Posts
Comments
Which one? I hope it's not the one where you have to put chocolate, because this is the most crucial instruction.
it's no biology lab
I'm afraid you're overestimating how well biologists follow the safety procedures. I wouldn't be surprised if we all had fluorescent bacteria in our guts.
Oh, that's a really good point. Actually, it might be common for chemists to work with panels of related molecules, while in clinical trials they only work with one purified drug candidate. This makes it less likely for them to discover things by accident. Surely a piece of the puzzle!
Sure, all these stories totally sound like urban legends, but the sweeteners are out there and I don't see how they could have been discovered otherwise (unless they were covertly screening drugs on a large number of people).
That's a great question, this is totally mysterious to me. There are a lot of examples of people putting thaumatin in transgenic fruits or vegetables (and somehow in the milk of transgenic mice because there's always one creepy study), but I don't know why it hasn't been commercialized. It sounds like superfruits would make a nice healthy alternative to palm-oil-and-chocolate-based comfort foods. Maybe it's a regulatory problem?
That sounds exciting! I hadn't seen Elisabeth's comment, I just wrote a reply. Do you think there are modifications I should make to the main text to clarify?
That sounds plausible, but I've not looked into the empirical research on that topic so I can't tell you much more!
(Sorry I missed your comment)
Here by "reproduce" I just meant "make more copies of itself" in an immediate sense (so reproductive fitness is just "how fast it replicates right now"). For example, in Lenski's long-term evolution experiment, some variants were selected not because they increased the bacteria's daily growth rate, but because they made it easier to acquire further variants that themselves increased the daily growth rate. These "potentiating" variants were initially detrimental (the copy number of these variants decreased in the population), and only after a long long time they took over the population. So, according the definition of reproductive fitness I used, they lead to a lower reproductive fitness – the reason they were eventually selected for is not that they're good for reproduction, but that they're good for evolvability. Of course, you can say that eventually they increased in copy number, but that would be defining "reproduction" in a different way, that I find less intuitive.
Now, is that other definition (how gene copy number increases over the long term) what evolution ultimately selects for? I'm not sure. To quote Kokko's review on the stagnation paradox:
"Trees compete for sunlight and attempt to outshade each other, but when each tree consequently invests in woody growth, the entire forest must spend energy in stem forming and—assuming time or energy trade-offs—will be slower at converting sunlight into seeds than a low mat of vegetation would have been able to. Every individual has to invest in outcompeting others, but the population as a whole is negligibly closer to the light source (the number of photons arriving in the area is still the same). This is why in agriculture, externally imposed group selection to create shorter crops has improved yields."
She gives other examples. In these cases, the number of individuals tend to decrease over time, even in the long run.
You're right. Honestly I wouldn't be able to talk about this in detail because this is getting far from the things I know best (full disclosure, my own research is on bacteria). The few papers I've cited give some general patterns, and my general point was "things can go in many different ways depending on the specifics, and even the well-known Bateman principle isn't universal".
That's unfortunately all can do: there's a whole world of things to say about how sexual dimorphism actually develops in metazoans, but it takes years of learning to get a deep understanding of what's going on.
Definitely post the papers you're thinking about! If you feel like making a new post about that, I can't encourage you enough to do it. This post was by far my most successful, so it looks like a lot of people are interested in the topic. I'm sure many people would enjoy your contribution (at least I would).
As for the Red Pill thing, I kind of regret mentioning it – I just thought it was funny, but it's not really that funny or useful. Maybe I should edit it out.
What do you mean by curating? So far I've tried to answer the questions and objections when I saw them, are there some I've missed? (Obviously I don't pretend to be able to answer everything). Also, do you think there are some clarifications that I should add to the main text?
I would guess that when organelles are inherited from both parents, the traitor organelle is disadvantaged by its burden on the host, but advantaged by it's ability to be the predominant organelle in the offspring. If the cost-benefit is favourable, then the traitor organelle will take over. OTOH, if only one parent transmits the organelle, the advantage disappears but the burden remains. So I'd expect that it makes it more difficult for traitor mitochondria to invade. Hopefully that makes sense!
Not quite, if it's less efficient at doing the normal mitochondria work, it puts a big burden on the cell, who is then less likely to reproduce.
Thank you for spotting this, I fixed it.
It's not so much the different types in themselves that prevent competition, but having multiple types make it possible to have a mechanism that forces all organelles to come from only one pre-selected parent. If all organelles come from the female, then a rogue mitochondria cannot take over by making more copies of itself or by poisoning other mitochondria, because the only way to make it to the next generation is to be in the female gamete, period. In other words, there's not much an organelle can do to increase in frequency, aside from improving the overall fitness of the organism. Does that make more sense?
The n°1 reason why I said not mention fungi is that I'm absolutely not a mycologist and I wouldn't be able to talk about them. So I greatly appreciate that you do it! Typically, I had never heard of glomeromycota, despite them apparently being involved in symbiosis with 80% of plants. I like to think that I have a decent understanding of the living world, and then I'm constantly reminded that I don't, and probably nobody does...
According to this paper, the "root" factor is how much effort each parent invests in caring about offsprings, as in some species the male is the primary caregiver. But that's really hard to measure and check empirically, so they instead measure "the maximum number of independent offspring that parents can produce per unit of time", and they find very good agreement with which sex faces the most intense competition.
On notable exception is the hippocampus, where the males both face intense competition and invest more resources in the offspring. Because of course it had to be hippocampi.
The "random sampling" that causes genetic drift is applied once every generation, asexual or not, so the optimal number of types depends on the ratio of generations that are asexual vs sexual. The Constable & Kokko paper has a mathematical model to quantify how many asexual generations you need for 2 being the optimum, and it turns out that most isogamous species are well into that regime.
That being said, you're entirely right when you ask "why is the equilibrium 2 instead of 3 or 5 or different for different species?" – Constable's model and empirical data is only for isogamous species like baker's yeast. It seems plausible that our isogamous ancestors were in the same regime, and then anisogamy evolved and kind of locked us into a 2-types configuration. But that's mostly speculation, I don't think we have any clear empirical data that confirms this hypothesis. That's still open to investigation.
Another thing I didn't mention is that the organelle-competition hypothesis naturally leads to 2 types, so it could simply be that.
In their book on social dominance, Sidanius and Pratto make a relevant point: first, they cite a bunch of audit studies where researchers send fake resumes to employers, and find a marked bias against employing African American. Then, they point to Gallup polls asking people whether African American face any discrimination, and almost half of African-Americans themselves say they don't. Same for discrimination in justice or housing. So, when the book was published back in the 90s, many black people didn't believe in racial discrimination, even though it affected them personally in their life.
This means that people's perception of discrimination is not so much influenced by lived experience, but by what the dominant ideology is at a particular time. Back in the 90s, racial discrimination wasn't emphasized in the dominant discourse, so people thought it wasn't very important. In that case, standpoint epistemology just entrenches the dominant beliefs.
Another example: today, there's a vast body of research showing large discrimination against men when applying for housing. If you've ever applied for a place to rent, this has affected you personally favourably or not. But were you aware of it? If you did an online survey asking men what discrimination they face, how many of them would bring up housing discrimination? They don't know about it, because the dominant ideology doesn't talk about it.
I see this as a critical failure of standpoint epistemology (and the "lived experience" approach in general). Here is a 2011 survey where white Americans claimed that white people face more discrimination than black people. I don't think this gives us any valuable information about how much discrimination white people actually face.
(I do see the value of lived experience for hypothesis generation, of course.)
That single broader idea belongs in a single paragraph. Do not split ideas unnecessarily; and certainly do not combine them.
That's interesting, I usually don't think about this when writing. I will in the future.
On using words precisely: I find it more useful to think about how the reader will use the text to make an inference about what's going on in my head. Of course words have official labels that say what they are supposed to mean, but in pratice what matters is how you think I think, and how I think you think I think (you'll recognize a Schelling point). This may be correlated with the dictionary definitions, but it doesn't have to. For example, the word pratice doesn't exist, yet you can understand the meaning of this paragraph just as precisely as if I had written practice. Maybe it's just my experience, but thinking in this way makes writing feel less constrained.
Could there be a signalling component? Nobody you see online would ever be in favour of conversion therapy, so there's no risk for you to be mistaken for one of them. The ideology where one excludes anyone who doesn't support gay rights become the baseline, the least sophisticated ideology, so it's tempting to be a meta-contrarian and argue against it, which signals intelligence and freedom of mind. But IRL, you see that there are pretty homophobic people around (could be some family members at the Christmas dinner), so being a meta-contrarian is no longer an option, as it would just signal intolerance.
I agree on the point that open source software doesn't have to be more secure. My understanding is that they are less likely to send user data to third parties as they're not trying to make ad money (or you could just remove that part from the source). For the exploits-finding AI, I can only hope that the white hats will outnumber the black hats.
I'm also a postdoc, and my institution more or less requires having a smartphone because you can't do anything without their proprietary 2-factors authentication. The other proprietary thing that seem mandatory is Zoom, have you found a way to escape from it?
This is just my humble opinion, but I found this post hilarious.
I see your point, and you're right. Data leaks from big companies or governments are not impossible though, they happen regularly!
Android is (partially) open source but it's not "free as in freedom", which is a technically narrower thing: https://itsfoss.com/what-is-foss/
I agree, though it depends on whether rational design of genomes is even possible, and can do at least as well as natural selection. Can we ever come up with something like an ATP synthase? (Tbh, just maximizing the traits we know about may be enough to stay in the game for a while)
Fixed, thank you.
Could an AI secretly make money on Onlyfans?
I was thinking about how a misaligned AI with write-access to the Internet could gain power without anyone noticing. It looks like porn would be a decent option. It's definitely easier than ordering peptides that turn into nanobots, and less noticeable than blackmail. All the AI has to do is generate realistic pictures of a hot girl and fish for horny subscribers. And then, when you've got money, you've got power.
If the pictures are convincing enough, it would be very hard for anyone to figure out it's fake, especially because it's normal for models not to give any identifying information.
Maybe we can actually try to do it. Just pretend we are an AI, and see how much money we can raise. If it works, it can be a compelling way to demonstrate that there are ways for AIs to silently gain power using only an Internet connection. If it doesn't work, well, it's good news: the world will not immediately end due to porn.
Unfortunately, I don't have any skills in AI-generated stuff, so I wouldn't be able to do it on my own. But if someone here knows an easy way to produce realistic pics of a fictitious model, maybe we can start something.
I'm only seeing this now, thanks for all the info!
That's interesting, thank you.
I was thinking of making a TripAdvisor for concepts, in a kind of tongue-in-cheek way, like "this philosophical concept is worth 3.8 stars out of 5". But then, I figured that it would lead to every reader learning the same ideas. It's better if people explore random ideas independently, so they can make unique connections and come up with novel things. Still, a power-ranking of concepts could be very useful as long as you keep in mind that it doesn't replace exploration.
Next year, we should give the Sneerclub reddit a big red button to destroy LW, and have a big red button here to destroy Sneerclub. Nuclear war is more fun when it's not all like-minded people.
According to the paper, that would be the same individual making several shoots.
You mean seeds that can only work if there's fire? If you have any example in mind, I'm very interested.
Yes, I'm pretty sure there is some diminishing return after some decades (though, apparently, they hit pretty late for bamboos). Now if we stick to the absurd model with no diminishing returns, we can imagine a mutant that almost never reproduces, but when it does, it suddenly covers the entire planet, erasing all the other strains that have been growing exponentially in the meantime. The limit where it doesn't reproduce at all is when a bamboo in a forest appears dead, but will eventually turn the entire universe into copies of itself when comes the Armageddon.
You're right! Corrected. As where the extra resources are stored, I don't know enough about botanic to tell, but here's what they say in the paper: "First, plants that wait longer to flower may accumulate greater energy resources to invest in producing more seeds, and/or seeds that are better protected (Fenner 1985). (The latter scenario, involving better-protected seeds, seems less applicable to bamboos, whose ancestral fruit type is a caryopsis, i.e. fruits with seeds that are generally less well protected than those of many other flowering plants.) In bamboos, this investment might, for example, take the form of increased shoot production between masts". So, at least from that paper, it doesn't look like there's a clear mechanistic explanation, aside from more bamboo.
Many people would think that the second case is more fair. It depends on how these dollars were obtained.
Yeah, there was a bit of hyperbole, but overall when I see what gets in my RSS feed, it's clear that it tends to converge on a particular kind of people and I have to spend a lot of energy finding contradictory sources. On topics like the gender pay gap, I find that one "side" is basically always wrong, and it's hard to tell if it's just because they are really always wrong, or because of my confirmation bias. Also, by "favourite political tribe" I just mean that there are tribes with which I disagree more often than others. I don't think anybody really likes every tribe equally.
It's the second option in both cases (e.g. P(pay|gender) and P(pay) are equal). In addition, since they do the same work with the same productivity, it's assumed that they have the same circumstances and dispositions.
The commuting gap is what I was thinking of. Well done!
Nice try, but it's neither. To clarify, the company doesn't take gender into account when calculating the wages.
I didn't expect that. Do you think there are cases where an unfair society is better than a fair society?
In the riddle's world, there are indeed competitive labor markets. The firm would hire the woman if they could. It turns out they can't, because of the mystery phenomenon.
Oh yes, that's basically the same point! I love this channel btw. Now that you mention it, this post by Jason Crawford on precognition is also relevant.
This is the third time today I read something partially written by a language model. Also, I don't know why, I have a gut feeling that I should start making paperclips. Weird.
I like to imagine a future where people get bored of automation and use transhumanism to start working again. Instead of drones, delivery is done by humans in exoskeletons running through the country at the speed that maximizes fun. Intellectual jobs are done by humans with augmented intelligence that feels just like normal consciousness. Even if it doesn't generalize, I'd expect at least a few people chose this way of life. Isn't it tempting ? (I, for one, will be the running delivery guy)
Could you give more details on how you used the language model? (Great post btw)
This sounds plausible, there were a few experiments ten years ago about this: https://doi.org/10.1111/j.1559-1816.2010.00588.x
I meant the initial example as a justification for investigating the past in the first place, as a reminder that you don't need to be a full-on conspiracy theorist to be suspicious of the historical record. When you say "shifting some facts forward", I would also count that as the victors altering history. Had the US collapsed instead of the USSR, I suppose the facts that would be shifted forward wouldn't be the same.
A competent conspiracy trying to frame someone for murder could have taken a few hairs and placed them at the crime scene.
Yeah, in this case I think we can only use genetic testing for the timeframe where conspirators didn't know genetic testing would ever be possible. You're right that you don't need DNA synthesis if you can plant hair from someone else.