News for hackers. Stuff that matters.
Re:The C standard formalized in Coq by pron on Monday November 30, 2015 @08:51AM (#10649319)
> To say that type-theoretic theorem provers (or the Coq system specifically) are merely academic toys compared to TLA+ is disingenious.

I never said that Coq is an academic toy. Some projects done in academia and used in the industry have been verified with Coq, but I am not aware of actual (direct) use of Coq in the industry (but would be interested to learn of any).

> I'm not sure I understand why "Computer Science-type math" is an unreasonable foundations for a program specification tool, which is what TLA+ is

I've never said it is unreasonable, only that it is foreign to most engineers (as well as most academic CS people), or rather, more foreign than basic logic/set theory.

I would say again that if you're comfortable with type theory, you might enjoy Coq better. If not, TLA+ would probably be more approachable. TLA+ has the added advantage of having a model checker, which is very useful when you don't have the resources required for a full proof, but still want a good confidence in the correctness of your algorithm.

> A formalization of the C standard is a HUGE undertaking, independent of what tool was used. Let's celebrate that achievement!

Oh, absolutely! It's just that stories like this make people interested in formal methods, and Coq, IMO, would be daunting for most. TLA+ is a good entry point to formal methods, even if you later choose to move to dependent-type approaches and tools.

Re:The C standard formalized in Coq by yaantc on Monday November 30, 2015 @09:01AM (#10649372)
> ... but I am not aware of actual (direct) use of Coq in the industry (but would be interested to learn of any).

The first EAL7 certified smart card certification used coq, see: https://www.commoncriteriaportal.org/iccc/9iccc/pdf/B2404.pd...

Re:The C standard formalized in Coq by pron on Monday November 30, 2015 @09:12AM (#10649431)
Oh, wow, they have verified a (JavaCard) JVM. Impressive! Thanks.

They mention > 117,000 lines of Coq with > 1600 proofs to get EAL7 for the JVM.

Re:The C standard formalized in Coq by brohee on Monday November 30, 2015 @09:21AM (#10649490)
This link is about an EAL4+ evaluation
Re:The C standard formalized in Coq by wolfgke on Monday November 30, 2015 @09:28AM (#10649536)
If you read slide 4, you'll see that the SIM card was evaluated at EAL4+ level and the design of the Java Card System at EAL7 level.