amluto

50 ms·

871.
by amluto 10mo ago
GnuPG should be extended (incrementally rewritten into something much better and turned into a library) and the original GnuPG should be extinguished.
872.
by amluto 10mo ago
> you have to supply people's public key each time Keyrings are awful. I want to supply people’s public keys each time. I have never, in my entire time using cryptography, wanted my tool to guess or infer what key to verify with
873.
by amluto 10mo ago
Passing an &uninit to a combinator seems like it would be strange. But if you told me that some strongly typed language wanted to have coroutines and futures, that coroutine bodies would not execute at all until first polled, and that i
874.
by amluto 10mo ago
Most of the problems mentioned in the article are not problems with using a content-addressed tree like git or even with using precisely git’s schema. The problems are with git’s protocol and GitHub’s implementation thereof. Consider vcpkg.
875.
by amluto 10mo ago
> For 1, I think it's hard to make a distinction between swapping an object, vs. swapping/mutating all of its fields such that it becomes equivalent to a different object. Privacy. If an object has fields I can’t access, but I
876.
by amluto 10mo ago
Some things I occasionally contemplate: (not that I’ve ever tried to work any of these out anywhere near completely) 1. Why isn’t there a variant of &mut that doesn’t allow swapping the value? I feel like it ought to be possible to len
877.
by amluto 10mo ago
Don’t use recursive page tables. - What if you want to port to an architecture that can’t do them? - How sure are you that all the bonus bits in the entries line up right? - What if you want to use huge pages of various sizes? - What if you
878.
by amluto 10mo ago
There are all kinds of reasons. One basic example would be if the vendor has a data sheet but does not have the right to grant anyone redistribution rights to the datasheet or may even be restricted from allowing anyone to read it without a
879.
by amluto 10mo ago
This is definitely not true. It’s even sometimes possible to negotiate contract and NDA terms with Large Corporations for the specific purpose of producing open source code based on NDA specs. Source: been there, done that.
880.
by amluto 10mo ago
It’s right there, clear as mud: from wifi_densepose import WiFiDensePose # Initialize with default configuration system = WiFiDensePose() # Start pose estimation system.start() # Get latest pose data poses = syst
881.
by amluto 10mo ago
I'll add some nuance here. In a classical computer, computing with coded bits is not particularly difficult. We've known how to do it with some degree of mathematical rigor for decades -- IIRC John von Neumann was interested in
882.
by amluto 10mo ago
I don’t really expect fancier codes to cause a huge jump in the number of logical qubits. At the end of the day, there’s some code rate (logical bits / physical bits) that makes a quantum computer work. The “FOOM” is the transition fro
883.
by amluto 10mo ago
I find the backward version slightly more intuitive. Here’s why: Suppose I want to uniformly randomly shuffle a deck of cards in a single pass. I stick the deck on the table and call it the non-shuffled pile. My goal is to move the deck, o
884.
by amluto 10mo ago
I'm amused by the burying of the non-lede until half way through the paper. I, too, can maintain a 59-bit repetition code for over two hours on my quantum laptop (it's just a normal laptop, but it definitely obeys the laws of qua
885.
by amluto 10mo ago
HTTP-only cookies ought to work fine for this. I imagine there’s a fair amount of complexity that would need to be worked out, mostly because the browser doesn’t know the suborigin at the time it makes a request. So Sec-Fetch-Site and all t
886.
by amluto 10mo ago
Or sending Content-Security-Policy: script-src 'none' for everything that isn’t intended to be a document. Or both. IMO it’s too bad that suborigins never landed. It would be nice if Discord’s mintlify route could set something li
887.
by amluto 10mo ago
I've had very impressive success running upstream OpenWRT on TP-Link hardware: I have Archer C7 access points running with literally years of uptime. That being said, for any new application, I suggest using at least an 802.11ax AP, be
888.
by amluto 10mo ago
> The point of a source tarball, as I understand it, is to be sure of what source you're building Perhaps, in the rather narrow sense that you can download a Fedora source tarball and look inside yourself. My claim is that upstream
889.
by amluto 10mo ago
I’m only familiar with the Fedora system. The build is hermetic, but the source input come from fedpkg new-sources, which runs on the client used by the package developer.
890.
by amluto 10mo ago
I’m not sure why this would make a difference. The only thing special about the head is that there is a little file (that is not, itself, versioned) saying that a particular commit is the head.
891.
by amluto 10mo ago
This seems no worse than GitHub Actions executing whatever random code people upload. It’s not so hard to do a pretty good job, and you can have layers of security. Start with a throwaway VM, which highly competent vendors like AWS will se
892.
by amluto 10mo ago
Shallow clones are a thing. And it’s fairly straightforward to create a tarball that includes enough hashes to verify the hash chain all the way to the commit hash. (In fact, I once kludged that up several years ago, and maybe I should dus
893.
by amluto 10mo ago
For most well designed projects, a source tarball can be generated cleanly from the source tree. Sure, the canonical build process goes (source tarball) -> artifact, but there’s an alternative build process (source tree) -> artifact
894.
by amluto 10mo ago
> For example, currently most Debian git repositories base their work in "pristine-tar" branches built from upstream tarball releases I really wish all the various open source packaging systems would get rid of the concept of s
895.
by amluto 10mo ago
> Perhaps you don't even need access to the state of anything except the penultimate hidden layer, since based on my vague reading of [3] the residual stream doesn't "lose information" as it passes deeper down the att
896.
by amluto 10mo ago
This design implicitly does something similar to something that I sometimes think conventional transformers should try: allowing later layers to query the KV data from earlier layers. As far as I can tell, with a conventional transformer, i
897.
by amluto 10mo ago
Whoops, that was a typo. Gas stoves really are kind of absurdly inefficient, though. You can buy silly pots with heat exchangers on the bottom that do better, though.
898.
by amluto 10mo ago
The socket API is all kinds of bad. The way streams should work is that, when sending data, you set a bit indicating whether it’s okay to buffer the data locally before sending. So a large send could be done as a series of okay-to-buffer wr
899.
by amluto 10mo ago
Excuse me? This is the basic rate: https://www.pge.com/tariffs/assets/pdf/tariffbook/ELEC_SCHED... You pay $0.40317/day for the connection but you get back $58.23 twice per year. That’s $30.70 per
900.
by amluto 10mo ago
The US average residential electricity price is 18.07 c/kWh [0]. Natural gas is $15.39/thousand cubic ft [1]. 1k cubic feet of gas is about 300kWh (this varies because natural gas is not always the same and because the higher he

More ›