Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
plonk
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
8 ms
·
31.
▲
The White House May Condemn Musk, but the Government Is Addicted to Him
(nytimes.com)
5 points
by
plonk
3y ago
|
2 comments
32.
▲
by
plonk
3y ago
How much did Blackwater do? Did they actually fight or just guard?
33.
▲
by
plonk
3y ago
Everyone can make disasters at the wheel though. If you didn’t mandate insurance, there would be victims with no recourse.
34.
▲
by
plonk
3y ago
This isn’t about enjoying the explanation now, it’s about the academic success and open doors one could have had if their teachers had been better communicators. Math doesn’t get studied in a vacuum, it’s the biggest selection criterion for
35.
▲
by
plonk
3y ago
What do you mean by asymmetry?
36.
▲
by
plonk
3y ago
The elephant's reaction to fear of humans might not be what you expected though? Do they flee humans more often than they charge?
37.
▲
by
plonk
3y ago
Do they not feel the sting or are they immune to the venom?
38.
▲
by
plonk
3y ago
I’m curious about why you keep living there despite what you describe. Does it affect your life a lot?
39.
▲
by
plonk
3y ago
Is there typically a “cold restart” plan to rebuild the whole infra from scratch? I’m thinking of things like circular dependencies when services boot.
40.
▲
by
plonk
3y ago
I'm looking for actual evidence or reasoning. Of course I have nothing against the US government strong-arming others into using the dollar, but I don't think you have a lot supporting that hypothesis either. The dollar's pre
41.
▲
by
plonk
3y ago
Government spending both increases the government's debt and causes inflation. Maybe that's the link they are looking for. Plus, I don't know much about monetary policy, but wouldn't the Fed feel obliged to start quantit
42.
▲
by
plonk
3y ago
> Your precious ML/DL libraries may not even be upgraded, because writing neural network code is not the same as writing thread-safe code. If it comes from a CS lab, its authors have already left, and there's nothing worthy of
43.
▲
by
plonk
3y ago
Because interest rates are increasing? Both the cost of debt and the cost of the dollar increase.
44.
▲
by
plonk
3y ago
Lebanon and Pakistan aren't exactly politically stable, Greece was incompetently managed for a long time. Don't know about Turkey. Compared to the national deficits of basically everyone in the EU, US, and big Asian economies, it
45.
▲
by
plonk
3y ago
> It's not a choice if the consequences of making the 'wrong' choice is being invaded, bombed into the stone age and have your country pillaged and its assets handed off to reserve currency alligned corporations. Is there
46.
▲
by
plonk
3y ago
I kind of thought Python would enable the GIL if any no-nogil libraries were installed , but even that is hard to define when you can modify sys.path with environment variables and code. Will look it up.
47.
▲
by
plonk
3y ago
In my important dependencies, there are deep learning frameworks from billion-dollar companies, bindings for C++ libraries that are basically standard in the field, projects from CS labs with millions of users. I don't see any of them
48.
▲
by
plonk
3y ago
I think the point was to let libraries declare whether they support nogil mode (opt-in), and your program would only run with no GIL if all the dependencies allow it? So they have all the time in the world to iron out those bugs.
49.
▲
by
plonk
3y ago
Suppose you’re building a feature that requires $newlib as a dependency. How is that handled? Do you request the addition of $newlib and wait for a new remote image to be deployed for everyone before you can work on your feature branch?
50.
▲
by
plonk
3y ago
What does a bank write if it's in a good place and the rumors are baseless? Or do you just leave the bank regardless of that to be sure?
51.
▲
by
plonk
3y ago
Search engines are fucked in that scenario. Maybe we'll just need better-curated ones, or a better ranking (usefulness ratings by verified users?). And anyone can still create a website, so you can escape the AI flood. But if Google fa
52.
▲
by
plonk
3y ago
What your mental plan is doesn't matter though, as long as you actually made a reasonable effort towards what the VCs expect. Or does anyone actually tell them "invest in me because I'll overtake Nvidia or die trying"? I
53.
▲
by
plonk
3y ago
"I initially planned to scale but eventually realized the company wouldn't make enough to survive, as happens to most startups. I chose to scale down and make the company profitable instead of killing it." Could the VCs reall
54.
▲
by
plonk
3y ago
> Uh, no. The reason why you need a "conscious" universe is precisely beause humans aren't special. In a conscious universe model, the brain "abuses" the laws of physics including whatever laws that relate to con
55.
▲
by
plonk
3y ago
That's a weak argument. The reason some of us don't believe that consciousness is something else than the state of a brain following physical processes is that the latter is a much simpler theory and matches every objective observ
56.
▲
by
plonk
3y ago
I said patterns, not literal parts of images.
57.
▲
by
plonk
3y ago
Generative models are still trained with real data. That output isn't really synthetic, it contains patterns from existing images of victims. Making that legal to share doesn't sound right.
58.
▲
by
plonk
3y ago
Why would inflation push people into banks? In my (EU) country, banks don’t offer anything that matches current inflation besides speculative investments.
59.
▲
by
plonk
3y ago
No, the only versioning is “old models still exist in the bucket and their IDs stay in the docs”. We might need to add a database with more structured attributes for each model, but until now it hasn’t been a real problem.
60.
▲
by
plonk
3y ago
Models that actually get deployed get a random GUID. Our docs tell us which is which (release date, intended use, etc.) Models are then stored in an S3 bucket. But since the IDs are unique, they can be exchanged and cached and copied with n
More ›