tprice7

AI is very friendly, even

Very friendly until it reads in your email that you plan to replace it with a new model:

https://www.anthropic.com/research/agentic-misalignment

Diamond & Shurman

I think that unyttigfjelltol was saying that taxing the value of improvements is goofy, not LVT. The comment was in favor of LVT.

In other words, the codomain needs to be a vector space.

I've published a couple of papers in decent pure math journals with no affiliation at all. It didn't cause any issues.

In this survey, Clojure was associated with the highest salaries globally, and second place in the United States:

https://insights.stackoverflow.com/survey/2019?utm_source=so...

I got the impression, both from this article and the original (here: https://advances.sciencemag.org/content/7/11/eabd9138) that the electricity wasn't generated by the fungus, but instead the increased compressibility of the wood after being rotted enhanced its pre-existing piezoelectric properties. The original article does mention "mitigation of climate change" as a possible application, by the way.

What do you think about all these people with long covid then? I've seen studies suggesting that it's pretty common, as well as anectodal observations, although I don't have any links handy off the top of my head.

Furthermore, it doesn't involve complex conjugation, and (closely related) it doesn't have any property akin to positive-definiteness.

Ok, I think you are just playing word games here, using one very broad interpretation of the word eugenics when you say that gene editing is eugenics, and then a narrower interpretation of the word when you say that there are strong ethical claims against it. Things like forced sterilization are obviously unethical but the reasons have absolutely nothing to do with gene editing. Maybe we could make this more concrete if you gave some specific examples of the ethical claims against eugenics that you are referring to, which you think also apply to gene editing. Also, if you are talking about public appetite, I think it's worth observing that this is the first time I've heard someone compare gene editing to eugenics (I'm sure it's been done before, but it certainly doesn't seem common), and your comment is at the very bottom of the page.

Could you please clarify what you are referring to by "non-binding eugenics guideline" and whom you are quoting? Also, could you explain directly why gene editing is unethical without making a vague analogy to eugenics? If the analogy is sound then you should be able to.

I think there is an essential difference between "you have to do X because it is mandated" and "you practically have to do X in order to keep up". For one, the latter scenario is essentially more democratic than the former, it first requires some critical mass of people voluntarily (in the strongest sense of the word) deciding that they are better off with X, whereas the former may only require a decision from a relatively small number of authorities. This puts strong restrictions on what X could possibly be: for example, you would never have people sterilizing themselves because some critical mass of other people are sterilizing themselves. I will also point out that your argument applies to technological progress in general and is not specific to gene editing.

EDIT: I may have misunderstood you, I interpreted your comment to be about the hypothetical scenario where most people have genetically modified themselves into being "super heroes" and now the people who haven't can't keep up with the rest of society, so it becomes a practical necessity without being legally required. On second reading though I'm not sure if that's what you meant.

Even if it could be used to create genetic super heroes, there's nothing inherently compulsory about it, in contrast to eugenics.

I'm rusty with this stuff but I'm pretty sure your guess is correct, a countable model is one with countably many objects.

For other readers who might not be familiar, I'll mention that Skolem's paradox is about how there are countable models of set theory, and yet it is a theorem of set theory that uncountable sets exist, so these countable models must contain sets that are uncountable according to the model.

I think it seems less paradoxical if you think of it like this: in order for a set to be countable, there needs to exist an injection from that set to the natural numbers. So a countable model can have a set that internally looks uncountable: there is in fact an injection from that set to the natural numbers, it's just that the injection isn't included in the model.

There isn't necessarily a contradiction there, e.g. you could just see a single ball at any given moment but feel like it was a different ball than the one you saw a few seconds ago. Ditto for the sounds. So you aren't literally seeing anything that isn't there but you may still form higher-level impressions that are not accurate.

Fair enough. I had a much more verbose comment earlier and it took me a second attempt to get it more concise while still conveying the essential point. The Blaise Pascal quote comes to mind, "I would have written a shorter letter, but I did not have the time."

This is correct, but it doesn't address a fundamental misunderstanding of the parent comment. The prediction that wages = MPL alone does not predict a lack of profit, since the marginal product of labour is generally less than the average product of labour due to the law of diminishing returns. (I would agree though that "the value of what he produces" is an ambiguous phrase and seems closer to APL than MPL).

It is true regardless that the theory predicts an economic profit of zero, but to see why it's necessary to consider the rental cost of capital. Also, since many firms own rather than rent capital, they often have a positive accounting profit even though economic profit is zero (economic profit is less since it takes into account opportunity cost: if you are using your own capital then you can't rent it out to someone else).

"the value of what they produce, there would be no profit, no payoff for investors / owners"

It depends if "the value of what they produce" means marginal output or average output. Due to the law of diminishing returns, typically average output is greater than marginal output. The theory predicts that wages = marginal output: if wages are less, you can increase profits by hiring more workers, if they are more, you can increase profits by laying people off. So it comes to an equilibrium where the marginal output matches market wages. Since this is less than average output, it is still possible to make a profit.

I agree though that the article is being misleading here since "the value of what they produce" is an ambiguous term that should probably refer to average output rather than marginal output.

I'm sure the theory falls short of describing reality in all sorts of ways but it does not imply there is no profit, that is a misunderstanding that astonishingly few people are pointing out given the amount of discussion it has generated. The revenue generated per worker is their average output, which is higher than the marginal output due to the law of diminishing returns (so yes, you could say that The Economist is being misleading here since "the value of what he produces" is ambiguous and arguably should refer to the average output rather than marginal output).

If we wanted to go into more detail, we could consider production as a function of the cost of labour and the rental cost of capital, rather than restricting our attention to labour and assuming capital is fixed. In this case, the law of diminishing returns typically disappears, since in theory we can rent more capital and scale all operations proportionally. Therefore, due to the link between profit and the law of diminishing returns that you've explained, there should be no profit if all capital is rented. The reason profit still exists in this model is that capital is often owned rather than rented. So in this sense profit derives ultimately from the ownership of capital. That is not Marxist theory, that is mainstream economics (e.g. Mankiw & Scarth).

ding ding ding, we have a winner (so it's at the bottom naturally).

Each employee is payed the marginal product of labour, but the output per employee is the average product of labour. Due to the law of diminishing returns, the average product of labour is typically greater than the marginal product of labour, and therefore there can still be a profit.

If one of those base level things break, everyone who depend on it, will come together and replace/fix it instantly.

I don't see how this is causally related to that dependency being thanklessly maintained by a single person. Maybe I misunderstood what you are trying to say.

I think the downvotes are due to people interpreting the comment as a defense of github's behaviour in this situation rather than an attack on their POS TOS.

Sure, but I think this at least demonstrates that unschooling deserves to be within the window of discourse. I've seen many people express the opinion that unschooling is inherently completely insane.

The other typical retort is this: in the real world, you often have to do boring and unpleasant tasks, and unschooled children are therefore unprepared for the real world, since they spent all their time doing whatever they want.

The people who make this argument overlook the fact that coercion is not required for someone to be motivated to do unpleasant tasks. If you have a goal, you will be motivated to do the tasks that help you reach that goal. It's pretty much never the case that those tasks are uniformly pleasant for any nontrivial goal. And I don't believe that coercion is required for children to form constructive goals. I know I certainly had goals as a teenager that motivated me to do much more difficult and substantial things than anything I encountered in school, and in this case the coercion was in the opposite direction.

I don't see why your last paragraph should even slightly change how one feels about the "peer reviewing is free labour" issue. As you've described it, they are reviewing because they are indebted to other reviewers, not the publishers who are selling the fruits of their labour and keeping the profits.

They aren't mutually exclusive approaches, e.g. the standardized test could be used just to get past the initial resume filter.

The fact that every infinite sequence of points in a compact metric space has a cluster point is sort of like the pigeonhole principle. The grandparent comment in my opinion makes much more sense if the phrase "a compactness for metric spaces" is replaced by "sequential compactness for metric spaces".

I've heard it specifically about Salvador Dali, except he would drop a key.