dahart

15 ms·

211.
by dahart 5mo ago
> Income inequality isn’t the same things as government resources available per person. Correct. You clearly understand that your citing of averages papers over the poverty rate and conflates the gains of the rich with the plight of the
212.
by dahart 5mo ago
Oh, then you’re of course aware that many many people did in fact return, and that your earlier estimate of the number of people coming to LA after 2005 that hadn’t lived there before was over-estimated. According to this study, almost a th
213.
by dahart 5mo ago
Louisiana has the highest poverty rate of all the states in the nation, according to the 2023 Census, at 18.9%. https://www2.census.gov/library/publications/2024/demo/acsbr... Louisiana ranks 43rd in per
214.
by dahart 5mo ago
Why are you assuming the rapid increase in LA’s population from 2006 to 2010 did not have a significant portion of temporarily displaced people moving back?
215.
by dahart 5mo ago
Is it possible the land is sinking because of the rising water level?
216.
by dahart 5mo ago
Have you bothered to check @nekzn’s comment history? Attacking new people here is going to ensure we’re left with only bots. This is one of the reasons why HN has the guidelines that you’re repeatedly breaking. If you do suspect a bot after
217.
by dahart 5mo ago
Looks very cool assuming all the comparisons are correct & fair and there’s no major failure cases. Quick link to the HTML version of the paper to save you a couple of clicks: https://arxiv.org/html/2605.05148v1 Si
218.
by dahart 5mo ago
> I was trying to point out the absurdity of correcting illegal activity by simply eliminating laws. Isn’t this straw man? Who said anything about eliminating laws or being inconsistent about legal immigration? The top comment was only p
219.
by dahart 5mo ago
Yes that’s the original spelling & meaning. But using the spellings foobar, foo, bar, and sometimes baz, have been used for decades in programming as examples, temporary names, stand-ins etc. I just assumed that spelling it foo was mean
220.
by dahart 5mo ago
Yes making fun is a subset of joking, so when I ask if you’re joking and you say no and argue, you’re being at best intentionally obtuse, and also incorrect about English word meanings. This isn’t nitpicking, I’m revealing the root cause of
221.
by dahart 5mo ago
> The first message was making fun of the title ... It's, BTW, not exactly the same thing as joking. Oh, really? Why do you believe that? https://dictionary.cambridge.org/us/dictionary/english/make-...
222.
by dahart 5mo ago
That all depends on what you’re doing. Like I said, if a high level lang or lib supports and fits your goal well, then yes you should use it. I don’t know what most people are doing, but it’s fair to say that a lot of people can use a highe
223.
by dahart 5mo ago
I haven’t tried them all, but I suspect Warp is the easiest; it’s ridiculously easy. I’m sure there are some tradeoffs, so once you learn a little CUDA in Python it might make sense to switch from Warp to Numba or CP depending on what you’r
224.
by dahart 5mo ago
So you are in fact claiming you were joking? Why didn’t you say that from the start instead of arguing about English? For multiple messages in a row, there’s no indication you were making fun, and the big problem with the joke, the reason i
225.
by dahart 5mo ago
Regarding the section on Python and high-level CUDA, anyone interested should maybe first take a peek at Warp, which I’m guessing is too new to have a book yet. Warp lets you write CUDA kernels directly in Python, and it’s a breeze to get s
226.
by dahart 5mo ago
It’s not about whether you work at Nvidia. Avoid writing CUDA kernels if there are higher level libraries that do what you need. Do write CUDA kernels if you want to learn how, or if you need the low level control, or to micro-optimize. Bei
227.
by dahart 5mo ago
It’s a bit late in your argument to plausibly claim you were joking, if that’s what you are implying; if that was your first reply I’d maybe have believed it. Outside of that, you did use English wrong; to a native speaker “Fecal transplant
228.
by dahart 5mo ago
A retina transplant is a treatment for blindness, and you can optionally leave out “a treatment”. Note that saying “treatment for blindness” doesn’t actually fix anything, it doesn’t mean you treat someone so that they become blind. You see
229.
by dahart 5mo ago
A transplant in a medical context is a type of treatment. I’d be willing to bet using ‘for’ like this works the same way in your native language.
230.
by dahart 5mo ago
That’s all true and I agree with all of that. But again, that doesn’t address why publicly funded research ought to be locked away for private gain. This is not to look a gift-horse in the mouth; open access publication has really taken off
231.
by dahart 5mo ago
What do you mean? Research funding absolutely is an investment. Research that is productized is a venture, and the universities involved absolutely do invest in the venture, and sometimes see a handsome ROI. In terms of why the public funds
232.
by dahart 5mo ago
To be pedantic (apologies), I’m certain you know this, but not everyone thinks about floats - precision loss is almost always occurring in practice with every floating point operation you do, and happens often even when combining numbers of
233.
by dahart 5mo ago
This is of course why we have audio compressors that lower everything while it’s loud and bring it back up when it’s quiet and try to do it quickly but without anyone noticing too much. It’s theoretically possible the same trick could wor
234.
by dahart 5mo ago
Wild to see multiple references in one place! That was one of the more memorable scenes ever…
235.
by dahart 5mo ago
That’s valid English. You take ibuprofen for headaches, or go to the hospital for a broken bone. Assuming the word ‘for’ means ‘causes’ is incorrect.
236.
by dahart 5mo ago
Funny thing is, that may be a bigger concern here, that the research is often publicly funded while the uses of that research and the profits that come with them are kept private. It’s complicated and I buy the notion that research improves
237.
by dahart 5mo ago
That code is absolutely terrible! Never do that. The range is awful, and the relative error is awful. If you want a double in 32 bits, convert to single precision float. This will beat the relative error of the code you linked to by orders
238.
by dahart 5mo ago
> so you should keep people on pay-roll because they "might actually be doing something if you look hard enough" ...? don't think so The point was that you might be making incorrect assumptions that they aren’t doing work,
239.
by dahart 5mo ago
Which kind of meritocracy - the original definition that points out that “merit” is a function of wealth and social advantage, or today’s sanitized version that attributes merit to achievement while ignoring all the advantages that lead to
240.
by dahart 5mo ago
Absolutely agree!

More ›