fasterik

27 ms·

511.
by fasterik 3y ago
It doesn't seem to me like he is generalizing from his experience and applying it to the general population. He's pushing back on the very specific claim that working fewer hours leads to higher productivity. He admits that this m
512.
by fasterik 3y ago
That list was compiled anecdotally by a single person, so I think we should take it with a huge grain of salt. It could be subject to confirmation/selection/reporting bias. For all we know, there could just as many or more people
513.
by fasterik 3y ago
You're talking like there is some objectively best way to live life. My point is that meaning is subjective. Maybe for Carmack, resting and "enjoying the small things" would make him less happy than working on projects. Why d
514.
by fasterik 3y ago
Carmack recently started a new company at age 52 with the goal of creating artificial general intelligence. Whether or not that is a realistic goal, I don't think a modest pace is going to maximize his probability of success.
515.
by fasterik 3y ago
Just because you have a clock doesn't mean you have to play blitz. Why not play a classical time control like 90+30? The benefit of a time control is it keeps the game fair (one player can't spend more time than the other) and ens
516.
by fasterik 3y ago
Fundamentally it's about meaning. People like Carmack find it meaningful to do ambitious projects and push their abilities to the limit. Some goals require more time investment than 40 hours per week. Obviously not everyone finds that
517.
by fasterik 3y ago
What I said doesn't depend on that assumption. I can make the claim that life is qualitatively better while being agnostic about whether we can attach a number to it.
518.
by fasterik 3y ago
Yep. Also known as the golden age fallacy. It's nostalgia for a past that never existed. Some people can't accept the fact that human life has always been terrible, and it's better today than it ever has been.
519.
by fasterik 3y ago
They serve different purposes. You might not be able to find even one person in your local area who shares your specific niche interests, but on the internet you can find hundreds or thousands of the most engaged and knowledgeable people on
520.
by fasterik 3y ago
Google results have become so bad that I use "site:" for a majority of my searches these days. I have a bunch of Chrome search engine keywords set up so that I can go straight to results on Wikipedia, Economist, Reddit, Stack Over
521.
by fasterik 3y ago
Using AI to reduce cognitive load is both a gift and a curse. We can use ChatGPT to write essays for us. But the reason why we write essays is because the act of writing exercises our critical thinking and communication skills. Offloading t
522.
by fasterik 3y ago
It's fair use if the work is "transformative". GPT-4 isn't publishing the content of the books, it's publishing a model derived from the entire corpus. I'm not a lawyer, but I think there's an argument tha
523.
by fasterik 3y ago
That's interesting to know, but that doesn't by itself imply that it's illegal. For example, Google Books, which has massive amounts of scanned PDFs of copyrighted works, is considered fair use under US copyright law.
524.
by fasterik 3y ago
Was GPT-4 trained on data that was acquired illegally? Or was it trained on data acquired legally that OpenAI didn't have the rights to redistribute? There is a difference. In the latter case, whether it counts as "stealing"
525.
by fasterik 3y ago
I have no idea what you're talking about with "general purpose array abstraction" and "map your types over the memory". A struct in C is just a fixed size chunk of memory where the fields have a fixed offset. Access
526.
by fasterik 3y ago
>Projects where you need to think about memory are a pretty small percentage of projects. Performance matters in every project, and the way you get performance on modern processors is by using the cache effectively. That requires a cer
527.
by fasterik 3y ago
"More control over memory than C" is nonsense. C allows you to do anything the machine's instruction set is capable of, all the way down to inline assembly. Undefined behavior is a completely separate conversation. UB is nece
528.
by fasterik 3y ago
>We should work less because it allows us to be better humans. This seems like a false dichotomy. If your work is a net positive for humanity, then doing more of it makes you a better human (up to a point of diminishing returns). Maybe
529.
by fasterik 3y ago
I never said that it does. I was pointing out a logical flaw in that person's argument. Also, why are the creators of GPT-4 authorities on what does or doesn't count as reasoning?
530.
by fasterik 3y ago
This analogy seems pretty disingenuous. Static memory safety isn't an unambiguous good like handwashing; it has tradeoffs. There are many projects where control over memory is critically important and using a language like Rust is a no
531.
by fasterik 3y ago
>No there's no empirical research because ChatGPT 3 is what - 3 months old ? Right, so these claims are based on speculation, not evidence. >Are you seriously saying this tool can't replace people in many many tasks? I c
532.
by fasterik 3y ago
The claim is that GPT-4 can reason sometimes. Evidence that GPT-4 fails to reason sometimes isn't a counterexample.
533.
by fasterik 3y ago
>Despite reason being a metaphysical property of the training data, the process of optimisation means weights are metaphysically reasonless. Therefore, any output, as it is a product of the weights, is also reasonless. This seems wrong
534.
by fasterik 3y ago
Claims about the near term impact of AI are always vague. Exactly which jobs are going away, and how many of them are there? Is there empirical research supporting these conclusions? From my point of view, the impact of AI in the past coupl
535.
by fasterik 3y ago
>It's obvious that human general intelligence operates in frontiers that are plainly outside the scope of computation. IMO, the two most important properties of human cognition are that it evolved through natural selection, and th
536.
by fasterik 3y ago
>I'm also noting that those are already influences on the current formation and existence of the structures around us, so assuming that this is a new idea isn't helping anyone. [...] I am hearing that you don't seem to be
537.
by fasterik 3y ago
I never said change can come fast in terms of a human lifetime. I was pushing back against this idea, which seems to be widespread, that corporations and power structures are inherently bad. The alternative viewpoint which seems more reason
538.
by fasterik 3y ago
The thing you're ignoring is that humans have reason , which is able to produce behavior that goes against human nature and its corrupting influences. There's a reason why some people give away large sums of money to reduce suffe
539.
by fasterik 3y ago
Yeah I get that it's more hip to be cynical. The problem is that cynicism is often an unconsidered attitude without any real substance or rigorous argument behind it.
540.
by fasterik 3y ago
I don't see why that implies we shouldn't take it seriously. The world #1 didn't want to play, so we got a great match between the world #2 and #3 with some very interesting and complex games. Do you also think that Karpov sh

More ›