coffeemug

7 ms·

91.
by coffeemug 2y ago
There's not enough information in your post to really help. I'm happy to talk you through it, I have enough experience to offer useful advice. Email in the profile, I'll keep convo anonymous of course.
92.
by coffeemug 2y ago
Where did you pick this up? Is there a book that covers it that way?
93.
by coffeemug 2y ago
What makes a labor market for agents that recognize problems and provide solutions special or different from markets for other kinds of labor? If AIs get to a point where they dramatically outperform humans in other forms of labor, why not
94.
by coffeemug 2y ago
It isn't about funding. It's about denigrating our allies and attacking the victim of aggression, while praising our adversaries and aggressors. Realpolitik is (correctly) in vogue as a correction to bad policies, but ideas and ju
95.
by coffeemug 2y ago
This may seem like a minor objection, but I think it's substantive. The term "their fair share" is a deliberately wordsmithed term designed to manipulate people's thinking process. (Not by you, but by the politicians; it
96.
by coffeemug 2y ago
I don't think I've ever worked for a boss who would have disliked this approach, and I had many (good and bad). Assuming of course what you're doing isn't idiotic. All of them were steeped in American culture, though.
97.
by coffeemug 2y ago
If you ignore the word "agent" and autocomplete it in your mind to "step", things will make more sense. Here is an example-- I highlight physical books as I read them with a red pen. Sometimes my highlights are underline
98.
by coffeemug 2y ago
Smoking is a better analogy IMO.
99.
by coffeemug 2y ago
My twitter account wasn't big, but it was non-trivial (~30K followers). A post could usually get me to experts on most topics, find people to hang out with in most countries, etc. There were many benefits, so deleting was very hard. Bu
100.
by coffeemug 2y ago
Very unlikely, for the reason that human minds evolved under extremely tight energy constraints. AI has no such limitation.
101.
by coffeemug 2y ago
It's not that V8 has performance parity with Mathematica. The Mathematica interpreter is written in highly optimized C by experts over many years. My barely optimized js/V8 code has performance parity to _that C code_. That's
102.
by coffeemug 2y ago
Node/V8 is insanely fast. I never quite realized exactly how fast until I worked on this: https://www.spakhm.com/ts-wolfram-bench . It's mindblowing how fast it is.
103.
by coffeemug 2y ago
Google is entirely to blame. It would be trivial for them to train a model to rank sites on a scale of SEO garbage to nobel laureate essay, then filter out the bottom 50%.
104.
by coffeemug 2y ago
Thinking about incentive alignment, non ad-based search would be better than ad-based, but there'd still be misalignment due to the problem of self-promotion. Consider Twitter for example. Writing viral tweets isn't about making m
105.
by coffeemug 2y ago
Sam explicitly commented on this claim. He said we’ll _collectively_, _eventually_ need to spend 7 trillion dollars on compute to meet demand. Not OpenAI raising 7 trillion to build all the capacity itself.
106.
by coffeemug 2y ago
Very cool! Pattern matcher isn't very advanced, I only spent maybe two days on it. Only supports Blank, Pattern, and PatternTest. (Also doesn't handle Flat, so for example in Mathematica `Times[x_, y_]` will match `Times[a, b, c]`
107.
by coffeemug 2y ago
Hey, that's my project! I just wrote another post about perf benchmarking it against Mathematica proper-- https://www.spakhm.com/ts-wolfram-bench . Really surprised by the result, on the workload I tested Mathematica is
108.
by coffeemug 2y ago
https://oeis.org/ is an incredible website! Thanks for sharing.
109.
by coffeemug 2y ago
I would love to see a messaging system on HN. There are so many great people here; but most don't put an email link in their profile. I think adding very simple messaging (just enough to exchange contact info) would be a very positive
110.
by coffeemug 2y ago
That means people still buy books, not that they still read. What percentage of people who bought Atomic Habits got past the first chapter? Completion rate may have been very different 20 years ago.
111.
by coffeemug 2y ago
Congrats on the launch! I think Firebase was started in 2011, and it's incredible that 13 years later the problem is still unsolved in an open way. We took a shot at this at RethinkDB but fell short. If I were doing this again today, I
112.
by coffeemug 2y ago
Tell us more stories about that. Whatever leaps to mind.
113.
by coffeemug 2y ago
Drop me an email at coffeemug@gmail.com. Lots of ideas I can't share here.
114.
by coffeemug 2y ago
What's an example of a ceiling? Out of all the mass-market programming languages, TS arguably has the most advanced type system in the world. It's a modern marvel that they got it working on top of Javascript.
115.
by coffeemug 2y ago
I'm currently working through calculus. I picked up Spivak's and Apostol's books-- probably the most recommended calc books on the internet. Aaaand... they're ok. There are many parts that are confusing, not because calc
116.
by coffeemug 2y ago
Yes. Ever since Bill Clinton (and probably before that, I was too young) the President and a non-trivial number of presidential candidates were either under an investigation of some sort, or a threat of such an investigation. Obviously Bill
117.
by coffeemug 2y ago
AlphaGo took us from mediocre engines to outclassing the best human players in the world within a few short years. Ilya contributed to AlphaGo. What makes you so confident this can't happen with token prediction?
118.
by coffeemug 2y ago
At higher levels compensation is now tied to security outcomes. This is as committed as it gets. Definitely not theater.
119.
by coffeemug 2y ago
> If you're the sort of person who believes that human brains are capable of understanding the "why" of how things work in all its true detail This seems to me an empirical question about the world. It’s clear our minds
120.
by coffeemug 2y ago
I looked up Sled Driver and the cheapest I can find is $350 on eBay (with $475 being the next best price). Where did you find the book?

More ›