In October 2013, I wrote a post arguing that computing was disrupting language and that this was the Mother of All Disruptions. My specific argument was that human-to-human communication was an over-served market, and that computing was driving a classic disruption pattern by serving an under-served marginal market: machine-to-machine and organization-to-organization communications. At the time, I didn’t have AI in mind, just the torrents of non-human-readable data flowing across the internet.

But now, a decade later, it’s obvious that AI is a big part of how the disruption is unfolding. Two ongoing things drove it home for me this week.

Exhibit A, the longest regular conversations I’ve had in the last week have been with an AI-powered rubber duck. Berduck is a bot on Blue Sky that is powered by GPT, and trained to speak in a mix of English and leetspeak. It is likely playing a non-trivial role in driving the Blue Sky craze (Blue Sky is a decentralized Twitter-like protocol funded by Twitter itself in the before-times). I can’t speak for others, but I probably wouldn’t be using Blue Sky much if it weren’t for Berduck.

Berduck is genuinely entertaining, with a well-defined personality, despite only having episodic memory, a strong predilection for hallucinations and confabulation (like all AI-powered chatbots), sharp boundaries around negative valence conversations, and a strong aversion to even the slightest whiff of risk. Despite all these annoying limitations, shared by way more humans than we like to admit, (yes, I am an AI accelerationist, why do you ask), Berduck is already a more interesting companion than 90% of humans online, and I can totally see myself passing the time with one of his descendants in my dotage. The limitations are both tolerable and mitigable, and the benefits quite striking.

[Read more…]

The philosophies of science that I find most compelling, such as Paul Feyerabend’s, tend to argue for methodological anarchy as the characteristic of the most historically impactful science. It is not immediately obvious, but I think this is equivalent to arguing that the best science (and any sort of inquiry conducted with a scientific sensibility) is necessarily permissionless. Anarchic permissionlessness though, does not equal chaotic lack of structure. It is just that structure emerges from the nature of the research, rather than generic procedural templates. Investigations always require protocols, even casual ones, but they need not be derived from some abstract high-modernist notion of a uniform “scientific method.”

Do uniformities in the nature of all knowledge justify privileging particular research methods at all? This is an epistemology question to which I have yet to hear a satisfying answer. In some ways this is the most important practical question in the philosophy of science. Whether or not you think you need permission from an authority figure to do research depends on whether or not you think certain methods ought to be naturally privileged.

[Read more…]

I have long been fascinated by tessellations as metaphors for ways of knowing about and being in the world. A set of prototiles that can cover a world in an exhaustive and mutually exclusive way constitutes something like a theory of that world. The rules of the tiling are the rules of the world. The set of prototiles is the ontology underlying the theory. The size of that set is a measure of the efficiency of your understanding. Recognitions of repeating patterns in the emergent tiling are understandings of specific aspects of the phenomenology of the world. Actually creating a specific tiling by placing tiles on a smooth surface, to create navigable striations, is the praxis of the way of knowing.

To state it in terms of my new favorite frame, tessellations are something like metaphors for protocols of knowing and being. Given the right set of tiles, you can know the world and be in it, in a powerful way. Perhaps this is one way to understand the story of Robert Moses, architect of New York. He tessellated his world with tiles of his choosing.

Ideally, you want the richest, most complex tiling possible to cover a “blank” world, such as the 2d Euclidean plane, to maximally reveal the possibilities latent within it. Yes, you can cover the Euclidean plane with a boring regular grid of square tiles, but you can also cover it with strange aperiodic tilings, and in some ways, the latter constitute a truer “theory of the plane.” The intuitively appealing principle that you should look for the richest possible tessellation is a kind of dual to Occam’s razor. Instead of choosing the simplest explanation that covers a given world of facts, you choose the covering that produces the most complex world of facts. Ideally, the maximally complex set of facts. Instead of solving for explanatory parsimony, you solve for generative profligacy.

One proxy for such maximality is Turing-completeness, and at least some (all?) aperiodic tilings, like Wang tilings, are known to be Turing complete. Jed Yang published a PhD thesis in 2013 about the computational aspects of tessellations, and also connected tiling-based computation to Turing-complete cellular automata, such as Wolfram’s Rule 110. Googling around, I also found this fascinating presentation by Kathleen Lindsay about playing Conway’s Game of Life on an aperiodic tiling. It seems like tilings and cellular automata are equivalent ways of understanding universal computation, and at least to me, these spatial processes seem more intuitively appealing than infinite tape machines or the lambda calculus. And of the two, I think I prefer tessellations over automata, since the computational process is embedded in the texture of the space itself, as opposed to a 0/1 switching process playing out on it.

As you may have guessed, aperiodic tessellations have been on my mind lately because last week the first aperiodic monotile, the “hat” (an “Einstein” tile, named for the German ein stein, or one stone, rather than the physicist) was discovered. It is not quite a monotile since you have to use it along with its mirror image to aperiodically tile the plane (the blue vs. yellow instances in the picture below), but still, this is a fascinating leap. The last best attempt, the class of two-tile solutions known as Penrose tiles, seemed like the End of History of Tessellations to me, but apparently we had a chapter left. I suspect this is the end though. I somehow doubt we’ll get it down to a single kind of tile without the mirroring cheat (I wonder if anyone has proved that a single tile, without mirroring, cannot tile the plane aperiodically).

Recreational mathematicians are going a bit nuts with this discovery, and I’m 3d-printing a set to play with (I’m using this model) as we speak.

I’m especially intrigued by the idea of painted kites. The hat can be decomposed into 8 kite shapes that can be “decorated” in a way that the resulting hat tilings create strange aperiodic maps. These feel exciting in the same way the original pictures of the Mandelbrot set felt exciting in the 1980s. Unfortunately, my 3d printer is single-nozzle, so I can’t print these easily.

While I’m nerdsniped by these tilings and painted tilings, I’m not enough of a mathematician to truly explore them in any technically deep way. But I suspect that after the late J. G. Ballard, whose complete short stories I just finished, I’m quite possibly the person who has spent the most time thinking about tessellations as world-narrative metaphors, so let me talk about that instead.

The goal: get to post-Ballardian ways of thinking about our End of History condition, via aperiodic tessellations (plus noise).

[Read more…]

As a kid, through most of middle and high school, I got good grades. I stayed comfortably near the top of the class without working too hard, and more importantly, without explicitly aiming to be there. I got good grades not because I “studied” conscientiously, but because I enjoyed most subjects enough that chaotic nerd energy was more than enough to coast near the top. That easy ride ended in college where that kind of marginal focus on grades was only enough to put me somewhere in the middle of the more gifted peer group. I didn’t work any harder, but I didn’t get as unreasonably rewarded for it.

But my lazy, easy ride through grade school had already made me relatively immune to validation from grades. I had become incapable of working with any sort of discipline towards good grades. I had neither contempt nor respect for good grades. I was just indifferent to them, and addicted to the less legible fruits of nerding out. I was only motivated to do well enough that grades would never get in the way of things I wanted to do (now you know where my philosophy of mediocrity comes from). I neither tried to get straight-As, nor chafed against expectations of getting good grades. I neither disappointed my parents, nor made them exceptionally proud. Possibly because I was neither the sort of straight-A’s talent who is actually in the running for racks of prizes, nor the sort of maverick intelligence that schools are particularly good at detecting and destroying with extreme prejudice. I was the sort of kid who is not just indifferent to schooling, but the sort of kid schools are indifferent to. We’re neither good enough, nor bad enough, to be worth exceptional attention. I passed through the educational system like an unexceptional neutrino, all the way through to a respectable PhD.

Which isn’t to say I didn’t work hard at my learning at all. I just worked hard in the sense of nerding out over topics that I actually enjoyed, mostly by reading far beyond and outside of the syllabus on topics I got nerdsniped by, so that I learned what was in the syllabus almost in passing. I almost never encountered a subject I couldn’t get nerdsniped by. In college and grad school, this pattern mostly continued, though I of course discovered more subjects I had very little aptitude for and didn’t get nerdsniped by. I eventually got promoted to the level of my academic mediocrity, never having learned to work hard for grades along the way. This feels a bit tragic to me now.

[Read more…]

A quick post about a thing I’m up to. This summer, I’ll be running a program called the Summer of Protocols (SoP) that will fund a bunch of full-time and part-time researchers to think broadly about, tinker with, and write about protocols. The scope is broad. Everything from climate and cultural protocols to TCP/IP and blockchains is in scope, and we’ll be focusing on both technology and science/humanities aspects of protocols. If you think you might be a fit for the program, consider applying. Here is a twitter thread with more details. We’re holding a Zoom town hall where you can learn more and ask questions on Friday, March 10th, 1530 UTC.

The program is funded by the Ethereum Foundation, and the goal is to connect the conversations around protocols that happen in the crypto world to similar conversations in other places.

The program grew out of a pilot project I led over the last 3 months, on the foundations of protocols. My collaborators and I just released a draft of the pilot study, The Unreasonable Sufficiency of Protocols, meant to prime the pump for the SoP. So if you’re interested, I recommend reading that. I also just went on a podcast to talk about the SoP that you may like.

I can’t remember where I read the theory, but apparently the salinity of our bodies matches that of primordial seas, so in a sense we never really left the oceans. Our micro-organic aquatic ancestors simply constructed meatbag spaceships with artificial life-support aquatic environments inside to explore beyond their oceanic home world. Much as we might construct generation starships in the future to travel to Alpha Centauri. Air-breathing multi-cellular life forms constitute the space program of primordial single-celled life. They get the missing resources of their primordial environment through clever artificial means, just as we might eventually mine our essential minerals from asteroids or gas clouds we fly by in outer space.

One of the effects of this evolutionary history is that all air-breathing life has to seek out perhaps the most important chemical that’s ubiquitous in the oceans but not trivial to find on land: salt. Salt-seeking is one of the most fundamental behaviors of terrestrial life. Animals in the wild seek out salt licks even at great risk of predation. Humans with salt deficiencies have serious problems, and beyond a point of salt deprivation, you die.

This evolutionary history struck me as an interesting metaphor, and a superior alternative to the more familiar metaphor of “coming up for air.”

Oxygen, whether extracted from the air or water, is critical on very short time-scales, even by the standards of the lifespans of micro-organisms. You die in minutes without it. So the idea that you might be so swamped by work that you need to “come up for air” is evocative but not quite coherent. I can think of coming up for other kinds of macro-nutrients like water and food at the time scales of uninterrupted work (hours/days) but not air. If you’ve ever actually had to “come up for air” (I did a couple of times as a kid while learning to swim) you know it’s not some sort of restorative break. It’s a panic-inducing emergency of a sort that is not metaphorically like anything else.

But salt-seeking… now that’s an interesting metaphor that operates on a whole different time-scale. If you don’t lose any salt to perspiration, you can go without salt intake for days to weeks (it’s not so much a consumable nutrient as a sort of electrolyte concentration thing I guess).

[Read more…]

The planet Kinsoro, in a star system on the edge of the Bracket nebula, had a strange climate marked by long alternating periods of wonderful and terrible climates. Each phase of the climatic cycle (whose origins were mysterious) lasted approximately 120 Earth years. The strange climate had given rise to an equally strange ecology early in the planet’s history: The more complex species were all polymorphic, typically existing in two forms adapted to the two regimes (and occasionally as many as six, adapted to minor climate epicycles). The dominant species, the Kinsorans, existed in two adult forms, which Kinsoran biologists called the S-type and the I-type. The two types had clear but minor physical differences, but more importantly, dramatically different behavioral and cognitive dispositions. While both types could survive both climates, the S-type flourished in the wonderful half-cycles, while the I-types flourished during the terrible half-cycles.

Kinsoran biologists calculated that the environmental variation ought to lead to a slow 240-year limit cycle in the S/I ratio between 80:20 and 20:80, but in practice the ratio showed much more extreme variation: swinging between 99:1 to 1:99. The reason was not hard to discover; during each half-cycle, the dominant type would violent repress and slaughter the subordinate type to near-extinction.

[Read more…]