While there are zealots who genuinely do think that the brain is literally a digital computer, I think most people would, when pressed, admit that it’s an analogy. The prevailing technology of the day has a long and distinguished history of being used as a metaphor for describing thought–early modern philosophers loved to talk about mental events as a kind of clockwork mechanism, for instance–but those analogies are not generally to be taken literally. The mechanists of the 18th century didn’t think that there were literally gears inside your skull; they just thought that the idea of an unimaginably intricate clockwork mechanism was a useful way to think about the functioning and organization of cognition (which it is, at least in some ways).

There’s maybe more literalism about it these days than is standard, but even most of the people who take this analogy very seriously aren’t saying something so trivially false as “your brain has a literal CPU and works exactly the way your laptop does.” That’s pretty obviously not true. But the analogy is a useful one in many ways, and can help us understand what the hell is going on in there that lets a big chunk of meat give rise to such an extraordinary phenomenon as consciousness. The observation that when I do something like add two and two in my head there must be something going on that is, in some relevant sense, functionally identical to what goes on in a desk calculator when I enter 2+2 isn’t totally vapid. It’s possible to take all this too seriously, and moving from “there’s some amount of functional parallelism here” to “these two systems are functionally identical in general” is (I think) an unwarranted one. But, again, I don’t really think that’s the interesting thesis here.

Computers, quite literally, process information – numbers, letters, words, formulas, images. The information first has to be encoded into a format computers can use, which means patterns of ones and zeroes (‘bits’) organised into small chunks (‘bytes’). On my computer, each byte contains 8 bits, and a certain pattern of those bits stands for the letter d, another for the letter o, and another for the letter g. Side by side, those three bytes form the word dog.

They really don’t, at least no more literally than brains do. Computers don’t “know” anything about numbers, words, formulas, images, or algorithms. They don’t even know anything about 1s and 0s or bits and bytes: every single one of those things is an abstraction that helps us track the real pattern in how an extraordinarily complicated system changes over time. Computers are cleverly designed arrangements of metal, plastic, and other physical material that, when subjected to certain boundary conditions, will evolve over time in predictable ways that we can then use to model various patterns. They’re physical models in about the same sense that a model airplane in a wind tunnel is a physical model of a full-sized airplane in the open air–they’re just more complicated by far. There’s nothing magical about this kind of arrangement that makes it “real” information processing while the brain (or anything else) is ersatz; that is, information processing just is that kind of stable, predictable physical change. Computers process information in exactly the same sense that brains do and in exactly the same sense that any other physical system does. Computers (and brains) have the virtue of being a combination of complex and stable that lets them process a lot of information across a wide variety of contexts, given appropriate inputs and boundary conditions.

Consider the difference between a modern digital computer and something like Babbage’s analytical engine. There are enormous physical differences between those two systems. The analytical engine was purely mechanical: it took its input via punch cards and stored its internal state via wooden or metal pegs inserted into rotating barrels. Is something like that “quite literally processing information” on this view? Is it encoding data as bits and bytes? Or is that something that only electronic digital computers can do? This strikes me as an obviously silly question: there are ways in which the analytical engine and my laptop are functionally similar, and ways in which they are different. Whether the similarities or differences are more salient depends on what you care about, or what kinds of things you think are important to track. Either they’re both doing information processing, though, or neither of them is–there’s nothing special about electronic digital computers that makes them “real” instances of computation and everything else just a simulacrum. But if the analytical engine can process information despite huge differences in material constitution and operation from an electronic computer, then surely the brain can as well. That doesn’t mean a brain is a digital computer, just that (again) there are elements of similarity between the two, and that the formalism of information theory can be a useful lens for understanding the operation of both.

The court heard that Woods was told during the encounter that some deputies were allergic to bees, and that she replied: “Oh, you’re allergic? Good!”

C-suite executives, meanwhile, are facing the heavy burden of squeezing blood from AI’s proverbial stone. 72 percent of all surveyed execs said their company’s AI strategy is causing them stress or anxiety, 32 percent of whom characterize their stress as “high” or “crippling.”

lol, lmao

If we stay on the current trajectory, I think more-or-less complete collapse by 2060 at the outside.

Definitely not belittlement, but my committee didn’t necessarily agree. Climate scientists are often rather conservative–not so much in the political sense as in the “we think that the middle-of-the-road ensemble model prediction is probably basically right” sense. When you say something like “I think that even the more pessimistic model forecasts are underselling how bad the situation is” they tend to be pretty skeptical. I got pushback, but not really disdain or belittlement. A lot of the new work coming out seems to be vindicating the position I was taking though which (again) sucks because it is very bad.

I had a whole chapter in my PhD dissertation arguing that we were much closer to this than was commonly believed. That was a little more than a decade ago. Sucks to be right.

It’s looking more and more like Trump is going to inadvertently cause a worldwide lurch toward green energy, which will never not be funny to me.

“clicks”*

*latest fad

Eventually, my car from 2010 is going to bite the dust and I’m going to have to get a newer one and I’m preemptively pissed about it. I’ve got an ebike and use it when I can, but I live somewhere that it isn’t practical to use for all transportation and I need a car, unfortunately. 2008-2014 or so is the technological sweet spot for me: my car has power windows, remote start for when it’s 15° outside, and Bluetooth for my music. It also has no screens, no “assistive technology,” all analog controls, and can be worked on by a normal mechanic without special software. It sucks so much that all this shitty tech that makes cars less safe to drive and harder to use/work on is now mandatory. On the other hand, my wife drives an ancient Toyota pick up truck that’s older than I am, and it doesn’t even have power steering. It’s also a nightmare to drive, but one of a different kind (though you can disassemble the whole thing with a screwdriver and socket wrench, which is neat).

reign down

Extremely well, actually. I’ve got one student with a large amount of family in Iran (plus it’s a class on rhetoric and media), so it’s been on our joint radar for a while. A few of them had seen it, but most hadn’t. They’re still broadly critical of the Iranian regime (fine) but the contrast between the communication coming out of Iran and the communication coming out of the US regime–along with the behavior of each–is pretty apparent. They mostly landed on “this would be an amazing piece of propaganda if only Americans could read,” which is a very good take in my opinion. They pretty much all already thought the US was a dogshit empire though because they’re smart kids.

Going to make all my students listen to me read this out loud tomorrow.

It won’t pay for it, but it’ll service the debt they had to go into in order to pay for it for a while. Cutting your grocery bill to free up money to make your payment on your new Cybertruck.

Oh yeah, it’s quagmire time!

Ah, I misunderstood your question. Yeah, most of the overall energy goes into heat for sure. Since the accelerator is using massive particles in its collision, it’s probably not right to say that energy is being transformed into mass: the total mass of the system is conserved, and you’re just “whacking” an antiproton off the iridium nucleus via inelastic scattering. You could get that kind of transfer if the colliding particle were massless (like a photon), which happens with cosmic rays sometimes. It’s harder to control in a number of different respects though, and you’re not guaranteed to get the kind of anti-particle you want. By using a proton, we can guarantee we get the flavor we want, since all the quantum numbers are also conserved.

If you mean does it cost energy to produce it, then yes an incredible amount. The only way we have of making it is to smash particles together in an accelerator at enormous speeds. The LHC at CERN uses protons and, for this kind of experiment, probably some kind of dense metallic atom like iridium (though I don’t know off the top of my head). That takes a truly incredible amount of power to run–something like half the power consumption of a small city. That energy doesn’t disappear of course: some of it goes into the beam, but most of it goes into operating the magnets used in acceleration. The actual creation of the particle anti-particle pair conserves energy just like everything else.

Generally yes, they’re always produced in pairs (which is why it’s so hard to get them to stick around, as they tend to more-or-less instantly annihilate each other). There are some exotic exceptions to this mostly involving neutrinos that we don’t fully understand yet, but this kind of an experiment will generally get you a particle/anti-particle pair. When you depict the process with a Feynman diagram, the symmetry is really evident, which is part of why they’re such great representational tools for this kind of interaction.