> However, for some reason, both Gemini and ChatGPT tend to argue with me so heavily and inject their own weird stupid ideas on things
do you have examples of this?
asking because this is not what happens to me. one of the main things i worry about when interacting with the llm is that they agree with me too easily.
one issue that comes to mind is the nature of the determinant. when one considers the determinant defined by the recursive definition, it seems like a highly contrived object that is difficult to work with (as it is from that definition!). avoiding that confusion requires that a lot more scaffolding be built (ala Axler in the "Done Right" book). either way you have some work: either to untangle the meaning of the weird determinant or get to the place where you can understand the determinant as the product of the eigenvalues.
it really depends on what you mean by rich: the surest path to end up with $2-5M over ~10 years is job at ~FAANG, do it well to get promoted, and manage your savings/investments well. that path is very unlikely to get you to $10-100M in the same 10 years, and starting a startup seems to be one of the best ways to do that.
i was going to sit this one out, but i want to +1 the above with more than just an upvote. Austen seems to be arguing and explaining all the happenings in good faith with someone who appears to be hell-bent on "catching" him with something by misconstruing things.
i have never understood why Lambda/Bloomtech has so many haters but the other commenter here appears to me to be in that camp.
this is all accurate and i'd add that it is deceivingly hard to make something that sustains itself via advertising. when youtube was acquired google had a large part of that already figured out.
one of the things that helps generate new ideas that can be cultivated is the ability to be playful. once a given problem or subject is sufficiently loaded onto a brain, if that person can relax and have child-like naivete about poking and prodding, novel insight is usually not far.
cultivating this ability is fairly well understood in a lot of domains, i think. two examples that are top-of-mind are improv and jazz.
> The author of this article was paid $42 million last year. Uber's top 7 executives received: "$11.4 million in salary and cash bonus, plus $71 million worth of equity awards."
it's misleading to call equity compensation "getting paid". there's a strong argument that executives are being over-payed, but say the CEO receiving a salary of ~1.6M (11.4/7) is a lot less egregious than them being paid $42M. surely the equity conferred to the execs gets them some cash availability in various ways, but it isn't exactly the same as just giving them money (eg what would happen if the CEO uber sold off all of his stock?).
maybe the cash portion of exec salaries ought to be much lower or maybe their equity compensation ought to be much lower, but i don't think either of these things by itself or in combination is going to solve the structural problems of the american economy such as healthcare being tied to employment.
i agree wholeheartedly with the spirit of your comment, but saying "nutritional science is complex" or "the dosage alone makes it so that a thing is not a poison" does not help people figure out what foods to choose or how much to consume.
looking at the glycemic index is a useful heuristic and some of the research on canola oil makes me skeptical of it, and i think it is good for the author to point that out, as it suggests that oat milk might not be as safe as it is marketed to be.
another heuristic is biasing in favor of food that is processed less eg eating a bowl of oats is probably a better idea than drinking the analogous amount of oatly. similarly, it is a good idea to eat foods that people like you have been eating for long times, which in the oats vs oatly example favors the oats.
the precautionary principle suggests that the onus is to verify the safety of a given new food, not to prove that it is unsafe.
i was a physics PhD student once and i might have some thoughts on how to learn physics, but i think the parent commenter is spot on -- the only reliable way we have is to have an apprenticeship with someone who works well with you.
there are a tons of posts along the lines of this one, which are largely lists of books that people like to romanticize or as the parent comment says, lists of things that someone thinks you should know.
the Feynman lectures are a good example of that, actually. i've read them a couple of times and they are wonderful -- like most things Feynman. but very unhelpful. even Feynman was disappointed at how the students they were given to weren't quite learning the physics. Feynman is to explaining science as, eg Joshua Bell is to performing music for the violin. but in terms of actually trying to learn what a physical theory says, to what degree it holds, what are the current open problems with it, and what might be good approaches to solving them, Feynman explanations outside of highly technical works (like his papers or maybe his lecture notes on statistical mechanics) are not too useful. and the more technical works are very difficult to approach by yourself, you'll need a mentor.
the CFR is only as low as 0.5-1% when there is adequate medical care and the population is otherwise healthy. in NYC the CFR for the 18-45 cohort is ~5%[1].
context: former theoretical physics grad student, dropped out ABD to start a company.
this comment is right on the money on several fronts:
- the Feynman lectures are great after you already understand some of the mechanics of "doing" physics and have some other exposure to the topics, Halliday & Resnick is a better place to start on any one topic
- with infinite time, i'd always follow the approach of learning the math first and then the physics, the book by Boas is pretty good for self study of the minimum necessary math
- there's no real reason to follow a traditional grad school curriculum ala Fowler unless you need to pass quals in a traditional grad school setting
- simulating physical systems on computers is a pretty good exercise, but very time-consuming, avoid if you already spend a lot of time in front of computers
and some thoughts of my own:
- "get a strong foundation in Physics" is a bit too vague to be a useful goal, some examples of potentially better goals: "be able to pass a classical mechanics qual in the allotted time", "be able to write down the standard model and explain it", "be able to grok N papers from the X section of arxiv per week", "be able to write down toy classical field theories and calculate their predictions", ...
- if you are looking to avoid computers, try supplementing your reading with simple experiments either by buying educational kits or by hacking together things
- the books by David Griffiths (esp the E&M one) are awesome
- try to follow curiosity instead of a program: trying to answer "how do superconducting materials work?" for yourself is better than "follow the grad intro to condensed matter that's available online"
- use the physics stackexchange and other forums: asking and answering questions can be very helpful
it's a bad model to think of measures responding to covid as scalars. they have magnitude and direction.
everything below is me spitballing -- weakly held:
my thinking is that we should err on the side of caution, which means that if anything, in terms of flattening the curve, we haven't done enough, yet, but there is also that any it takes ~1-2 weeks to see the effects of any one action. once the curve is flattening, i think we ought to relax on this front as to minimize the economic damage.
in terms of structural changes to medical system (think FDA approvals), i think that we're also under-reacting, but i don't know enough to really say.
in terms of stimulating the economy, i think we've also under-reacted so far, though the latest bills look a bit promising on this front and i think if they pass they are of about the right magnitude.
> The primary value of a business is their assets, brands and expertise. All of those things will be just as valuable after the pandemic as they were before.
this doesn't seem true to me. at least the brand loses value over the pandemic, and there's likely complications from cashflow that eat into other assets.
i agree with everything you said, except for the part of your being unable to help. i think OP is asking for what information informs schema decisions, and what are some heuristics to use. expanding on the differences between your two examples would be very valuable!
> Whenever I see a productivity blog by someone who is enthusiastic about recent changes they've made to their life, I immediately assume the writer is quite young.
> As people get older, they realize that you only have the "right" to write a blog on productivity if you've actually found a system that keeps you productive for years on end.
i broadly agree with you -- i'm not sure it has to do with the author's age, per se, but productivity advice based on recent changes to the author's systems ought to be taken with a huge grain of salt.
that said, i do think it is good to share what one is trying and whatnot -- blogging is cheap. for example, while the reader-beware above applies to this post and most of the ideas were familiar to me, i had never thought to record my screen while i code to get feedback, despite the fact that i record myself regularly to asses performance in music! that bit alone made it worth the 5 min it took to read.
this is one of my favorite (bc it is pretty harmless) cases of regulations failing spectacularly: the reason the companies do this is that it is a legal requirement for H1B visas. the idea, IIUC, is that before giving the job to an immigrant, the company should do a proper search to fill the position with an american candidate. what happens instead is that the companies make no additional effort to find american candidates after they've decided to sponsor the H1B (largely because that's just not how the hiring pipelines work), and instead everyone wastes time and effort "complying" with the requirements that law has for a proper search.
do you have examples of this?
asking because this is not what happens to me. one of the main things i worry about when interacting with the llm is that they agree with me too easily.