Thread by @eevee

university of minnesota today coming out with groundbreaking research that the best C programmers in the world can't tell if you're giving them bad C

maybe C is bad

"actually c is fine if you just never make any mistakes" Actually chris the developers of the god damn linux kernel are not even this superhuman so i seriously doubt you are

but also to be clear, deliberately trying to sneak vulnerabilities into someone else's software without their knowledge is pretty fucked up, what

and to the kernel's credit they are experimenting with rust /right now/

all our most critical infrastructure is written in a language where if you swap the order of two lines, a very common failure case is "let anyone on the internet run whatever code they want on your machine"

fair point — the paper is not so much "wow, look how easy it is to get bugs into linux" as it is "wow, look how easy it is to lie when you already have a good reputation"

though i suppose nothing guarantees that an arbitrary student is trustworthy

https://twitter.com/11rcombs/status/1385004181967175685

checking with the IRB about a potential experiment where i put on overalls and go to someone's house saying i'm there to fix a leaky faucet, and i do fix it, but before i leave i also set their kitchen on fire. no humans are involved, only a house

look y'all have got to stop responding to "X facilitates serious problem Y" with "well it's impossible to eliminate Y entirely" as though that has anything to do with anything. i can't eliminate water damage entirely but my house still has a fucking roof