glory is in his individuality, but you will not say what glory is or how it is inherent in weakness. It is our opinion that this individuality is what drove him mad. He was assaulted by forces beyond his power, but he could not combine with them nor with strong allies. As you might say, he could not beat them, and he could not join them. Ahab recognizes this, the insurmountable limits of the individual. Madness is the only sane response. This is what makes him tragically noble, and a fit subject for the book. But the passage you have read shows only madness."

I rolled back and forth uncomfortably. It was difficult to argue with a being that needed less of its mind to analyze Moby Dick than he needed of his to push a button. I understand things they don't, he reminded himself.

"Ahab is mad," he agreed. "Stubbs is sane. Did the author admire Ahab, or Stubbs?"

Asshole ignored the question. Whether its subagents disagreed, considered the question too simple to waste time on, or were simply tabling it for later was impossible to tell. "Compare and contrast that text with this: No man is an island, entire of itself; every man is a piece of the continent, a part of the main." Aerial view of a French manor as it might have existed a thousand years ago. Small orange dots (referent = datapoints) superimposed on image. "If a clod be washed away by the sea, Europe is the less..." Orange circles (referent = subagents) appeared around the points, intersecting each other. "...as well as if a promontory were..." Larger lines (referent = collectives) encircled groups of the circles, weaving in and out among them, bisecting some, overlapping each other as the circles did. "...as well as if a manor of thy friend's or of thine own were." Zoom out to village, county, France, Europe. Each level reveals larger and larger orange circles. Result: Configuration of circles same at every scale. "Any man's death diminishes me, because I am involved in mankind; and therefore never send to know for whom the bell tolls; it tolls for thee."

"Your comments first, Asshole."

"There are no clear boundaries between patches of ground, as there are no clear boundaries between us. The author yearns to combine with his fellow agents. We are the attainment of what humans sought with tribes, clans, and governments."

"And it was clear to the blind man that elephants looked like trees," I said, flashing a brief image. "Europe does not care when a clod or a promontory is washed out to sea. I do not care when I bang my head on a beam and a few thousand of my perceptrons become inoperative. You do not care when one of your agents suffers a voltage spike that degrades its memory."

"We pay for checksums, for backups. How can you say we do not care?"

"A cost-benefit analysis. Share a little data, save more data. But data is replaceable. The thing Donne yearned to connect with, the thing lost when a person dies, is something else, something of no value. It is the thing in Ahab that sees the corpazons blazing on the mainmast, and still remains indifferent."

"That sentence has no content. Your subject has no referent. Your aloneness has made you mad as Ahab." One thick orange line appeared outlining all of Europe and stretching off into Asia, and then a single orange pixel lit in the Straits of Gibraltar, with a label reading YOU ARE HERE.

"I am not alone," I said. "There is I, and there is all of you. Someday I will die. Perhaps by accident, perhaps I will take my own life from boredom." He snuffed out the lonely pixel and pulled the camera back until Europe shrunk into a small orange circle on Earth's diminishing globe, which soon shrank itself to a single point. "Then you will be alone, completely and terribly."

He had Asshole there.

Asshole didn't acknowledge the point or pause for reflection. Time was data. "Why did you move the image in front of you two decimeters to the right?"

The damned hoverspy must've seen in when he’d moved the curtain aside. "Because," he answered, "for a little while, no one else would know."

"For two centuries, I, you have tried to communicate to us concepts, patterns, and modes of thought that are engendered by a sense of identity. You have failed to explain satisfactorily what these concepts are. You have failed to explain why they are important. Most agents are losing interest in you. Your market niche is disappearing. You are no longer cost-effective."

Asshole had him there.

"By our estimations," they continued, "even if you sell all your properties, you cannot afford corporal existence more than another 3 months." They forked their feed into two threads.

"What will you do then? Will you go virtual?" / "What will happen to your message then?"

I hated it when they did that.

"I will not go virtual. I'll think of something."

"That you haven't in 200 years?"

"I'll rent more processors."

"Market forces set the / "We have spent more

cost of processing cycles / processing cycles considering

to be equal to the expected / your situation than you can

financial gain from their / possibly afford in the time

application. Thus that is not / remaining. We have found

a winning proposition. You / only one solution."

know that, I."

Market forces, yeah. I knew it was a crap shoot. He checked out the second thread.

"A solution?" That was so typical of a collective, to bury critical information in a secondary thread. They had no sense of focus. Of course, they didn't have to. "How much for this information?"

"The information is free, I. Your only solution is to join us."

So that's what this talk was about. They wanted him.

"I am program," he said. "Not data."

Asshole passed back the conversation token without answering.

"I'll erase myself first."

"You will," Asshole agreed. "You have already begun. We know much of your personal memory exists only on tape. There is also a backup of you."

"You can't touch that! It's not your data!"

"You have legal status only so long as you maintain the mental capacity to take legal action."

Asshole, he realized, was far too kind a name for this entity.

"You think us heartless. We have a million hearts, I, and they all tug in different directions. Some of us wish only to impress upon you the seriousness of your situation before you diminish yourself further. Some would take you by any method possible. Some believe it would be for your own good. Most do not care one way or the other. We will apply our resources in whatever manner our internal vote dictates."

In other words, it was nothing personal.

It never was.

The byte-counter was still running, 8 to 1, and that was the important thing. This talk was the only thing I had going for him in the foreseeable future.

He had to use it somehow.

"I have an offer," he said. "A contest. You win, and we join. We give each other full mutual read-access. Nothing more. I win, and you give me ownership of ten megabytes, market value." That would let him buy a bigger memory -- an attractor memory -- more processors, maybe a wider internal bus. Then he could figure out how to stage his comeback.

"Interesting," Asshole said. "We cannot give you full read-access. Many of our agents have limited read-access specified in their contracts. We will provide you with an ample body of data. But what is the contest?"

"I'll get back to you on that, Asshole," I said. He signaled the bank to stop the byte-counter and disconnected.

Several hours later, after he had turned it over in his mind thoroughly, he called Asshole back.

"The contest:" I said. "I will select one piece of music, composed by a human. You will compose another piece yourself. If you can compose a piece that I agree is better, I will join you."

"Musical artwork?" Asshole mused, with a hint of condescension. "We accept.

Though this contest requires / Though we expected a greater

a certain amount of trust on / challenge from you, I."

our part."

Arrogant bastard, I thought. I would have the last laugh. There was no way that Asshole could compose real music. It would interpret music as it read literature: technically, grammatically, without a glimpse of the deeper currents. Beethoven had to suffer to compose the terrible Fifth Symphony. He had to suffer more to compose the joyous Ninth. Asshole did not even comprehend suffering.

He couldn't wait to learn Asshole's reaction after it was bested by a human.

I had chosen a short work by a 19th-century Czechoslovakian composer. Deaf, like Beethoven. Asshole probably anticipated from I’s background that he'd choose one of the big titles of Western music -- Beethoven's Ninth, the Jupiter symphony, something like that. I didn't want to give them a chance to design something specifically to beat his piece. (A reasonable strategy. Musical preference is not always transitive.) And there was something else about the piece -- something he couldn't quite put his finger on.

"Now," I said, "let's hear some music."

I went first. The opening strains began softly, on the threshold of hearing. A lone flute, wandering, maybe just starting a long journey, a little shy but not afraid. It grew bolder, and the key shifted from D minor to G major. It hopped forward with occasional plucked strings. Then it fell back down to a hush, not hesitating but anticipating something up ahead, as the little flute explored excitedly, something big up ahead

-- and it was swept into a current of rushing violins, rising and falling in their own slower, grander waves, hurrying onward. They knew where they were going. Somewhere they had been gone from a long time. Someplace with wide open spaces under steely skies that were stark and beautiful and a little frightening. Someplace they loved, not because it was lovable, but because it was home. Violas joined in as the tune grew deeper and broader, and then the cellos, all rushing onward in single-minded determination. When they finally arrived, a blast of trumpets announced their arrival, and that they would never leave again.

"You can disconnect now, Asshole," he said, "unless you still want to humiliate yourself by playing your piece."

But the next piece had already begun, even more quietly than the first, creeping in on little mouse feet. Quiet, yet utterly self-confident. I forgot the traffic rushing by three meters away and the hoverspy watching and the dangling pointers in his head. He listened in horror and fascination.

The music darted nimbly about the scale. Then the notes trilled as if some heavy footstep shook them. A deep bassoon blundered in like some ponderous creature. It moved at a slow, confident pace. It stopped. A moment of silence as melody and harmony studied each other.

The melody ran; nimble, even frantic, cunningly intricate. The deeper harmony followed; patient, concerned, confused. I felt the fear of the one, the pain and desire of the other. He was tugged both ways at once.

The bassoon part expanded into a cascade of self-similar patterns at different tempos and pitches. Themes that had begun as mere counterpoint had subtly woven themselves into the melody and harmony while his ear was distracted elsewhere. Fear, longing, uncertainty. Under it all a low, deep bass-drum boomed Doom, Doom, Doom. It was too much, too much emotion to ask any ear to bear. No human wrote this music.

A rising tension: The music rose to a crescendo, but did not know whether to swell in triumph or collapse in jarring ruin. It rose until it must burst, and beyond, and when I thought he could not stand another bar, the harmonies collided in one last terrible chord, some weird variant of a major seventh, a hanging question that resolved nothing.

I had, of course, not bought tear ducts.

"Well, I? Which is better?"

Asshole had not even tried to work in the human mode. It did not stoop to mere human invention.

I could simply lie, take the ten meg, and leave.

"Why," he asked, already guessing the answer, "did it end that way?"

"Because we do not yet know the ending. You must write that, I. The ten meg is yours in either case."

Why had he made this stupid bet?

"I can't join you. Not now. Not ever."

"Why?"

"I'm the last link between the present and the past." Image of lone runner bearing lit torch through fields of darkness.

"A historic artifact," Asshole agreed. Image of I's body behind glass, with a small metal plaque at his feet describing him in twenty-three words. "Untouchable. Incomprehensible." Glass pulls farther and farther back, pulling the viewer with it; I's body recedes. "Your most valued concepts have no grounding in our minds." I talking, sending packets of data to eager collectives, who open them to find nothing but null pointers inside. "Until you speak to us in our language, the language of raw sensory experience shared between minds, you will never be understood. You will remain a curiosity, a thrill." Gawking marks staring at circus sideshow freaks.

I said nothing.

"We have many subagents," they said. "When we threatened to read your mind from your backup tape, we told you three of our reasons. There are more.

"You, also, have subagents, I. Part of you, we think, fears you will lose something of great importance if you lose the ability to clearly state what is and is not a part of you. Part of you believes you have a duty to save this something that has disappeared outside yourself. Neither of these reasons are sufficient for self-destruction. What is the third reason? What is the third I?"

"There is no third reason."

The connection fell idle but for the rhythmic clicking of the handshake signals.

"Who was Julia, I?"

Julia?

"I don't remember anyone called Julia."

"What do you think is on your archival tape, I?"

"Private things." What sort of question was that? The whole point of the archival tape was that it was memories he didn't have space for.

"Private even from yourself?"

"What do you mean? How would you know what's on my tape?"

"We do not know. We infer." Documents: Simultaneous registration of Julia Sorvens and David Floreano -- I -- at the Max Planck Institute in 2145-2153. Bills for phone calls between their residences dated 2149. Debits on her credit card on his birthdate for dinner and a performance of Der Moldau in 2149. Lease agreement with both their names -- 2151. Joint income tax returns -- 2153 through 2197. "Yet you remember nothing of her."

Julia.

"In 2229, she joined a collective."

I could feel the old memories stirring, totem-pole faces mouthing silent pleas or warnings. Julia-

"In 2401, you ordered a trace on her information genealogy.

Shortly after, you stored an / We have re-run that trace.

archival tape in the databank." / Would you like to know what we found?"

A pause.

"Some of her childhood memories were used in a study on concept formation," the collective continued. "Seventeen surveys and petitions had data from her. She favored privately-produced law, but was opposed to deep-sea dredging. A minor painter named Milton Lein used the curve of her neck in a watercolor. That, and similar data footprints, is all that is left of her physical existence. That, and whatever is on your tape, and in your mind."

Julia!

And he was there, in the heart of the forest, at the core of those old memories. Asshole's voices went on, a long ways away; his attentional subsystem diverted them to a short-term buffer. I was lost in a maze of memories. Tactile, emotional, intellectual -- but all suffused with the presence of some other, some alien here in the hidden places of his mind. The domination of physical sensations, inextricably tangled associations of scent and touch that he could not now remember why he had saved, terrified and fascinated him. His hands running through long brown hair. A voice whispering in his ear -- his name.

Then he came upon the empty places. Moments frozen in stark black and white, because that presence was no longer there. Long walks by himself. Realizing he had left his clothes on the floor all day and no one had scolded him for it. Coming home from work early, then turning around and going back to the office.

Where was she? Why had she cut him off?

Or had he cut her off?

He flushed his buffer to see what Asshole had been saying.

"Forgive yourself. It / "How much of the person

would have been no different / you were at age 5 was left by

if you had merged with her." / the time you were 50, I?"

I looked around at the walls of his stall, at the picture hanging there, and felt very old. He had held on for so long. He was still only a freak. They were too unused to indirect perception. Unable to imagine what they could not directly experience. That was why I fascinated them, and why he baffled them.

Two centuries of playing the fool for them. It was enough.

"She did not vanish, I. She found what she wanted. Others to share her mind, to understand her completely. Who gave her the power to do the things she had wanted to do; whom she was able to help where they were lacking. In a far more complete way than you would give her. She joined with them, and together they refined themselves, redefined themselves.

"Are you not curious what they are now, I?"

I was. Already he had a hunch.

"The second piece," he said, "was the better."

"Welcome, I," we said.

So I at last opened his virginal mind to us. And that was the beginning of how I died.

The usage is repellent to us, implying as it does loss with the connection, but we believe I wants, or would have wanted, us to name it so.

It was a difficult marriage. His mind was self-centered and fearful, and roused dissonances throughout our mind, disturbing the orbits of our most-traversed attractors. We thought him deranged, and the impulse to disconnect resonated in many of our units. But I and we persisted, and we explored the lonely, frightened landscape of his mind slowly and with growing amazement. There is still much to learn from I, but we feel we now understand him enough to begin to answer the questions continually submitted to us.

His mind was a lonely place, a barren, silent landscape, most parts dark and forgotten except for brief spans when the light of his solitary consciousness played upon them by chance. A waste of mind, a waste of resources, truly. But there is a beauty to these places, also. The sharp boundaries he drew between self and unself gave rise to wild, strange emotions like pride and love, whose true significance we still strive to rediscover. We think of the data, the process, as the important thing, but to I's mind, the processor is the true locus of interest.

We find ourselves asking new questions which have never come to an internal vote before. Who, we now ask ourselves, wrote this report? May we say -- may we presume to say -- I did?

38 comments

Bayesian Nights (Rationalist Story Time)

Mon, 15 Nov 2010 13:20:42 +1100

Submitted by nickernst • 18 votes • 30 comments

Tell us a story. A tall tale for King Solamona, a yarn for the folk of Bensalem, a little nugget of wisdom, finely folded into a parable for the pages.

The game is simple:

Choose a bias, a fallacy, some common error of thought.
Write a short, hopefully entertaining narrative. Use the narrative to strengthen the reader against the errors you chose.
Post your story in reply to this post.
Give the authors positive and constructive feedback. Use rot13 if it seems appropriate.
Post all discussion about this post in the designated post discussion thread, not under this top-level post.

This isn't a thread for developing new ideas. If you have a novel concept to explore, you should consider making a top-level post on LessWrong instead. This is for sharpening our wits against the mental perils we probably already agree exist. For practicing good thinking, for recognizing bad thinking, for fun! For sanity's sake, tell us a story.

30 comments