News for hackers. Stuff that matters.
Re:11 year old unsoundness bugs found in Rust borrow checker by vrighter on Tuesday October 13, 2026 @08:56AM (#50070949)
not of the implementation. Of the type system. And a verification of subsets, but not a verification of their interactions == the whole system is unverified.
Re:11 year old unsoundness bugs found in Rust borrow checker by afdbcreid on Tuesday October 13, 2026 @09:24AM (#50071344)
The eventual goal is verification of the full system. Not the implementation, yes, that's infeasible. But this will mean all soundness bugs will be either found and fixed or we'll know how to fix them once they're found. In any way they won't be fundamental (which is believed to be the case today, but not proven).
Re:11 year old unsoundness bugs found in Rust borrow checker by vrighter on Tuesday October 13, 2026 @09:55AM (#50071821)
so why put them in before knowing if they work reliably? That's the YOLO part, to me
Re:11 year old unsoundness bugs found in Rust borrow checker by cscheid on Tuesday October 13, 2026 @10:44AM (#50072550)
The same combination of reasons we've been doing these kinds of heuristic implementations forever:

- people still get value off of an incomplete implementation, because we are allowed to be mindful of the issues (see, eg, UB in C++). In other words, many, many more bugs have been prevented by the borrow checker working as intended than have been caused by it working poorly. It's not particularly close.

- theory almost always moves slower than practice

The mutable state situation pointed out by the grandparent poster was a famous issue in typing SML under mutability that required serious changes in the language, see Section 4.2 here: https://smlfamily.github.io/history/SML-history.pdf

These were Turing-award caliber people figuring it out and, according to your definition, "yolo"ed it. This stuff is _hard_, and these driveby comments truly aren't helpful.

Re:11 year old unsoundness bugs found in Rust borrow checker by vlovich123 on Tuesday October 13, 2026 @10:46AM (#50072585)
Because there’s real value delivered even before that? Perfect bring the enemy of good and all that
Re:11 year old unsoundness bugs found in Rust borrow checker by dwattttt on Tuesday October 13, 2026 @10:43AM (#50072547)
You're claiming that this bug that reproduces back to Rust 1.4 (2015) is a sign they're adding too many features to make sure they don't introduce bugs? Or do you have bugs that are due to these recent features?
Re:11 year old unsoundness bugs found in Rust borrow checker by streetfighter64 on Tuesday October 13, 2026 @09:09AM (#50071144)
It holds that ¬(∀subsets: isverified(subset)) but it's not true that ∀subsets: ¬isverified(subset), which it think most people would interpret "the whole system is unverified" as
Re:11 year old unsoundness bugs found in Rust borrow checker by vrighter on Tuesday October 13, 2026 @09:59AM (#50071886)
If I independently verify that subsystem A properly consumes apples, and that subsystem B is properly emitting oranges, but never verify that subsystem A is receiving apples, and just connect subsystem B to it, then yeah, the system is not verified.
Re:11 year old unsoundness bugs found in Rust borrow checker by samus on Tuesday October 13, 2026 @11:42AM (#50073500)
That's not the issue though.
Re:11 year old unsoundness bugs found in Rust borrow checker by cmrx64 on Tuesday October 13, 2026 @11:44AM (#50073534)
you’re deeply confused but the words “validation” and “qualification” may help untangle the web, there’s a lot more to assurance than what your model seems to hold (and you’re drawing batshit deductions from it)