It sounds like only a hosted service was compromised, not the open source OpenCode binary that users run on their own machines. So this wouldn't explain that.
There's no self-serve endpoint, but OpenCode could send the inference providers the leaked keys and ask for them to be revoked, which is a thing the inference providers can do. It's possible that this'll happen, apparently remediation is still in progress.
Well, if you write a library in the scriptc subset, it will also work in regular JavaScript runtimes, for free. (Unless you rely on, e.g., catching runtime errors thrown by failed casts, but that's easy enough to avoid.) That seems like a clear advantage over creating a new language.
What I do think would be nice would be if all the tweak-JS-semantics-for-optimization-purposes projects (scriptc, Static Hermes, Nova, possibly Closure Compiler) got together and agreed on a minimal set of rules that TypeScript code should follow in order to be compatible with all of them. Then there could be a push for commonly used npm packages to adhere to those rules, and advertise that they do so in order to be fully optimized.
Those offer a more streamlined developer experience than traditional cloud VMs, but still require just as much compute resources (apart from being able to scale to zero if you can tolerate cold starts). E.g., you still have to choose how many data centers they run in and which ones, which is laden with tradeoffs, whereas every Cloudflare Worker runs in every Cloudflare DC, for free.
Running Rust code in V8 via WebAssembly works today, though the tooling could benefit from a bit more polish. Go sort of works, but less well; IIUC fixing this would require adding more features to Wasm, most especially interior pointers, so that its garbage collection behavior can be made semantically compatible with Go's.
From the article: "He was also at pains to point out that the reverts were not down to 'vibe coding' in development, as the technique was not available when most of the new features were written."
From the announcement: "Deno Deploy will continue operating for six months before shutting down. We will provide migration support for paying customers moving to Cloudflare Workers."
If it were deterministic but chaotic, which AFAIK is the case for LLMs running at temperature 0, then you would have to likewise check every time you changed anything and couldn't meaningfully correct for it.
Julia is not actually doing anything unusual here. When their documentation talks about UT1, this is just a fancy/confusing way of saying that their datetime library isn't aware of leap seconds, just like every other mainstream programming language AFAIK. The abolition of future leap seconds will only reduce the number of weird edge cases this generates, which is all to the good. I suppose they'll need to change the docs, though.
The concept of determinism is useful in computer-science contexts other than automata theory. See, e.g., https://en.wikipedia.org/wiki/Deterministic_algorithm. It's not incorrect to use the term to refer to those things.
The actual rationale here is that it's not worth trying to figure out whether a hypothetical leap hour in a thousand years would be worth it, or how to pull it off if so, because no advance preparation for that scenario that we can do today has enough value to outweigh our ignorance of what the world will look like then. If they decide not to do it, they can just change the rules again; there'll be plenty of advance notice.
My read of the author's position is that the people who built Mononoke at Meta did all the hard engineering work that made Scion possible, while the work that happened at Anthropic was mere code-monkeying unworthy of note. This is decidedly not the sense you would get from reading Anthropic's blog post, so they feel that that post misrepresented what happened. I am skeptical that this is really an accurate characterization, but I can see how a person might get there, especially if they were involved in the work at Meta and so have a natural bias towards weighting its significance more heavily.
Programming languages take decades to get traction. We've had LLMs capable of real software engineering for less than two years. It's way too early to make pronouncements about this.
That's not really relevant; the U.S. government has enough leverage over the global (fiat) financial system that they can cause any account in any First World country to be shut down.
The actual reason to discount this is that account shutdowns are flashy rare events that reliably summon HN outrage, and so the kind of person who posts here overweights how likely they are, at least for people not on the margins of society. (If that weren't the case, the system wouldn't work and normal people wouldn't use it.)
The safety net makes you less secure against the U.S. government, but more secure against everyone else. That's the right choice for almost all users, but I still think it's at least kind of good that people have a choice.
The problem turned out to be that I had a very outdated curl-bash installation of uv on my system that was shadowing the up-to-date package-manager one. This is why curl-bash installation is evil.
Oddly, when I tried running this, it used version 3.15.0a2, which made some of the puzzles not work because the required new features hadn't been added yet.