naasking

41 ms·

721.
by naasking 1y ago
Nuclear operators do need insurance to cover incidents, so they're already paying that cost on top of the absurd regulatory burdens. Current gen nuclear designs are basically meltdown proof, so not only should insurance costs be lowe
722.
by naasking 1y ago
> fragile, unreliable, expensive nuclear plants. Nuclear is currently expensive, but you're 100% wrong on those other two. Also, more people die from installing and maintaining solar and wind turbines than have ever died from nuclea
723.
by naasking 1y ago
> Even the simplest business problem may take a year to solve, and constantly break due to the astounding number of edge cases and scale. Is this really because the single problem is inherently difficult, or because you're trying to
724.
by naasking 1y ago
Nuclear costs are largely due to regulatory burdens created for reactor designs that are not safe. That is no longer the case. Also, attempts to exploit economies of scale could also improve baseline costs, although these attempts haven&#x2
725.
by naasking 1y ago
Definitely true at some level: https://en.wikipedia.org/wiki/Predictive_coding
726.
by naasking 1y ago
Sometimes things can just be "done", and the paper is pretty good documentation if the implementation is faithful to what is described there.
727.
by naasking 1y ago
Really interesting to see how long ideas take to go mainstream. From my recollection, Oleg and Chung-chieh Shan did this first back in 2009 as a library in OCaml [1,2]. [1] https://groups.google.com/g/fa.caml/c&#x2
728.
by naasking 1y ago
> And its going to get worse! That isn't clear given the fact that LLMs and, more importantly, LLM programming environments that manage context better are still improving.
729.
by naasking 1y ago
> You change the prompt slightly, where a human who understands the topic would still give the right response trivially, the LLM outputs an answer that is both wrong/irrelevant and unpredicably and non-humanly wrong in a way that no
730.
by naasking 1y ago
> LLMs just spew words. It just so happens that human beings can decode them into something related, useful, and meaningful surprisingly often. This sentence is inherently contradictory. If LLM output is meaningful more than chance, then
731.
by naasking 1y ago
> My point is that this pattern is not object oriented programming. I think the "is/is not" question is not so clear. If you think of "is" as a whether there's a homomorphism, then it makes sense to say that
732.
by naasking 1y ago
Is this a troll article? The article asked basically the same question: I also wonder how many organizations have had certificates mis-issued due to BGP hijacking. Yes, this will improve the warm fuzzy security feeling we all want at
733.
by naasking 1y ago
I agree with the article, this is "potentially very dangerous". Potential is not actual though, and I'm asking about what damage has actually materialized. Is there a cost estimate over the past 20 years vs. say, memory safet
734.
by naasking 1y ago
> 2) x86 also heavily relies on micro-ops to greatly improve performance. This is not a "penalty" in any sense. It's an energy penalty, even if wall clock time improves.
735.
by naasking 1y ago
On the vulnerability ladder since SSL was introduced, how common and how disastrous have stolen or fraudulent certs really compared to other security problems, and by how much will these changes reduce such disasters?
736.
by naasking 1y ago
> Well, actually there could be a separate step: understanding is done during and after gathering requirements, before and while writing specifications. Only then are specifications turned into code. The promise of coding AI is that it c
737.
by naasking 1y ago
I think the signal-to-noise is demonstrably higher with AI than with a legion of monkeys on typewriters. I think an interesting philosophical question is, is there some threshold of signal-to-noise that by itself would qualify a system as &
738.
by naasking 1y ago
Stones are typically not very dense. Iron or lead or 2-4x denser than typical stones, and so will net you better energy density.
739.
by naasking 1y ago
I think abstract principles deserve extreme skepticism. Pursuing material improvements will yield conditions that improve dignity, but pursuing dignity will not necessarily yield material improvements. Dignity is a luxury for those who don&
740.
by naasking 1y ago
> Nothing in the statistical observation that variables tend to be correlated suggests we should somehow reject the moral perspective that that its desirable for a model to be based on causal rather than merely correlated variables, even
741.
by naasking 1y ago
That's still cutting edge research, not a polished product for a well understood domain. These are like the first steam engines, not a Ferrari.
742.
by naasking 1y ago
GenAI is barely out of the research phase and is only now being fine-tuned for specific applications. Check back in 3 years.
743.
by naasking 1y ago
> But who's to say humans are any different than simple databases with probability and statistics firing neurons?? Nobody has said they're simple databases, they would obviously be complex databases.
744.
by naasking 1y ago
> The Bible may be the deadliest book ever written. Books are excuses for war that would have already happened. They would have just found other reasons if the Bible didn't exist.
745.
by naasking 1y ago
I'm not sure "economic incentives" are what most people classify as a "planned economy", unless you want to take the broadest, most expansive possible understanding of that term. All Western nations would have plann
746.
by naasking 1y ago
I don't see how that could be considered a planned economy, you're describing individuals creating startups of their own free choice and the government backing them with no strings. Individual choices are driving economic progress
747.
by naasking 1y ago
True, and in the case of Solomonoff Induction, incompleteness manifests in the calculation of Kolmogorov complexity used to order programs. But what incompleteness actually proves is that there is no single algorithm for truth, but a coll
748.
by naasking 1y ago
Define causal reasoning?
749.
by naasking 1y ago
> Because reasoning tasks require choosing between several different options. “A B C D [M1] -> B C D E” isn’t reasoning, it’s computation, because it has no mechanism for thinking “oh, I went down the wrong track, let me try something
750.
by naasking 1y ago
> I want it to have empathy so that it can understand what I'm getting at when I occasionally ask a poorly worded question. I'm not sure what empathy is supposed to buy you here, I think it would be far more useful for it to as

More ›