vlovich123

5 ms·

91.
by vlovich123 2mo ago
If you’re talking about the simulator, no. That doesn’t virtualize - it literally just runs it locally on your machine. Less RAM overhead and CPU since it uses your normal instruction set and doesn’t run a separate kernel and userspace.
92.
by vlovich123 2mo ago
I’ve read the readme a few times and still have no idea how this is useful. The motivating example of Sort => SortBy has no description of what’s highlighted and no description of how. Is it running the old test suite with the new change
93.
by vlovich123 2mo ago
The government spent drastically more money on the space shuttle (which is widely acknowledged as a mistake of a design) than what the private industry ended up delivering. It’s rent vs build and I’m not so sure the government benefits so m
94.
by vlovich123 2mo ago
That’s because infra telecom is done incorrectly. Look at Sweden’s municipal fiber model how you make infrastructure cheaper and make competition actually work for customers. Having each carrier have to run lines to each customer is the wro
95.
by vlovich123 2mo ago
Is it just me or does this create a huge potential security vulnerability where it’s a lot easier to trick the application into mutating itself. There’s also the alternate problem where if the file is placed in a privileged location, you wo
96.
by vlovich123 2mo ago
I would posit there’s no way in hell they’re getting sufficiently cheaper on a short enough time frame vs how much demand is sky rocketing. AI companies are seeing quarterly doubling of revenue if not more.
97.
by vlovich123 2mo ago
All Jevon’s paradox says is that as a resource becomes cheaper total consumption of that resource increases. It applies equally well to the inputs of token production as it does to the tokens themselves. The former would describe the effect
98.
by vlovich123 2mo ago
There's a lot to criticize, but this argument seems extremely misleading to me. The government wants to put things into space. SpaceX says "this is how much we're willing to take to do so" and bids on the government cont
99.
by vlovich123 2mo ago
> Please don't pick the most provocative thing in an article or post to complain about in the thread. Find something interesting to respond to instead.
100.
by vlovich123 2mo ago
Did you actually read the article? It lays out a pretty compelling case that the CIA established funding sources and magazines to popularize American modern art and make it a worldwide phenomenon / a cultural export to have Americans b
101.
by vlovich123 2mo ago
The lack of mmap is a problem, especially given how dynamic link heavy Linux distros are. It’s going to cause a lot of extra RAM usage and the kernel won’t know to dump those pages efficiently and will swap them to disk. The extra RAM usage
102.
by vlovich123 2mo ago
Setting aside the morality of these vs other force options, these run at $1650/pair if I’m reading it correctly - $66k for 40. Anyone else find it like an insane amount of money for a device that’s so simple?
103.
by vlovich123 2mo ago
> because the node you hit is rolling/broken/overloaded, but the others aren't Retries in such a situation should be handled internally with the client at most responsible for failing over with a circuit breaker to another
104.
by vlovich123 2mo ago
It doesn’t take away from the point that build time often has access to secrets the runtime does not.
105.
by vlovich123 2mo ago
So? Runtime code requires actually executing the malicious code path which isn’t an immediate 100% hit rate for everyone that includes it in the dependency chain. For build.rs it’s a 100% compromise of everyone it’s in the dependency chain
106.
by vlovich123 2mo ago
Defaults matter. It’s nice you can set this up using a plugin to protect yourself, but that doesn’t protect the ecosystem, most of which doesn’t use cargo deny. I also disagree a build scripts is a mild convenience. A build script always ru
107.
by vlovich123 2mo ago
Disallow lists are ineffective - better to disallow by default and require opt in. Also, crates.io can defer serving up newly uploaded scripts that have a new build.rs / proc-macro dependency and warn publicly that a version introduces
108.
by vlovich123 2mo ago
I’m disappointed crates.io doesn’t have a stricter bar for serving a crate that has newly acquired a proc macro or build.rs. That seems like a trivial mitigation.
109.
by vlovich123 2mo ago
They can get RAM at drastically more expensive rates. And not all critical applications have unlimited budgets to absorb the increase. It’s the reverse Jevon’s paradox.
110.
by vlovich123 2mo ago
Did the distro apply a patch to glibc that broke things that others did not? Or did they pick up a newer version of glibc that other distros hadn’t? Yes, different distributions run on different versions of glibc. And different versions of
111.
by vlovich123 2mo ago
Not sure what you’re trying to show with that bug report but it’s not a case of cross distro glibc issues. If I read correctly it’s a vanilla behavioral change that exposed preexisting UB in flash. Not sure how the comments about alpine or
112.
by vlovich123 2mo ago
I haven’t heard of this and I don’t think you’re right. Glibc, for all its faults, as a general rule does backward compatibility well. The problem is if you compile against a newer glibc (common in CI by default) and try to run on a distro
113.
by vlovich123 2mo ago
That’s a bold statement to make in the 21st century when nearly everything is run on computers, including cars, planes, hospital networks, etc etc.
114.
by vlovich123 2mo ago
I have this problem with an 8gib 2080. Nvidia drivers have bugs with Wayland or Vulkan which is at least partially a memory leak. Here’s one for example: https://forums.developer.nvidia.com/t/kwin-6-7-x-causes-mass...
115.
by vlovich123 2mo ago
No, that’s a common misconception. MoE despite the name doesn’t have an “expert” slice that’s an expert in any particular thing. It’s just trained models such that a subset of all memory weights is active on a single token. Often, the weigh
116.
by vlovich123 2mo ago
He says: > From that position, the difference between a ten cent part and a one dollar part is not a rounding error and it is not a detail you get to wave past on the way to the interesting discussion about encodings Yet earlier: > I
117.
by vlovich123 2mo ago
All I’m saying is that lattices are newer and more complex. Bernstein is proposing a lattice scheme AND saying it should be hybrid. ALSO. The proposal is to replace cyclotomics with Gallois field precisely because they are better understood
118.
by vlovich123 2mo ago
My point is not about a specific algorithm or application. I’m talking about the family elliptic curve vs lattice. Bernstein isn’t asking people to run hybrids because the mathematics is simple and well understood. Also Curve25519 is just a
119.
by vlovich123 2mo ago
We seem to be reading different history books. The first elliptic curve paper (1985) precedes the first lattice paper (1996) by 11 years. 2005 is when LWE was published which provided the first theoretical foundation to construct lattice en
120.
by vlovich123 2mo ago
Why do you then argue from both sides that no one is proposing lattice only for standardization while simultaneously clearly pushing that it should be lattice only? And clearly there is a concern for the lattice only recommendation, even if

More ›