Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
tomp
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
45 ms
·
541.
▲
by
tomp
4y ago
Yes. When you're in a "tight loop" (e.g. a matrix multiplication, which is basically 3 nested loops that only load data, do math, write data), Java's virtual threads just won't yield. So if you write your app in the
542.
▲
by
tomp
4y ago
The fact that they’re cooperatively scheduled makes them quite problematic. Basically a leaky abstraction, you have to think whether your code will block or not. Go went through the same process, before eventually figuring out that they do
543.
▲
by
tomp
4y ago
These are more like “prime brokers” rather than “counterparties” in trade. How often do you assume that your prime broker will go out of business before your hedge fund does? Unless you’re a big one (Citadel, Millenium), probably “never” (i
544.
▲
by
tomp
4y ago
Better link to same content https://old.reddit.com/r/ChatGPT/comments/11wkw5z/has_chatgp...
545.
▲
by
tomp
4y ago
They are if you arrest someone in response to mere allegation.
546.
▲
by
tomp
4y ago
With this kind of mindset, we'd still be using lead pipes and letting blood. Doctors and plumbers might make society work, but technology drives society forward.
547.
▲
by
tomp
4y ago
try googling "<race> person face" where <race> is "white", "asian" or "black" and you get pretty representative results of what people mean by "race". (same if you do "dog b
548.
▲
by
tomp
4y ago
we can't define "night" either.
549.
▲
by
tomp
4y ago
Thanks for reminding me! It works now. The difference is striking!
550.
▲
by
tomp
4y ago
The killer app is waiting for the right hardware. Imagine doing remote surgery, with your headset showing, in 3D, the patient's body. Or in-person surgery, with a 3D rendering of the CT/MRI scan overlaid the patient's body. T
551.
▲
by
tomp
4y ago
From what I’ve read, they would have to mark the portfolio to market (i.e. report its value in terms of current market prices), not say “we intend to hold this to maturity” and account it at face value. Like, the latter actually makes more
552.
▲
by
tomp
4y ago
Well then there’s 4: The regulators failed massively, allowing banks to mask their duration losses. Obviously this needs to change, but they probably need to fix the other banks first (that are still hiding their losses).
553.
▲
by
tomp
4y ago
GPT-4 gets it. https://twitter.com/tomprimozic/status/1635720278578692152
554.
▲
by
tomp
4y ago
What's wrong with it? I'm a native speaker of neither, but they do sound similar (of course, gracias being pronounced South American style, not Spanish style). Except the fact that Spanish and English "r" differ, but
555.
▲
by
tomp
4y ago
It’s a faster conda. Strongly recommended
556.
▲
by
tomp
4y ago
Which insurance company can pay out a $40bn risk event?
557.
▲
by
tomp
4y ago
hm... well... It definitely runs . It uses almost 20GB of RAM so I had to exit my browser and VS Code to keep the memory usage down. But it produces completely garbled output. Either there's a bug in the program, or the tokens are dif
558.
▲
by
tomp
4y ago
5 hours ago my comment was flagged for saying Yellen is a liar ("there will be no bailout"). I'm glad lies were exposed so quickly, and my comment vindicated. https://news.ycombinator.com/item?id=35120888
559.
▲
by
tomp
4y ago
You don’t actually need torchvision. I just used mamba create -n llama python==3.10 pytorch sentencepiece
560.
▲
by
tomp
4y ago
Better instructions (less verbose and include 30B model): https://til.simonwillison.net/llms/llama-7b-m2 I’m running 13B on MacBook Air M2 quite easily. Will try 30B but probably won’t be able to keep my browser open :
561.
▲
by
tomp
4y ago
Same here. running 4-bit LLaMa 13B on my MacBook Air M2: "possible resolutions to the Fermi paradox, the question of where is the evidence for extra-terrestrial civilizations: - " answer: possible resolutions to the Fermi para
562.
▲
by
tomp
4y ago
Fines are just “cost of business” for companies. Either bankrupt the company (fine is 20% of yearly revenue) or jail the executives and everyone responsible.
563.
▲
by
tomp
4y ago
Sounds like there’s a gap in the market. I’m hugely enthusiastic hobbyist that would love to chat more about robotics, in particular how a hobbyist could get started with it (a robot arm + camera maybe?). I’d love to buy you virtual coffee,
564.
▲
by
tomp
4y ago
Still doesn't explain why green would be better than blue . Don't all problems go away if you're just using blue?
565.
▲
by
tomp
4y ago
Isn't color blindness mixing up red & green? So why would a green design be better than a blue design? Naively, it's strictly worse.
566.
▲
by
tomp
4y ago
Is it just me, or is this a fancy article with nice animations full of marketing speak, but doesn't actually mention the reason why they removed their previous blue brand/color?
567.
▲
by
tomp
4y ago
It’s not just a “functional programming language”, it’s a lisp! https://mobile.twitter.com/haxor/status/1580959518984241158 It will never be popular.
568.
▲
by
tomp
4y ago
Which of these basic hypotheses supports reparations? I'm pretty sure there needs to be a 3) blacks are poorer as well.
569.
▲
by
tomp
4y ago
Employment is, be definition, exclusive. More so for well-paying companies. I hate when companies lie about that, it massively lowers my respect for them.
570.
▲
by
tomp
4y ago
This is a general question for the whole "social justice" movement. If you're concerned about the wellbeing of black people, and you justify your concern with statistics pointing out that they're disproportionately poo
More ›