Recent submissions

Comment · 6 months ago
You can maybe, trust the user to handle it's own certificate in their own devices? Though I admit requiring attestation is probably a good default.
Comment · 7 months ago
Because it increases US workers relative (to other countries) wage. Though with current automation levels this may be a lesser problem.
Comment · 7 months ago
This is beautiful, incredibly sane, and awesome reference material. There's no way I'd use a 3500 lines init.el or most of the extras, but somehow I feel like a good chunk of the stuff here should be upstreamed if we one day consider it reasonable to change default behaviors in a major update.
Comment · 8 months ago
False, cf. ancient Athens.
Comment · 8 months ago
This is good and means no one can impersonate you using your phone number, but doesn't solve the recurring costs issue, you still need to buy a new number when someone registers yours, and every financial transaction puts you at more privacy risk. And is terrible UX, imagine having to add your contacts new numbers every other week.
Comment · 8 months ago
And what happens when the next guy buys that same number and registers on Signal?

Phone numbers are recurring costs. And to keep a truly private one you must keep paying without ever disclosing personal info and that is really hard. Signal is a privacy nightmare for long term use.

Comment · 10 months ago
We definitely need more humility. For starters, not to casually dismiss beliefs held by millennia due to some artifacts without even specifying them.
Comment · 11 months ago
>On paper it should be a good thing

Not really. Most people have terribly low time preference. Democracy for example is a very bad idea when you account for that (read Hoppe for a detailed explanation). Public company ownership is much better because it doesn't suffer from one vote per person, but still susceptible to much of the same management problems, specially in a society that already favors lower time preference by other means.

Comment · 11 months ago
FWIW, all of those links compare Rust to languages created before 1980, and are all projects largely and unusually independent of the crates ecosystem and where dynamic linking does not matter. If you're going to use a modern language anyway, you should do due diligence and compare it with something like Swift as the ladybird team is doing right now, or even a research language like Koka. There is a huge lack of evidence for Rust vs other modern languages and we should investigate that before we lock ourselves into yet another language that eventually becomes widely believed to suck.
Comment · 11 months ago
Disservice? Rust is taking over the world while they still have nothing to show basically (Servo, the project Rust was created for, is behind ladybird of all things). Every clueless developer and their dog thinks Rust is like super safe and great, with very little empirical evidence still after 19 years of the language's existence.

Zig people want Zig to "win". They are appearing on Hacker News almost every day now, and for that purpose this kind of things matters more than the language's merits themselves. I believe the language has a good share of merits though, far more than Rust, but it's too early and not battle tested to get so much attention.

Comment · 11 months ago
Author is definitely correct in pointing out the incentives for companies to buy hardware. What the article misses is that there is in fact a reasonable economic incentive to not invest in software even if LLMs were not an economic bubble. It is that every single company is developing the same thing, there are many of those who even develop them as open source, and the ones that are closed as well as any company who would hire this guy, have a bunch of industrial spies inside anyway. Buying hardware may increase your moat, but developing software just rises the sea level.
Comment · 11 months ago
Comment · 11 months ago
Just type C-h t (help-with-tutorial) and work your way through it.
Comment · 1 year ago
It's more about allowing a-library-fits-all than forcing it. You don't have to ask for io, you just should, if you are writing a library. You can even do it the Rust way and write different libraries for example for users who want or don't want async if you really want to.
Comment · 1 year ago
Why didn't crates.io maintainers apply the patch themselves? NPM does meddle with packages when an incident happens like they did with left-pad.

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton