Well yeah, it's big numbers! These sims (at least the ones I did) can never be compared to earth, because they make assumptions that start from maybe billions of years of simulated time, and then further make tons of ridiculous assumptions about life based on our observations from this one planet, and then they massively compress our physical reality from atomic and sub-atomic interactions to insanely granular abstractions in these sims.
("How much energy would it take to 100% accurately simulate our universe from the big bang to now" is a pretty interesting thought experiment though!)
I like that interesting period on Earth from maybe 4 billion years ago to maybe 500 million years ago, where we went from amoebas learning to duel to arthropods learning advanced dueling techniques. So any given one of my sims assumes a minimum of 10 billion years of priors, and assumes that it can compress millions and millions of generations and population diversity into a much smaller number.
I've had single sims where basic building blocks go from drunken amoeba to the earliest of pretty intelligent arthropods - so that's 3.5b years of evolution. But all sims are a massive tradeoff between fidelity and duration (fast/good/cheap), and the sims I like the most are the ones that cover an era that creates emergence in even a single aspect of the simulation. I guess most of what I'm interested in simulating is vaguely on the order of 100m years of evolution, over vaguely 7 days of real-time simulation?
Really interesting topic. For about 5 years, I spent much of my hobby time writing a-life sims and messing with them. I wanted an "aquarium" that looked interesting that I could have running 24/7, constantly evolving. I guess I wrote maybe 10 sims from scratch in Unity, Java, Processing, and p5.js.
I have had (limited) success in getting emergent control and coordination behaviours. A couple examples:
I played with a "battleground" simulation, inspired by the "Gladiabots" game, where the entities were NN-based fighting robots organised in teams (4v4, 20v20 etc). Random sensor systems, random fire control systems, signal emitters, etc, hooked up to random neural nets where winning teams procreate and evolve/mutate. Over a LOT of generations (maybe like 20,000+), I'd see teams evolve what looked like coherent tactics, including teamwork, role specialisation, and time-based tactics/formations.
I had a "jellyfish" simulation, similar premise, with small aquatic animals of various "species" who could speciate, etc. The aquarium had qualities like light levels, food production, temperature zones, etc. You always have to be careful to not anthropomorphize too much, but I would clearly see battlegrounds develop and evolve, and successful species rise and fall with behaviours that (again over many thousands of generations) initially look like drunken sailors but evolve into what looks like basic behvaiours, at least at the level you may see in small multicellular organisms.
In my experience, getting emergent behaviours needs a difficult-to-optimize balance between system complexity and system stability. Every neuron, sensor, output, or attribute you add to the sim can cause a super-exponential growth in the "problem space" of your denizens. Behaviours that require any form of coordination take so much longer to evolve than uncoordinated behaviours because you tend to need to "get lucky" where both a signal and response evolve/mutate at the same time.
Another issue is that feed-forward networks are not great at temporal processing. Many basic behaviours require more temporal processing than I expected when I started building these things. I got these to evolve by having features like memory nodes, timer nodes, and recurrent nodes, but if I were to build another one, I'd experiment more with LSTM-like neurons and other recurrent/memory-based architectures.
In my experience (and with my horrendously under-optimised codebases), the amount of evolution you need to simulate is super-exponential as you grow the complexity of the sensors, outputs, and neural nets. In my playing around, doubling any single dimension almost always led to more than a 4x increase in the time you need to see new emergence. In my most complex sim, it got to the point where even tweaking a basic variable like temperature or day/night cycle would need me to run the sim for at least a day to properly observe the impact, and sometimes more like a week or two. I never really knew when the "evolution ceiling" was hit, which was part of the fun for me -- I love seeing a new behavior suddenly appear.
This ALIEN project is really interesting and it has a lot of cool ideas. I haven't played with it yet but I think it's going to be fun! The project I have been keeping an eye on is Neuraquarium, which has similar design goals to what I have built in the past.
I get convergent evolution, but among large (let's say 10g+) organisms, I'm not aware of convergent evolution resulting in compatible species (with fertile offsping) that would not otherwise have been compatible?
I'm super rusty on this topic, but if there is theory that large organism actual DNA-level speciation (resulting in individuals who cannot reproduce together) has eventuated to convergence back to a new species (who can reproduce together with fertile offspring), I'd love a source. I definitely could have very rusty knowledge on this but it seems intuitive to me?
Sorry, as stated above I misunderstood "There is no such thing as a tree." to mean that OP though taxonomic trees were fundamentally broken in some way, I misunderstood OP!
That said, if we had perfect knowledge of the speciation process over the years, would our taxonomy not be extremely close to a perfect tree, where every node has 2+ branches, and branches don't converge to being species-compatible for breeding?
I get convergent evolution, but among large (let's say 10g+) organisms, I'm not aware of convergent evolution resulting in compatible species that would not otherwise have been compatible?
I'm super rusty on this topic, but if there is theory that large organism actual DNA-level speciation (resulting in individuals who cannot reproduce together) has eventuated to convergence back to a new species (who can reproduce together), I'd love a source. I definitely could have very rusty knowledge on this but it seems intuitive to me?
I definitely understand that the majority of taxonomies are problematic for the reasons you cited. When OP said "no such thing as a tree", I thought OP meant a taxonomic tree, not a literal plant tree, hence my example of binary numbers! Thanks for clarifying.
That said, taxonomic groupings can have both wide consensus and be useful, can't they? (Hand on chin... monotremes? hominids??)
> All taxonomies are broken, full stop. Your categories are gonna be completely wrong and everybody’s going to argue over every single thing. There is no such thing as a tree.
Is this a serious statement? If so, wouldn't it be incompatible with the theory of evolution? Would an alien taxonomy of human binary numbers not be a legitimate tree?
Why not both? The modern dismantling of antitrust regulation and enforcement is a significant driver of both. It's gotten way worse now than ever, to the point where private equity firms like Vanguard and Blackrock can be the biggest shareholders in direct duopoly competitors (KO and PEP are an example), or where investors and producers are permitted to fully horizontally and vertically integrate (AMZN is an example). And media (among many other industries) has been permitted to consolidate to ridiculous levels that would have been unimaginable to the architects of early-mid 20th century antitrust laws even on a practical level -- an Amazon-level entity would have been literally impossible to envision in the early 20th century because companies just practically couldn't get that big with manual technologies.
The largest US company in 1929 had revenues of about $1.5bn - roughly $30bn in 2023 dollars, which wouldn't even put it in the Fortune 100 today. Walmart today has revenues of about $600bn, which is roughly equivalent to 6x the top 50 companies in the US in 1929 combined!
It's mind-blowing to me that economics is still using these ridiculously old, flawed models whose assumptions are less true than ever. To me, econ 101 should be teaching how irrational consumers and lawmakers are, what state capture is, monopoly-seeking behaviours as the primary modern driver of profit growth, etc...
As a middle-aged long-time smoker who quit smoking immediately upon buying a vape several years ago, and having quite a number of friends who did the same thing, I'll point out a number of issues I have with this measure.
- I do see a lot of kids vaping, just like you used to see a lot of underage kids smoking or drinking. The solution there was not to ban cigarettes and alcohol for adults, or require a prescription, it was suitable age-appropriate regulation.
- I would not have seen a doctor to try out vapes. Under this regime, adult vapers like myself will now have to see a doctor, get a prescription which will likely need re-prescribing every 1-3 months, and most doctors won't renew prescriptions over the phone or without charge. My local GP charges me $80 a visit, that's now adding potentially $320-$1k/year to the cost of vaping, not to mention the additional fees that will be charged by pharmacists and whoever the approved vendors will be. (I also expect the approved products to kinda suck and certainly not include the only two flavours of vape that I enjoy.)
- This thing about "big tobacco" somehow being the beneficiaries of Australian vaping seems like nonsense to me -- at least where I live (VIC), the overwhelming majority of vapers use dodgy off-brand imports of disposable vapes that are sold over-the-counter, with nicotine, by unethical tobacco shops (extremely common around my area). For more "sophisticated" vapers who use refillables with hand-mixed flavour and nicotine, the nicotine is imported from somewhere like NZ from a reputable nicotine supplier, the flavours are purchased locally, and the devices are mostly from quality Chinese manufacturers. I have seen maybe 1-2 Juul's in the past five years, but I live in regional Australia so maybe things are different in the city. Overall I think big tobacco seems to have a lot more to gain from vape prohibition here, as do the big pharma companies who sell OTC nic replacement products (e.g. J&J and GSK).
- I have yet to see compelling evidence that vaping carries greater risk than other NRT products, modulo the Acetyl Acetate / "popcorn lung" issues that as far as I can tell originated almost entirely from dodgy THC vapes (very different product to nic vapes).
Personally, I think the better solution here would have been to regulate vapes like cigarettes. They are beneficial to long-term adult smokers who have tried and failed to use other methods to quit. My wife literally tried everything including the crazy prescription anti-nic drugs and never was able to fully quit until switching to vapes.
It's clear there is a huge under-enforcement problem right now (e.g. local shops selling OTC nicotine vapes, which is already illegal, and selling them to kids, also already illegal) but giving them the same regulatory framework as cigarettes, or maybe only allowing chemists to dispense them to 18+ users would have been preferable here.
I love my vape and what it did for me and my friends. I don't really plan on quitting vaping any time soon because based on my readings of the mainstream medical papers, there isn't sufficient evidence of personal risk for me that comes even close to other legal risky things I do like eating fast food and drinking wine.
I just dislike being in a position of now either needing to illegally import, or pay a huge effective tax to the medical/pharmacy industry, to keep vaping, as a responsible adult.
That said, I would support a simultaneous "grandfather" ban on both cigarettes and ALL other nicotine products along the lines of cigarette bans where the products are banned for people 18 and younger, and that ban grows by one year per year. That's a better approach than what Australia is doing.
I'm unsurprised it came to this, you could see the media coverage of vaping growing increasingly shrill / moral panicky over the past years, in contrast to the actual science on nicotine harm, where vapes may be on the "bad" side of other NRT products, but nowehere near the harm of cigarettes. It follows an eerily similar media narrative to what I've seen in other western countries (maybe other countries too, but I mainly read English media).
I would not be surprised at all if both big tobacco and big pharma lobbyists are high-fiving themselves over this.
("How much energy would it take to 100% accurately simulate our universe from the big bang to now" is a pretty interesting thought experiment though!)
I like that interesting period on Earth from maybe 4 billion years ago to maybe 500 million years ago, where we went from amoebas learning to duel to arthropods learning advanced dueling techniques. So any given one of my sims assumes a minimum of 10 billion years of priors, and assumes that it can compress millions and millions of generations and population diversity into a much smaller number.
I've had single sims where basic building blocks go from drunken amoeba to the earliest of pretty intelligent arthropods - so that's 3.5b years of evolution. But all sims are a massive tradeoff between fidelity and duration (fast/good/cheap), and the sims I like the most are the ones that cover an era that creates emergence in even a single aspect of the simulation. I guess most of what I'm interested in simulating is vaguely on the order of 100m years of evolution, over vaguely 7 days of real-time simulation?