December 2022 – belikeme

Doodle Abstracts 1

I doodle. Sometimes I elaborate my doodles, rarely I implementish them. As with Art Project 1, I suspect there will be a series of these, thus the numeration. Here are some ideas from doodles and for doodles.

Doodle Abstract 1. Superlongtermism. A red dwarf star with a full-capture Dyson cloud of techne orbiting. Actually, a just-sub-red-dwarf. Some cloud elements are projecting light nucleus plasma beams into it to keep the fusion going under positive control. Maybe it’s not a cloud, but an actual Dyson sphere, situated to provide a 1 g surface. The sphere captures all the star’s material output, recycling any light nuclei for fusion and retaining the heavier nuclei as a resource. The main material resource is the original stellar system, which has been re-assembled as the sphere, with any remaining atoms stored in a slag planet and its moons, and an ore planet and its moons. Any interstellar material traveling through is captured as a resource. In the far future this could be the motivation for war.

Doodle Abstract 2. Annealed Oulipo. For several co-authors, each author writes each paragraph of the same paper according to the rule. The papers are submitted to a rarity checker, and a ranked choice vote is taken of what the best rare words are. These are then required to be used in the final paper. The rewriting process requires each co-author to be assigned their share of parallel paragraphs, their task being to anneal them together, possibly producing 1 to 3 paragraphs, but including all the concepts and any required rare words. This becomes the rough draft. Then some method of passing the document through the hands of each of the co-authors at least once is used.

Doodle Abstract 3. Finite Tilings of Spherical Surfaces. The Platonic solids as avatars of the regular tilings, for the surface of a sphere rather than the infinite Euclidean plane. Definitely periodic. OTOH, you can randomly populate the surface of a sphere with vertices, and then fiddle with them until they are very nearly equidistant. With 4, 6, 8, 12, 20 vertices this will (probably provably) evolve to the Platonic solid vertices. With other numbers would you get the equivalents of the uniform tilings etc.? With yet other numbers could you set things up so that there are two different classes of equidistant points, either at the same distance even though there are two classes, or two different distances, etc. Or perhaps rational number distances only, optimizing for standard deviation and maximum deviation somehow. Would you get situations where each distance and therefore edge and therefore tile is different, or would you have subsets of equal lengths? An aperiodic tiling of an infinite number of tiles whose geometries vary according to some rule.

Doodle Abstract 4. Peer Reviewed Paragraphs. Citizen (and Establishment) microknowledge. They would have the equivalent of DOIs, and AIs would be constantly searching the available literature for paragraphs and sentences thereof easily transformable to the DOI version. If every artist is a thief, every sentence is a plagiarism. This is a mechanism to organize the data and the implications so that the wealth can be spread around justly.

Doodle Abstract 5. AI Old Pictures. Scan a bunch of decorative alphabet letters and engravings and such from old books. Use these as your training data for prompt-driven image derivations. Analyze them down to the splines, so you can manually or algorithmically optimize and tweak them.

Future Fitness

Any particular brain admits but a subset of instincts and behaviors that might arise in the totality of brains. Some brains admit an instinct for killing rabbits by neck-snapping, others can learn to retrieve dead waterfowl, some can master trapping by reading books. Some human instincts, selected in environments that no longer obtain, are today of potentially dubious, even negative utility. After generations of supersession by novel instincts or learned behaviors, these detrimental ones will perhaps atrophy or even disappear. The abilities needed for future advances may depend on genetically embodied learning instincts affording origination, transfer, exploitation, and defense against memes, the molecules of cultural transmission.

Obviously, different brains are better or worse than others at different subjects. Consider the ability to manipulate quantum mechanics and/or general relativity, the current extremes of our ability to comprehend what appears to be basal reality. Facility with abstract conceptualization and imagination, aligned with analogical reasoning is presently what is needed to excel at these two learned behaviors. Various people hit abstraction ceilings at various levels, for various reasons, maxing out at, say, algebra, trig, calculus, programming, etc.

Although able to calculate abstruse physics results, our brains were originally selected for handling the apparent physics of the mundane world, utilizing neural networks to rapidly simulate differential equations and game theory (walking, throwing, evading). Alas, these networks do not provide an instinctive ability to do integration by parts or find roots of unity. One reason math and science are “so hard” is the difficulty in setting aside instinctual physics for Newtonian, then setting that aside when necessary to exploit QM and GR.

Imagine, then, a world in which people have been uplifted and educated to the point where all brains able to handle intellectual physics can be so trained. A world in which the educated mustn’t constantly roll their eyes at or waste their time with the pitiable failings of the populist unfortunate. A world in which the greatest proportion of people who can rise beyond the highest abstraction ceilings is available and eager to so rise. Here I am unavoidably US-centric: metaphorically this might be something akin to bestowing upon all, worldwide, the privileges that were available mainly to white male geeks of the baby boom generation. Potentially this think force would be able to marshal the cognitive power needed to advance civilization towards an ever-improving environment for ethical self-actualization.

In such a world, fantasies such as establishing off-world settlements would be much less burdensome to both the stayers and the goers. These expanding far-flung societies might even become self-sustaining eventually, perhaps subject to selection pressures imbuing their descendants’ brains with an instinctive physics more suited to their environment, after a hundred thousand years or so even displacing the innate “cartoon physics” of yesteryear (to them: to us, presentyear) with variants making it easier to conceptualize QM and GR at their parents’ knees rather than in grad school.

Such an advanced intellectual basis might promote transcendence of currently unimaginable abstraction ceilings, helping our descendants get even better at as-godness.

Bond, Antibond

In quantum chemistry, electrons don’t orbit the nucleus the way that planets orbit stars. Rather, the odds of finding an electron at various points in the space around the nucleus are what’s calculated. For an atom in its lowest energy state, this probability distribution has a spherical symmetry, with the highest probabilities found in a shell around the nucleus, but with positive, if ever smaller with distance, probabilities for finding it anywhere in the universe. This arrangement, harkening back to the Bohr atom which proposed a more planetary style of motion, is called an orbital, to distinguish it from an orbit. I guess orbitoid sounded too kooky.

For higher energy atoms and in chemical bonds, the shape of the probability distribution is much more complicated. Nuclei, being positively charged, tend to repel each other, but some arrangements of orbitals and nuclei position much of the electron density between two positive centers, such that they shield the nuclei from each other and attract them somewhat. This produces the chemical bond and is known as a bonding orbital. Now, you need to balance positive and negative charges in order to have a neutral molecule, so one thing you do when setting up your calculations is just keep adding electrons (according to the rules) to the orbitals defined by some arrangement of nuclei until everything balances out.

It turns out that only the lowest energy orbitals are bonding. There are higher energy orbitals in which the rules place the electrons, but these do not shield the nuclei from each other the way they do in bonding orbitals. These are called, perhaps unsurprisingly, anti-bonding orbitals. Let’s say you have an arrangement with one bonding and one anti-bonding orbital, and you need to fill them with four total electrons to achieve electrical neutrality, two in the bonding orbital, and two in the antibonding one (one usually deals with electrons in pairs when doing such things). It turns out that “anti-bonding orbitals are more anti-bonding than bonding orbitals are bonding”. Thus, not only do you not get net chemical bonding, you don’t even get a sort of “meh” asociality between the nuclei. You get an actual repulsion.

All this is to try to set up a metaphor, or at least to create a conceptual space in which to seek a metaphor. In the popular science headline clickbait webiverse I notice that there appears to be a consensus that people tend to dislike the idea of losing a hundred bucks more than they like the idea of finding a hundred bucks. It’s a cognitive bias. No doubt there is endless evolutionistiod, genetic determinism type jibberjabber about how there must have been a survival benefit to thinking this way blah blah blah. Even if strong arguments could be made in that direction, it’s certainly not obvious that any such benefit obtains in our current social environment. This phenomenon seems to be consistent with, or perhaps even instrumental in, promotion of negative-sum rather than positive-sum outcomes.

I wonder how hard it would be to find additional examples of similar phenomena, perhaps within thermodynamics itself (my instinct would be to look for entropy-related arguments), but also within game theory, computational neural networks, or even such fascinating novelties as “Wolfram Physics”. I should try to scour various philosophy-of-mind works such as Daniel Dennett’s oeuvre to see if I can find a trace, or even explicit exposition, of this trope. Actually, before that I should follow the clickbait and familiarize myself with the actual work.

Art Project 1

I have numbered this post, as I imagine I will do more of these. I was actually going to schedule two normal posts, which I will get around to, but I have a new scanner, pottered around with it, and thought to do a Thing or two. This post comprises this paragraph and two Things.