layer8

3 ms·

31.
by layer8 6d ago
I would assume the parent understands that. The criticism is about the LLM behavior this results in. Being able to explain a behavior doesn’t necessarily excuse it. By some moral standards, you wouldn’t have trained an LLM that way, or woul
32.
by layer8 6d ago
Sure, but that’s different from “you do/don't care which one”.
33.
by layer8 6d ago
That rule isn’t universal though. “I go to the dentist every year” doesn’t imply that I particularly care about a specific dentist (or, case in point, about the specific dentist). “This was a nice evening” is also difficult to characteriz
34.
by layer8 6d ago
Both are being used and neither is incorrect. “An historic” used to be more prevalent in British English until around 1990, but “a historic” has become the standard choice now: https://books.google.com/ngrams/graph?cont
35.
by layer8 7d ago
I don’t see any conflict with what the root comment wrote. There are probably some differences between human and AI editorial suggestions that could be discussed, but that wasn’t the topic here.
36.
by layer8 7d ago
> My reason: code can be checked objectively. I can run it and confirm it works. Running the code only confirms that it works with the precise input, in the precise environment, under the precise circumstances you run it under. It doesn’
37.
by layer8 7d ago
I expanded my comment since.
38.
by layer8 7d ago
TFA is about having AI do the writing, not about having AI suggest editorial improvements. The parent comment argues against having the AI write the text and having the human merely review it, not against having the AI review human-written
39.
by layer8 7d ago
It doesn’t really matter if most app providers won’t be porting their apps to the new phone OS.
40.
by layer8 7d ago
Apple seems to be having more success.
41.
M6 Pro Chip Result on Geekbench Is Likely Fake(macrumors.com)
4 points by layer8 7d ago|0 comments
42.
CNN, MS Now and Politico Barred from White House After Trump Announces Ban(nytimes.com)
4 points by layer8 7d ago|1 comments
43.
by layer8 7d ago
In aggregate, yes. And even if LLMs didn’t exist (or had different idiosyncrasies), the article would still be exhibiting a repeatedly weird and stilted writing style.
44.
by layer8 7d ago
> I remember fondly the pre-LLM days when I used to love reading about programming languages. First I thought you were going to comment about the grating LLM-isms in the article, which made me end up not enjoying reading it.
45.
by layer8 7d ago
Meaningfulness comes from what you care about. I wonder why you were interested in programming-language concepts before but now (apparently) stopped caring about the code. The code still remains the language that communicates the actual pro
46.
by layer8 7d ago
> Tao is speaking Tao isn’t the article author, it’s a guest post.
47.
by layer8 7d ago
> Tao's concern The article isn’t by Tao, it’s a guest post by Grant Sanderson (aka 3Blue1Brown).
48.
by layer8 7d ago
No, it would be smaller, thinner, and lighter than the Air.
49.
by layer8 7d ago
> shaped for one hand Unfortunately, it is as wide as a regular iPhone (i.e. the 17e), so it’s not an iPhone mini/SE replacement in terms of one-handed use. At 158 grams it’s also heavier than any mini/SE.
50.
by layer8 7d ago
The purpose is to free optimizers from solving the halting problem (and similar undecidable propositions), which they can’t. So the approach is to reduce the allowable programs to those that optimizers can reliably reason about. By the very
51.
by layer8 7d ago
See https://news.ycombinator.com/item?id=49760653 .
52.
by layer8 7d ago
See https://news.ycombinator.com/item?id=49760653 .
53.
by layer8 7d ago
Because we want compilers to perform the optimizations, not programmers.
54.
by layer8 7d ago
This actually fixes a case that became a problem for Rust implementations: https://github.com/rust-lang/rust/issues/28728 So yes, it is good for Rust.
55.
by layer8 8d ago
Learning isn’t stealing, but if you’d set up a forum or hotline where paid staff would answer questions and write essay directly based on NYT content without a license for that content, that would probably be deemed illegal.
56.
by layer8 8d ago
I don’t think I’ve seen it expressed as well as this before. Thanks for your comments.
57.
by layer8 9d ago
Seems sensible, given that birds have higher body temperatures.
58.
by layer8 9d ago
Usually the hardware for which the respective music was composed is the most preferable. For example for Ultima VII, that would be a Roland MT-32 (plus a Sound Blaster Pro for speech synthesis). The SC-55 only provides incomplete emulation
59.
by layer8 9d ago
> Do you think human mathematicians will be entirely useless for any kind of input into the act of doing mathematics? Not at all. My point is that the Fields medallists’ letter didn’t provide good arguments, not that there aren’t any. An
60.
by layer8 9d ago
> we urgently need to come up with good ways of explaining the value of having a large pool of human mathematical experts, even if it is no longer part of their role to find new proofs of theorems. This is the main issue, and while I ful

More ›