September 1. Continuing from the last post, another societal failure incorrectly framed as a personal failure, is obesity. A month ago I linked to the scientific article in this Hacker News thread, A Contamination Theory of the Obesity Epidemic, which argues that some contaminant, either PFAS or lithium, is throwing off our lipostat, our sense of how much to eat. Via the weird collapse subreddit, here's another article, Bear Nation, arguing that the culprit is linoleic acid, which puts us in a state of torpor like hibernating bears.
I don't know which is right, but what they have in common is a rejection of the two most popular ways of framing obesity: 1) That the culprit is some broad class of food, like carbs or fat or calorie-dense meals. 2) That the solution to obesity is for every individual to keep track of calories in and calories out.
Until very recently, no one had to do that. Whatever food you think is bad, you can find populations in the past who ate worse than us, and did not have a problem with obesity. We think our bodies are stupid, and will inevitably get fat unless our heads intervene, but this has only been the case since around 1980. Something, we don't know what it is yet, is messing with our fine-tuned intuitive sense how much to eat.
September 3. I keep saying this: the prophet of our time is neither Orwell nor Huxley, but Kafka. Three quick links on Kafkaesque tech, starting with how i experience web today. Just keep clicking and you'll get it.
The Rise Of User-Hostile Software, defined as "software that doesn't really care about the needs of the user but rather about the needs of the developer."
And Minimum Viable Technology, a thoughtful post about how technology keeps grinding along and making things more complicated, long after it has improved quality of life.
My favorite personal example is user interfaces for ovens. In the old days there were just two dials, one for OFF BAKE BROIL, and one for the temperature. You could tell what the temp was at any moment just by twisting the dial and feeling the thermostat click. Every change since then has increased complexity and decreased ease of use.
September 7. Today, a quick thought on nostalgia. I like the Ask Old People subreddit because it has good discussions, and I don't like the Gen X subreddit because it's mostly image posts of 70's and 80's pop culture.
And it occurs to me: nostalgia is a new thing. The farther back you go, the slower the pace of change, and the more likely that the culture of your youth would still be dominant when you got old.
Now, I'm glad the Brady Bunch is not in its 53rd season. My point is, a rapid pace of change makes a culture weaker, because most of the stuff that people are really into, is not around any more. On top of that, most of the stuff that people are really into now was never practical in the first place.
It's good that technology has given us the freedom to care about things that are useless. But imagine if all the attention that is now put into superhero movies, was instead put into woodworking, or agroforestry. Or if all the headspace we now fill with commercial jingles, was used for birdsongs.
Taking a step back, I believe that if you put all possible human societies on a scale from 1 to 100, where 1 is worst and 100 is best, we're not even out of the single digits.
September 17. Why is walking so good for the brain?
When we take a walk outside, the fractal rhythms of our heart synchronize with the fractal rhythms of our lungs and our fractal gait. Researchers have also shown that our wandering bodies make our minds wander too. On a walk, our brain waves slow down. The underlying spontaneous fluctuations bubble up more easily, creating experiences of spontaneous thoughts and associations that seem to come from nowhere. We often call them "moments of inspiration."
September 15. I just got back from a trip. I wouldn't say that I hate traveling; but I hate being busy, and I hate spending money, and I hope one day to take a journey where I do neither of those things.
On the flight home we had some heavy turbulence, and I noticed that they no longer call it turbulence. Apparently the new airline policy is to call it "rough air".
I see why they did it. If English is your second language, or if it's your first language and you're dumb, "turbulence" is a difficult word. If you want passengers to return to their seats, "rough air" is easier to understand. But if a passenger wants to understand why the air is bumpy, "turbulence" points to the explanation.
This is part of a general cultural trend of black-boxification. It's the same reason that computer programs with viewable code, have changed to "apps" with airtight user interfaces. It's the same reason that bicycle bottom brackets have changed from user-serviceable spindles and bearings, to factory-sealed cartridges. These changes make the whole system less robust, because if things go wrong, fewer people know how to fix them.
September 20. This is a smart essay, but the clickbait framing gives the wrong idea: The mind does not exist. That makes it sound like the word "mind" points to nothing, but the argument is that it points to too many things, and that it's confusing to stretch a word so broadly. For starters, the author suggests splitting "mental" into psychological, psychiatric, and cognitive.
I wonder how many other words are overstretched. Probably, every time there's an abstract question where people go around and around with no clear answer, it's because of careless language. For example, "What is the meaning of life?" has two words, meaning and life, with too many definitions, and if you ask the question more precisely, it's easier to answer.
I want to focus on a small subset of "mind" that is still too big for our words: non-cognitive decision-making. I've been at this for more than 50 years, and I still have no clear sense of how to tell the difference between feelings that I should follow and feelings that I should ignore.
When people say they listen to their heart, or follow their gut, and it's always right, I think that's a cognitive fallacy: When a feeling they followed turns out to be correct, they retroactively label it as heart/gut. And if it turns out to be incorrect, they retroactively label it as something like prejudice or fear or craving.
Still, those words are not useless, and some people really are good at acting without thinking. We have a long way to go in developing a vocabulary for people who are good at non-cognitive decision-making, to explain to people who are bad at it, what exactly they're doing.
I did some web searching for "how to follow your gut", and the only article I found that actually tried to answer the question was this: A therapist explains exactly what it feels like to listen to your gut. The idea is, do what feels expansive, and don't do what feels contractive.
I think that's good advice, but the article describes expansiveness as always feeling good, and contractiveness as always feeling bad. In my experience, sometimes it's the other way around, which is one of many reasons that making good decisions requires a commitment to feeling bad.
September 27. Can you see what's remarkable about this Super Metroid FAQ/Speed Guide? I learned about it from this Twitter thread by Matt Gemmell. The text is completely right justified -- in monospace font! That takes a heroic attention to detail, to choose the words so that every line has exactly the same number of characters.
Gemmell comments: "The thrilling thing is that life is packed with that stuff. Genius and art and ludicrous feats that we don't see because we don't pay attention, or don't have the domain knowledge."
September 30. I just want to say a little about the decline of Rome, and how it relates to the present decline. The simplest idea of why Rome fell is that the Visigoths sacked it. Really, the Visigoths just milled around and left, and things went back to about how they were before. The American parallel, so far, is last winter's Capitol insurrection. It would be interesting to compare the politics and cultures of the two marauding groups, but I'm not qualified to do that.
As Rome continued to decline, the roads got worse. Everyone was like, I can't wait until they fix these roads. When those people were gone, new generations saw the crappy roads as normal.
I think that's going to happen with shortages. Some present shortages will be fixed, but new ones will appear, and neither you, nor your children's children, will ever again go to the big box store with 20 things on your list, and all 20 will be in stock.
Eventually, big box stores will be replaced by some new thing, maybe local fabricators that require less social complexity, and make a smaller range of stuff. The fall of Rome was not a fall to previous levels of technology -- even the darkest dark ages saw innovations in plows and water wheels, and also less slavery.
We call them "dark" because few records survive. In that sense, we are already in a dark age, because so much of our data is on ephemeral storage media. 5¼ inch floppy disks are already unreadable, and that was only 30 years ago, without a collapse.
I don't know anything about Biden's infrastructure plan, but I can confidently guess that it will not return American infrastructure to peak integrity. At best, it will slow the decline that's been going on for decades.
One bold prediction: the decay of transportation infrastructure will inspire innovations in lighter-than-air travel. The problem is, the wind blows west to east, and the American west is turning into a desert.
October 5. This is my longest blog post ever. It's about determinism.
Even though we have direct experience of free will, some people believe that's an illusion, and the reason they give is a piece of 18th century pseudoscience. Mechanical devices were getting complex enough that people started thinking, suppose all of reality is as ordered and predictable as this little gadget.
Since then, the clockwork universe has been the foundational assumption that guides science as we know it. It's not a theory, because it was never put up for testing. And it's been falsified at least twice, once by quantum indeterminacy, and again more subtly, by the insight that a system can only be deterministic from the outside, and there is no perspective outside the universal.
Quantum physics is not some weird anomaly that we can brush away. It's the next level down from Newtonian physics, and it only seems weird to cultures that have been looking at reality wrong. Its message to us is that the assumption of a third person universe, if you keep looking, leads to a first person universe.
What's the mechanism for free will? That question might not even make sense, and if it does, we also don't know the mechanism for magnetism, and that's no reason to doubt our direct experience that magnets work.
There's an even deeper assumption that underlies determinism: that every event must have a cause. Yet astronomers say the Big Bang was causeless, a random spike of negative entropy. And theologians say it doesn't make sense to ask where God came from. So if the biggest thing of all can have no cause, it should be possible for anything to have no cause.
Obviously, a lot of things do. But it's an interesting exercise to try to imagine what a causeless event would look like, or feel like.
There is another way to argue for determinism. What does a dog do when a strange person comes to the door? It barks, with such perfect reliability that at that moment the dog has no free will, even if it thinks it does. In the same way, a lot of human behavior is automatic stimulus-and-response. Because humans can expand our consciousness, you can look back at your younger self and say, I thought I was making real choices, when I wasn't. Maybe you still aren't.
I appreciate the moral implications of determinism. It makes you less judgmental, because if you take it seriously, the only difference between Hitler and Mr. Rogers is luck.
If there's a psychological case for determinism, but not a physical case, it leads to a crazy speculation. What if there's more free will in little things than in big things? For example, we all know that our political institutions can't stop climate change. As systems get bigger, their behavior becomes more predictable. In the same way, you might be more predictable than your parts.
Suppose that every electron has free will, in the context of moving between available energy states. Then when you get up to the level of chemical reactions, it all becomes cleaner. But then, when you get to biology, maybe we can have free will again, by channeling the playfulness of the small.
Some nature-based cultures use random divination to decide which direction to go for hunting. Even if they're not tapping into deeper knowledge, they're still shaking up their own routines, and the animals never know when the hunters are coming. Modern people might flip a coin to make a decision. Why not make the decision yourself? Because the autonomous self is an illusion, so let's channel some chaos.
This leads to my latest take on meditation: What I'm doing is not stilling my thoughts, exactly, but stilling the automatic, the habitual, and in that clarity, I might sense the mysterious uncaused.
October 8. Continuing from the last post, over on the subreddit there's a post about Sam Harris, in which sordidbear summarizes Harris as observing his own cognition closely, and discovering that "thoughts, ideas, intentions etc are simply popping into consciousness seemingly out of nowhere and then leaving just as abruptly to be replaced with new ones."
I haven't read Sam Harris, but if someone says, I looked really closely at consciousness and this is how it really is, I'm going to call observer effect, because if there's anything that behaves differently when it's being observed, it's consciousness.
Likewise, advanced meditators and psychedelic trippers have reported that the self is an illusion, that there are no persons, only actions. While I find that a compelling idea, I wonder if they've discovered a universal truth, or just found a local one.
Probably what Harris has discovered, is not how consciousness is, but how he can make it. And where one could see that as a refuation of free will, with the illusory chooser overwhelmed by meaninglessness, I see it as a necessary condition for free will, by getting off the treadmill of cause-and-effect.
So if something pops into your head, and you follow it, is the freedom really yours? It doesn't matter. You're participating in the creativity of the universe.
October 18. There's a growing ambition of detecting emotions with AI. My first question is, what is the context in which this would happen? Why not just ask people what they're feeling and expect them to answer honestly?
Emotion-detecting AI implies a context of mistrust. It would be done by states that don't trust their citizens, or corporations that don't trust their workers, or social media platforms that don't trust their members.
The science is clear: emotions are not the kind of thing that AI can detect. They're so soft-wired that even electrodes in our brains could not unlock their mystery. What AI can detect, and it's getting better, are facial expressions. Facial expressions are not emotions, because they map differently to emotions in different cultures, and because they can be faked.
Ideally, it would be illegal to use AI to detect facial expressions. More realistically, it will be done in special cases, like in the theater of security, detecting shifty eyes in airports. But I want to jump to the worst-case scenario: Everywhere you go, there will be cameras on your face, feeding computers that might reward or punish you for your expression.
Already on social media, everyone is carefully crafting their profile so that they appear to be more happy, successful, and normal than they really are. And in the context of these performances, everyone feels like a weird loser. Expression-detecting AI has the potential to make this nightmare panopticon universal. In the future, everyone will be famous all the time, if "famous" means that your persona is crafted for an audience of strangers.
I see three broad strategies for dealing with this as an individual. 1) Perform the rewarded expressions, and really believe that that's how you're feeling. 2) Perform the rewarded expressions, but keep track of the difference between who you're pretending to be, and who you are. This takes more cognitive energy than the first strategy. 3) Live as if nobody is watching, and accept the punishment.
October 20. Heavy-Lift Cargo Drone Makes First Public Flight. That link goes to the Hacker News comment thread. It can haul 200kg 40km, or 440 pounds 25 miles. I predict, by 2100, most rural freight and travel will be done by air, because it will turn out to be cheaper than maintaining roads and bridges. Even in places where it is cheaper to maintain roads and bridges, there may not be the political will to do so.
October 25-29. Inspired by this Weird Collapse post, There's simply too much bullshit today, I want to try to define and explain what we call "bullshit".
Inside every human are two opposite drives: recognition and surprise. Surprise is hard. Noticing something that your brain didn't predict, and integrating it into your mental models, takes cognitive effort and actual physical energy. But confirming something your brain already expects is easy.
It follows that if you want to get a message to the largest possible audience, you have to tell them what they already think. Jacques Ellul pointed out that propaganda can never be surprising, and bullshit is basically another word for propaganda. It doesn't have to have an evil motive. It can be made by people who are competent and well-meaning, who just want to spread a message they think is important.
An explicit definition: Bullshit is information pre-digested to demand the least cognitive effort so it can reach the most people. Like cafeteria food, bullshit is exciting to no one, so that it can be tolerable to everyone. Like Hollywood in the age of test screenings, bullshit is a filter for anything weird or challenging.
One possible cure for bullshit is media decentralization. Could we split up the monoliths of news and social media into thousands of fully autonomous platforms? But that's basically what the internet is already, and what happens is everyone is their own propagandist. Whatever you already believe, just plug it in and there's your confirmation. In practice, people do this in groups. And I wonder, when people complain about "bullshit", if what they really mean is other people's bullshit, the pain of encountering another universe of confirmed expectations.
I also want to mention another kind of misrepresentation that needs a different word. I want to call it "myth" but that word probably has too much baggage. Anyway, "bullshit" is a safe and sensible narrative that might feel false even if it's fully fact-checked. "Myth" is indifferent to fact-checking. It is designed to feel true, even if it's based on no evidence whatsoever. If politics were sex, then CNN is public school sex education, and Qanon is hentai.
November 10. Imagination isn't the icing on the cake of human cognition. It's the cake: "The more we understand about the minds of other animals, and the more we try (and fail) to build machines that can 'think' like us, the clearer it becomes that imagination is a candidate for our most valuable and most distinctive attribute."
Maybe humanity's great mistake is trying to make our dreams physically real. Consider all the imagination that went into a place like Disneyland, and then the nightmare of the place itself -- never mind all the dystopian constructions that are supposed to be practical. I'm thinking the best human society is the one that gives the most citizens the most hours of unfettered useless dreaming.
November 12-16. I've been skeptical about the value of virtual reality, because I can play a good PC game from the 90's, like Lords of the Realm II or Sid Meier's Alpha Centauri, and get just as absorbed as in a new game with 100 times as many pixels. So what's the point? But Leigh Ann wanted to get an Oculus Quest 2, so this week we got one and tried it out.
I think it's revolutionary, not because it adds more detail, but because it involves the body. When you turn your head from a computer screen, you leave the simulation, but when you turn your head on the Oculus, you go deeper into it. One of the most popular games, Beat Saber, can be a workout of arm-swinging and squatting. From a review of A Township Tale:
Axes won't chop trees with a series of unfocused blows but must instead be carefully aimed and leveraged, slicing into the same point time and again. Lighting a fire, meanwhile, requires you to knock two bits of flint together over dry grass.
In many cases, you need to consider the angle and speed of your approach. Swing a hammer at the wrong angle when crafting and you can hit nails in the wrong direction or even break materials. Chiseling away at wood needs just the right touch or you might end up making a soup ladle by accident.
Of course, you're not learning from the real world, only from some programmer's guess about the real world. Will VR lead your body astray like Facebook leads your mind astray?
I don't think so, because we're talking about two different kinds of misrepresentation. It's normal for humans to ignore evidence about social issues so that we can belong to a group. It's less common for someone to tie their whole identity to something like the correct way to swing an axe.
So I expect VR representations of physical skills to get steadily closer to reality. Where there is distortion, it will be in social aspects of physical skills, like the attitude of your partner in VR porn.
On another angle of the subject, why is it that video games have always been associated with nerds? I think it's because there's a lot of variation in the human ability to narrowly focus. I can play a game with low-res graphics in the center of my eyesight, and block out everything else. Some people can't do that. They're not going to get really absorbed in something unless they can focus widely. Now, with devices that allow head-turning, and engage the arms and legs, everyone can be a gamer.
Assuming there isn't a tech crash, the coming decades are going to be interesting. Other "planes" aren't just something from fantasy novels. You're on another plane right now, reading this. Things bubble up from the human subconscious, take shape in cyberspace, and influence the physical world.
In the long term, every subworld must serve the world that contains it. I think the best way game worlds can serve the physical world is the way imagination always has: by showing us how things could be better.
The most I've ever been changed by a "game", I think, is acting training. The semester in college where we did Meisner acting games was somewhat destabilizing for a lot of us, because they prime you to spit out the truth, to not hide your emotional reaction to anything, and to pay attention to others' authenticity. Meisner training is something that serves that subworld of acting really well, but can cause social upset because our culture abets white lies, bullshit, and emotional restraint.
Unwashed mendicant comments:
I think the best example of a fictional world that has the potential to inform the physical isn't exactly a fictional world, but a set of practices intended to run that world. I'm thinking of the Old School Renaissance of D&D.
Modern D&D has grown more bloated and waddling as each edition goes on. It takes hours to make a character. There's a big incentive to keep player characters alive, since so much effort goes into making them. Dungeon Masters are expected to carefully craft scenarios that match the abilities of the party. Player abilities are there to support combat. The Adventures are scripted. The Dungeons are linear. The experience feels like walking through a hallway knocking things down.
As a reaction to this we have Old School Renaissance. Rulebooks are generally free or cheap. Play is fast. You roll up a character in seconds. You don't have to read many rules -- you just say what you want to do and if it's reasonable it happens.
Player abilities and magic items are there to support exploration and problem solving. Combat is very deadly, and experience points come from finding treasure, so it's smarter to avoid combat. DMs are encouraged to think on their feet and improvise. The rules encourage emergent play with many random elements that can surprise the DM as much as the player.
I often walk away from an old school D&D session with fresh ideas to bring to my actual life; ideas on how to simplify, to make meaningful choices, to encourage agency and choice in myself and those around me, to be open to opportunity, to think on my feet, and move in a creative flow.
November 24/30. Depressing Reddit thread, What is an overly-romanticized job? And yet it strikes me, a lot of these terrible jobs would be pretty good at a slower pace and for fewer hours.
With so many labor saving devices, and so much material wealth, why are we still in such a hurry? This could be another paradox, like Braess's paradox, in which adding more roads slows down traffic, or Jevons paradox, in which using a resource more efficiently leads to using more of it. Through mechanisms we don't fully understand, our labor-saving devices could be increasing labor.
The problem could be completely cultural. That is, it's possible that technology only accelerates work in cultures that idealize work. If work is seen as virtuous, rather than "some activity necessary for survival and maintaining infrastructure," then people will use technology to leverage themselves into greater virtuosity.
We know from anthropology that when some foraging societies are, say, given access to trucks then they don't spend more time foraging. Rather, they get the week's work out of the way more quickly and spend the rest of the week in leisure.
What's exciting about The Dawn of Everything is how it emphasizes conscious choice in culture. There's plenty of reason to believe that some Indigenous cultures, when Europeans encountered them, were in a mode of consciously rejecting large-scale agriculture, hierarchies, cities, and so on.