Ran Prieur

June 17.

2020-06-17T17:30:43Z

June 17. Some stray technology links, starting with a smart one about AI: Human-Level Intelligence or Animal-Like Abilities? What I would say, which is not quite what the author is saying, is that everyone wants computers that are smart like we are, but the actual progress in AI is being made in systems that do not understand concepts in the way that humans do, but that work anyway. "Cats have navigation abilities that are far superior to any of those in existing automatous-navigation systems, including self-driving cars."

A couple weeks ago, on the same day as the Space-X launch, this also happened, and it might end up being more important: Electric-powered Cessna makes maiden flight, "the largest all-electric passenger or cargo aircraft ever to fly."

A weird idea that's obvious in hindsight: Stanford lab envisions delivery drones that save energy by taking the bus.

And after a retrofit, the Golden Gate Bridge makes trippy music when the wind blows.

June 15.

2020-06-15T15:10:14Z

June 15. I meant to post this more than a week ago, but other stuff got in the way: Weirdos during the depression is a short blog post about two characters in classic novels, one who pretends to be a miserable drunk so people won't bother him about his mixed-race marriage, and one who pretends to have a gruesome facial scar so people won't bother him about his beard.

Here's the Hacker News thread and the subreddit thread, where I said this: "If you break rules that other people are following, you have to pretend to be unhappy, or they'll get really mad, because they don't want to face the grief that they could have been breaking the rules themselves all this time."

Again digging into the archives, the rest of this post is from July of 2008:

One of my favorite bits in the Bible is the parable of the jealous workers: Some workers show up at a farm in the morning, and agree to work all day for a certain wage. Then some other workers show up in the afternoon, and they get the same amount of money for only working half a day. And the morning workers say, "No faaaair!" And Jesus says, "Morning workers, you consented to that deal and it's wrong for you to resent someone for getting a better deal."

To this day, religious scholars have trouble taking the parable at face value, but if you think it through it makes perfect sense. If the morning workers get their way, if it's not fair for people who came later to get a better deal, then the world can never get any better.

June 12.

2020-06-12T12:41:18Z

June 12. Double post today. First, there's a dust-up on the ranprieur subreddit, with some discussion about what kind of online community people want. Personally, I like the weirdcollapse subreddit better.

When I started this site in 2002, I named it after myself and not some subject, because then, whatever I end up writing about, the name is still accurate. But I still have to deal with the energy that bounces back from everything I've ever written. My latest project, to clean up that energy, has been to go through the entire blog archives and cut them down to stuff that I still like. I don't want people to ask "What did Ran Prieur think about this?" I want them to say, "Hey, this guy has some good ideas."

I purged almost all presidential politics, and a lot of wrong predictions. (I didn't notice any right predictions.) And I found myself cutting a lot of final paragraphs, where after I had the good ideas, I tried to say why the ideas were important. Every archive is now linked from the bottom of this page, and in the future I might be doing more explicit reposts, instead of doing new posts on stuff I've already covered.

June 12.

2020-06-12T12:40:49Z

June 12. For today's main post, I have more to say about tacit knowledge. First, from 2012, a piece with two strong examples of tacit knowledge, Chicken Sexers, Plane Spotters, and the Elegance of TAGteaching. In both chicken sexing and plane spotting, people can learn to reliably tell the differences between things, and teach others, without ever knowing how they know.

Second, the tacit knowledge article contains a method for learning to ride a bicycle, where you start with a very small bike, which enables you to learn the skills with zero risk of falling over. My thought is, I feel like I've lived my entire life on a tall fucking bike, falling over and falling over and falling over when I'm supposed to be learning. Maybe that's what's causing the epidemic of anxiety and depression: technology is changing our environment so fast that there's not enough room to make mistakes without being punished.

Third, the rest of this post is a repost from March 3, 2010:

Last week when I wrote about saving knowledge, everyone agreed that how-to-do-it knowledge is easier to preserve alive than dead -- through a living community of people teaching each other, rather than through written instructions. This is because if a skill is at all difficult, it contains subtleties that are easy to understand and transmit through hands-on practice, but almost impossible to transmit through words and pictures.

Dameon had the idea to take this basic principle and do some quantitative thinking. Over the last few centuries, as western civilization has grown more complex, it has depended on a larger and larger number of living skills. We are now orders of magnitude above the number of living skills that a forager-hunter tribe depends on. How was this possible? Through increasing population, and through specialization. There are people who know how to design a computer chip, but have no ability to feed themselves without a massive industrial infrastructure -- which now depends on computer chips.

Now, what happens when the population stops rising? Can we rein in and stabilize complexity so it doesn't overshoot our ability to know how to do everything? I doubt it. And what happens when some crisis forces specialists to generalize? If fiber optic technicians have to grow potatoes to survive, key skills for maintaining fiber optic networks could be lost. There are probably tens of thousands of skills equally obscure and important.

And if a skill dies, even if there are still books about it, the human attention required to resurrect it from books is much greater than the human attention that would have been required to keep it alive in the first place. So if we want to bring back a dead skill, without an increase in population or specialization, we have to sacrifice some living skills.

What we're looking at is catabolic collapse -- a loss of complexity that feeds back and causes more loss of complexity, and so on until the system finds a new point of equilibrium.

June 10.

2020-06-10T22:20:44Z

June 10. I can't wait until humans get politics figured out so I can stop writing about it. Today, some head-stretching links on other subjects.

From the "advanced technology is indistinguishable from nature" department, Slime mould simulates Tokyo rail network.

This Is What Happens When You Take 550 Doses of LSD At Once. The article covers several cases of accidental LSD overdose, and this is the one from the title:

When the drug finally wore off another 10 hours later, CB felt normal, and her chronic pain had completely disappeared. For seven years she had been taking morphine every day to treat symptoms of Lyme disease. After her LSD overdose, not only had her pain evaporated, she felt no withdrawal symptoms from the opioids she had been taking.

Going deeper into mind-body stuff, a good one from the subreddit, What we can learn from untranslatable illnesses. These are illnesses with physical symptoms, that only appear in certain cultures. When people say "it's all in your head," they seem to think that should make it easier. Really, mind-based illness is on a whole higher level of difficulty.

I don't have a link, but I heard about an illness that only appears in one Latin American culture. The symptom is that you have really bad luck, and the only cure is to go to a shaman.

What is emergence, and why should we care about it? This is a dense critique of reductionism, the idea that no matter how complex something is, it can always be understood by breaking it down into little parts. This is not science but faith: when in practice reductionists can't explain wholes in terms of parts, they insist that they just need better number-crunching.

Emergentism is the idea that the failure of reductionism is not merely practical, but fundamental:

that parts and wholes have equal ontological priority, with the wholes constraining the parts just as much as the parts constrain the wholes... that the universe is in some sense open to novel phenomena that cannot be perfectly anticipated using any scientific theory, but, once present, can still be studied using scientific methods. In other words, emergentism suggests that even our best quantitative theories cannot always tell us when qualitative changes will occur.

The same subject from another angle: Why Tacit Knowledge is More Important Than Deliberate Practice. The idea is, to get good at something, you have to learn skills that can't be described in words. More good stuff in the Hacker News comment thread.

June 8.

2020-06-08T20:00:45Z

June 8. This is really obvious: the police are used against lower-class crimes, and not upper-class crimes. When Facebook breaks a law, cops are not going to march into Mark Zuckerberg's office and tase him if he resists. I mean, if he does something really bad, the police will show their presence, but they don't need to, because he knows if he runs away, they'll finally start treating him the way they treat the lower class by default.

At the same time, big government can stand up to big money. So when someone says they're anti-government and pro-police, what they mean is they support force going down the pyramid, but not up it.

For all of history, force has gone down the pyramid while wealth has gone up it. America didn't change that, we just had enough prosperity to buy off the middle class. As the age of economic growth ends, the middle class cannot be bought off, and they're noticing that their interests align with the lower class.

I'm actually proud of how well America is handling this. We're working through a lot of shit in a short time with minimal casualties, and seriously talking about real reforms.

Four links on how policing could be done differently. First, Are Cops Constitutional?

Professional police were unknown to the United States in 1789, and first appeared in America almost a half-century after the Constitution's ratification. The Framers contemplated law enforcement as the duty of mostly private citizens, along with a few constables and sheriffs who could be called upon when necessary. This article marshals extensive historical and legal evidence to show that modern policing is in many ways inconsistent with the original intent of America's founding documents.

From the NY Times, Cities ask if it's time to defund police and reimagine public safety. Specifically, "many social welfare tasks that currently fall to armed police officers - responding to drug overdoses, and working with people who have a mental illness or are homeless - would be better carried out by nurses or social workers." That means fewer jobs for people trained in using force, and more jobs for people trained in engaging citizens without force.

Rolling Stone has just reposted this 2014 piece, Six Ideas for a Cop-Free World.

And What America can learn from Nordic police. Reading this makes me realize how really authoritarian America is. Our baseline culture takes for granted that torture works, that sentences should be long and prisons should be dreadful, that being nice to people is a bottomless sink, and humans will only serve the larger good under threat of punishment. Nordic countries are assuming the opposite, and it's working better for them.

June 5.

2020-06-05T17:30:18Z

June 5. In coverage of civic unrest, "violence" is a propaganda word, by which I mean, it's both morally loaded and sloppily defined. In practice, the word violence makes crimes against property seem just as bad as crimes against people.

Property is just a big game we're all playing, and it hasn't been fun for a long time, if it ever was.

Here's a thought experiment. Imagine the police all got together, and announced, "We're tired of putting our asses on the line protecting stuff. From now on, we're only enforcing laws that protect people." That's what good cops actually do when things get really bad.

The first thing that would happen is, any concentration of valuable stuff near poor people would be looted. Then the big companies would hire mercenaries to protect their warehouses. But if we break in, they can't shoot us, because laws protecting people are still being upheld. So if enough people gather to win a shoving match, they can take possession of anything.

At the same time, any business that's on good terms with most of the nearby people will be protected. More generally, any system of human activity that can justify itself locally will be preserved, while any system that relies on far-flung abstractions will be dismantled, and the whole economy will be stripped down to activities that make sense on a human scale.

Okay, but where does your friendly mini-mart get its merchandise? Who's going to make stuff if they can't make money from selling it? The answer is, anyone who enjoys making stuff. The manufacturing economy would be stripped down to home workshops and the happiest factories. I don't think any of them are making microprocessors, so high tech would be stripped down to stuff that's easy to scavenge and tinker with.

What about food? Farmers love their work, otherwise they would quit, because it's a hard job that pays basically nothing. But industrial farming would slowly fall apart without industrial manufacturing, and distributing food would be even harder. So we would need a lot of volunteer work, and really skillful organization, to stop a lot of us from dying.

But in trying to get through this, we would be constantly asking two questions of every task: 1) Is this really necessary for human well-being? 2) Am I enjoying it? Notice how rarely, under the present system, we hold our actions to those standards. Instead, the question we're always asking is: if I say no to this bullshit, how much will my life be ruined?

This thought experiment is an extreme simplification of a process that's going to actually happen over a really long time. My point is, the more we respect quality of life over claims of ownership, the more meaningful and enjoyable our lives are, if only we can overcome challenges that are mainly logistical.

June 3.

2020-06-03T15:10:56Z

June 3. Timely Terence McKenna quote, from this video:

We can embrace chaos, and see that chaos is the environment in which we all thrive. That's how I've done it for years. You think I could have gotten away with this in the Soviet Union? I don't think so. I require a society on the brink of social breakdown to be able to do my work, and I think a society on the brink of social breakdown is the healthiest situation for individuals. I don't know how many of you have ever had the privilege of being in a society in a pre-revolutionary situation, but the cafes stay open all night and there's music in the streets and you can breathe it, you can feel it, and you know what is happening. The dominator is being pushed.

It never succeeds, it never is able to claim itself. But on the other hand history is young. We may have a crack at this. A global society is coming into being, a global society made out of information that was not intended to be ours, but which is ours, through the mistaken invention and distribution of small computers, the printing press, all of this stuff. Information is power, and information has been spilled by the clumsy handling of the cybernetic revolution by the dominator culture, so that it is everywhere. Never has the situation been more fluid.

June 2.

2020-06-02T14:00:45Z

June 2. My latest take on Trump, standing on the shoulders of this excellent reddit comment from early 2016.

The nice thing about Trump is, he's not an ideologue -- he's a negotiator. He doesn't really believe in anything except his own power, so he found a movement he could get in front of, and he's very good at knowing what they want.

As their negotiator, he always starts with an offer as extreme as he can get away with, so that the eventual compromise will be more in his favor. [update june 3] For example, CNN thinks that when he had a street aggressively cleared of peaceful protesters so he could walk to a church, that was an act of clumsiness, when really he was carefully testing how far over the line he could step.

So America is being tested, to hold the line against the authoritarian personalities among us, and to push the line back, as we try to overcome our own history as invaders and slavers.

June 1.

2020-06-01T13:50:29Z

June 1. Two comments from a Reddit thread about what non-Americans think of the protests, both with a good thread below. From Luxembourg:

Nobody here understands why you are so unable to reform the police. The clear problem is a severe lack of police accountability, a lack of non-violent training, way too short training spans, an omnipresence of guns, a militarized police and a deep-rooting institutional racism.

And from the other side:

Most of the world does not have a chance protesting the police. The thought of doing so will get you locked up. If you think you have it bad in the US, you have no idea how it is overseas. The police in developing countries get away with everything.

Also, here's a repost of a comment on this page from 2014, about civil wars:

Civil wars are first and foremost about local score settling. The trigger isn't some guy going door to door saying "you know those Yazidis? We're starting a group to get rid of them, would you like to know why?" Everyone was already itching to kill the Yazidis. The trigger in most civil wars is the sense that the long-repressed vengeance on your nearest and dearest enemies has become possible.

Civil wars are almost never geographic at first. Syria was not divided into "rebel" and "government" territories until after several weeks of fighting. Why? Because the government troops and the people who hated them were evenly dispersed around the country. Once the shooting broke out, some local battles went one way, some went another, and each side eventually had to work out supply lines connecting places where they'd won. Your loyalties aren't determined by your residency, your residency is determined by which army you're running away from.

Civil wars aren't anybody's program. Usually the two sides each feel like they are legitimate, and can't figure out what the other guys are playing at. They think "shit, these guys are clowns, let's just get them out of the way." Everybody underestimates the consequences of their actions, the time it will take, and the dying that will happen as a result. Nobody in Syria in 2011 was saying "right, lets call a protest, and in three years we'll be holed up in a burning hotel shooting twelve-year-olds in the head as they pull their mothers' bodies from a drainage canal!"