Recent submissions

Comment · 5 months ago
Comment · 7 months ago
Worth noting that this is a bound on arbitrary search, but there exist some problems with structure (e.g. integer factorization) for which quantum algorithms are exponentially faster than known classical algorithms (a problem believed to be in NP and BQP but not P).
Comment · 8 months ago
Comment · 9 months ago
nethack.alt.org is conspicuously absent...
Comment · 9 months ago
Also worth noting that the verifier is under active development not only to verify more legitimate programs, but also to reject programs with exploits and side channels (and there are runtime defenses too, like dead code elimination and ALU sanitation).
Comment · 11 months ago
I worked a tiny bit on this experiment so I'm a bit biased, but it really is quite an impressive demonstration. I don't know if the methodology has changed since I was on the project, but in the past there was a whole process of sneakernet-ing the data from the game across an air gap and onto the ISS.
Comment · 11 months ago
The discussion of Richard Caley was warming, reminded me of Izchak Miller a little bit.
Comment · 1 year ago
Comment · 1 year ago
That's an excellent question, I thought the exact same thing when I wrote this article. From what I can tell [child](https://github.com/NetHack/NetHack/blob/ed600d9f0f3c37677418...) properly spawns a child process with the real uid/gid instead of the effective uid/gid. And to your point, there are actually a decent number of [CVEs](https://www.nethack.org/security/) due to NetHack being set[ug]id.

Live from the official Hacker News API

A modern reader for Hacker News · SvelteKit + Skeleton