rcxdude

12 ms·

31.
by rcxdude 23d ago
Yeah, this is almost the worst way they could choose to 'fix' the problem.
32.
by rcxdude 23d ago
I shouldn't need to drop to assembly to get an infinite loop that works!
33.
by rcxdude 23d ago
The CPU doesn't really see functions, it just sees instructions. Functions are a convention on top of the machine code. What happens in this case is the compiler emits essentially a malformed function: it ends without performing a retu
34.
by rcxdude 23d ago
If it's declared as static, maybe (well, usually, in my experience. You'll also usually get an unused warning). Otherwise the compiler can't assume some other compilation unit won't want it. Linkers can perform a garbage
35.
by rcxdude 23d ago
Also, the visualisations are to some extent an artifact of how you arrange the equation: rearranging terms between the left and right side won't change the line but it will change these visualisations. The first example illustrates it
36.
by rcxdude 23d ago
If you're doing that you might as well use public/private keys. The issues with passkeys are around user experience and education about how they are stored, and a lack of good backup options like enrolling a not-present device as
37.
by rcxdude 23d ago
Which is a laudible goal, but something that makes "oh just fall back to the standard insecure recovery process" not a very good response to the concerns about users being able to actually hold on to their passkeys easily and reli
38.
by rcxdude 23d ago
If you're using a USB hardware token your knowledge of it is at least an order of magnitude better than the median user's. I know where my passkeys are stored, I don't know where my family member's passkeys are stored
39.
by rcxdude 23d ago
Not always. And which vault? There can be multiple on a given device. This isn't some hypothetical 'mollify the user's worries' question, this is an important practical question of what do they need to worry about losing
40.
by rcxdude 23d ago
Ah yeah, Microsoft has this awful habit of just hiding the buttons for actions you don't have permission to do, which makes me feel like I'm being gaslit sometimes when a guide says 'there's a button right here' and
41.
by rcxdude 23d ago
Cool, where are they stored? (I know the answer: 'it depends', and that's the big problem with their usability: most users haven't a clue what the answer is and most tech support can't answer that question straightf
42.
by rcxdude 23d ago
I've seen it. Websites are pretty inconsistent about the flow but I've had a few just try to install a passkey when I log in (amazon does it. Paypal seems to do it but only try to use it on the same browser it was installed on, ev
43.
by rcxdude 23d ago
They're also not supporting device attestation which would allow websites to insist on particular implementations of passkeys.
44.
by rcxdude 23d ago
To me it seems easier than proving the code does something useful without pinning down what that actually is.
45.
by rcxdude 23d ago
Superhuman does not mean actual superpowers. You still need a path to victory and it's not obvious that there's one that we're just too dumb to see.
46.
by rcxdude 24d ago
You're right that this is a generational thing, but the expectation for your kids to move out like this is also relatively recent: historically it was a lot more common for your children to continue to live with you, or if not, to live
47.
by rcxdude 24d ago
Yeah, it's more that they can, if designed correctly, be made very difficult to clone. (and that if in the middle is pretty important!)
48.
by rcxdude 24d ago
Well yeah, why would they offer an legal guarantee if they don't have to? They could have a watertight proof that their software works perfectly in all situations (lol) but the lawyers still would prefer to disclaim as much liability a
49.
by rcxdude 24d ago
Tests are guaranteeing that the product doesn't fail under the test conditions. Formal verification is guaranteeing the product doesn't deviate from the formal spec under a given set of assumptions. Both of them are useful but dep
50.
by rcxdude 24d ago
the vast majority of such annotations are checked though. There's generally three kinds of annotations in formal proof systems: assertions about inputs that cannot be checked by the system, statements of propositions that want to be
51.
by rcxdude 24d ago
Claude's style is obvious enough I don't think you really need the checker.
52.
by rcxdude 24d ago
Indeed. This feels like it's just slut shaming.
53.
by rcxdude 25d ago
A paper plane does have some important similarities to a full-sized aircraft, though. I don't think 'biological brains are really complex' makes it obvious that an LLM is conscious or not.
54.
by rcxdude 25d ago
Good luck convincing others of these truths you know without evidence or argument, though.
55.
by rcxdude 25d ago
That's pretty much the definition that the term started with (arguably it was a bit more general in some sense: just along the lines of 'politically aware and engaged'). It seems to have morphed, but I find that when it'
56.
by rcxdude 26d ago
I think a lot of the actual issues they're encountering are that the underlying POSIX APIs have a lot of sharp edges which the older versions of the tools have had enough time to work around, while the newer ones are generally running
57.
by rcxdude 26d ago
but GPLv2
58.
by rcxdude 26d ago
Probably the largest part is that this kind of code quality is not that important to a lot of devs and applications. Personally I would not describe the majority of code I work on as particularly hard, the challenge was always in getting wh
59.
by rcxdude 26d ago
The basic idea is that a lock should be something the OS is aware of, so that while a thread is blocked on a lock, the scheduler never tries to wake it at all, and when it is unlocked, the scheduler can wake up the thread that's waitin
60.
by rcxdude 26d ago
There's two sides to efficiency: one of them is how much you put in and the other is how much you get out. I was asking the question because the answer of the relative efficiency of a human vs an LLM strongly depends on what you put on

More ›