Thanks to a little luck + good genes + some means, you had a reasonably happy childhood, graduated from college, and ended up with a job that you're good at. You like the work you do (most of the time), because people like doing things they're good at. And you...
As we approach AGI, we also approach the risk of alignment failure: due to either mismatch between intended goals and specified goals ('outer misalignment'), or mismatch between specified goals and emergent goals ('inner misalignment'), we end up with a catastrophically powerful failure mode that puts us all in a very...
The problem with this list is it fails the "fluff test". (My name for not-my-idea - sorry, I can't locate the original source.) The "fluff test" is: a suggestion "do X" is useful only if the suggestion "don't do X" is also plausible. Otherwise it is fluff.
So e.g. most political speech immediately fails the fluff test: "if you elect me I will focus on reducing crime" is fluff because "if you elect me I will focus on increasing crime" is not plausible oratory.
Apply the test to each item on this list and you will see that it's all fluff.
I mentally added an epilogue to the post saying "of course, everything on this... (read more)
I think that when you say "I love cats", you mean more than you prioritize specific cats. At some points, those cats will no longer be with you, and if you are like most cat lovers, likely you will then go on to own and prioritize different cats. So while the cats you prioritize at any moment in time may be specific, if you are like most cat owners, over the course of your life you will generally prioritize cats. ("I choose to eat cake every day above other desserts, but I don't in general love cake, just the cakes I eat on a given day" sounds like more like someone who... (read more)
German Autobahns with no speed limit have been designed to be safely driven at high speed. For example, wide lanes, long straight sections, very large radius of curvature for non-straight sections, minimal layout changes, good drainage. And also features which minimise the impact if accidents do occur, e.g. strong central barriers.
It does not therefore follow that removing speed limits on typical American freeways, which have not been designed for high speeds, is a sensible thing to do.
Plus, the way US politics works, if you did any kind of no speed limit trial, it would not last long. Let's say you're a politician who somehow gets approval to push through a policy to trial no speed limit on a freeway. An accident happens (regardless of whether speed was a cause), and you'll be out of office the next day, and that's the end of the trial.
Can you think of examples of mainstream use of the word 'love' for which prioritization isn't an essential component? It seems to me that prioritization is the key thing that binds together what would otherwise be disparate uses of the word, not just in the relationship context. (e.g. the 'love' in "I love reading" and "I love my wife" mean very different things, but are both effectively statements of prioritization)
While 'love' isn't of course well defined, it seems a central component for most usages is one of prioritisation (of time, money, emotional bandwidth etc.) in the face of constrained supply.
So in romantic love, priority of (usually) a single person is a necessary component. Loving a romantic partner is providing a guarantee to yourself that you will prioritize their needs over competing demands (which will therefore reduce in priority). Sex is often one of those needs, but not necessarily. It's the guarantee of priority that matters, not specifically what is prioritised.
The love a toddler has for their toys has this same necessary component. You can't play with all the toys in the... (read more)
I think one of the reasons why punitive damages sometimes make sense is in recognition of the fact that the total 'damage' (in the colloquial, not legal sense) can sometimes include not just an economic component, but a societal component beyond that.
Here's an example: suppose a company is concerned about wrongful death suits. There are basically two levers available here: (1) spend $X on making the work environment safer, (2) put $Y aside to cover the cost of such suits, and of course it's not either/or here. But although this didn't turn out to be the case in the 'factories' example so things turned out well for society, in a given situation,... (read more)
I don't think productivity ever increases with contiguous time spent on a given activity, at least in the short term. (Yes, longer term you can identify and implement working patterns that increase overall productivity.) All the effects push in the opposite direction: low hanging fruit gets picked first, you get tired, you get hungry, you've done all you can on your part of the critical path and need to wait for others to do their bit before you can continue, etc.
So I think there are two explanations here if someone on Hacker News does claim that the last hour is the most productive. The first is post-rationalisation, i.e. if you have been... (read more)
I like the idea of the 4-day work week, but this post is actually a quite separate argument.
The 4DWW idea is: work less, and you'll be happier as a direct consequence.
The argument in this post is: if you want to work X hours a week, whatever that X is, go for it! But rather than spending X on one job where you're almost certainly spending a significant proportion of X in the diminishing returns regime, split it into e.g. 0.8X on that job and 0.2X on a completely separate job. The main effect of this will be productivity gains, which in turn will lead to increased happiness as a side-effect.
Yes. I don’t think the argument requires that the work be hard (or that you work hard at it, whatever that really means). I believe it’s quite generally true that for most activities (howsoever achieved), productivity drops as hours spent increases. Then the rest of the argument follows.
Thanks to a little luck + good genes + some means, you had a reasonably happy childhood, graduated from college, and ended up with a job that you're good at. You like the work you do (most of the time), because people like doing things they're good at. And you also work a lot of hours, because people find it easy to spend a lot of time doing things they like.
And because you work a lot of hours, you'll be pretty far to the right on the transfer function curve (x-axis time, y-axis total work output) where the gradient - the marginal return in work output for the time you spend -... (read 209 more words →)
As we approach AGI, we also approach the risk of alignment failure: due to either mismatch between intended goals and specified goals ('outer misalignment'), or mismatch between specified goals and emergent goals ('inner misalignment'), we end up with a catastrophically powerful failure mode that puts us all in a very bad place.
Right now, we don't know what kind of model will lead to AGI. We can guess, but a few years back people didn't have a lot of hope for LLMs, and now look where we're at. A lot of people were deeply surprised. Solutions to alignment failure may be model dependent, and AGI may emerge from something entirely new, or something... (read 286 more words →)
I personally don't find anything on the list disagreeable (including the summarization and mentoring items).
Summarization is a pretty well established memory consolidation technique to improve long-term recall of information. The OP does not explicitly state this is the aim, but that was my assumption, and if so I think it is uncontroversial that that is beneficial.
Regarding the mentoring, the item on the list was "you would have a good mentor" (which I agree with) and then underneath is "One way to do this is to email people" (which I also agree with in the sense that emailing is clearly one way to do this - I do not necessarily feel this is... (read more)