derefr

65 ms·

601.
by derefr 2y ago
There’s even a make-up trend of “enlarging” the eyes by painting the waterline of the lower eyelid white, that could be used as a justification for walking around like this even in a totalitarian police state.
602.
by derefr 2y ago
If what you’re trying to do is to publish prepared images of yourself , that won’t be facially recognized as you, then the answer is “not very much at all actually” — see https://sandlab.cs.uchicago.edu/fawkes/ . Adver
603.
by derefr 2y ago
What about the hypothetical regulatory reality where companies can cite “20 years of clinical practice / OTC marketed sales in another country X” as justification enough for considering a medicine to be GRAS until proven otherwise (sim
604.
by derefr 2y ago
Unlike those other things, which are external contaminants that find their way in due to non-sanitary processing conditions, cadmium and lead in chocolate come from the soil the cocoa trees grow in, and end up in the beans themselves, in a
605.
by derefr 2y ago
I presume the counterargument is that inference hosting is commoditized (sort of like how stateless CPU-based containerized workload hosts are commoditized); there’s no margin in that business, because it is parallelizable, and arbitraril
606.
by derefr 2y ago
They wouldn't be "forced to continue living like that." They'd be free to seek opportunity anywhere in the world. (Unlike today, where North Koreans are literally the only people in the world who are not allowed to leave
607.
by derefr 2y ago
Isn’t the answer, then, to just give North Korea 10x as long as a “managed” colony before unification? --- Or, alternately, given that literally everything in NK would need to be rebuilt at some point... rather than "swallowing the ele
608.
by derefr 2y ago
> Should South Korea manage to unify North & South it would bankrupt them. What if they didn't do that (at first), but instead (as an intermediate step) annexed the north as a colony with its own economy but South Korean governo
609.
by derefr 2y ago
You're right! Edited. (I think I never wrote Erlang for long enough before switching to Elixir to find this syntax. Also, I don't think it's ever come up in any Erlang code I've read. It's kind of obscure!)
610.
by derefr 2y ago
They're not saying Skia itself is browserless; they're saying that "Skia Canvas" is a non-browser-based implementation of what people would normally call the browser Canvas API.
611.
by derefr 2y ago
For quite a while after Windows already became popular, any program that did networking on IBM PCs still had to be an MS-DOS program. In 1993, (regular, consumer) Windows didn't really "support" networking; (limited) OS sup
612.
by derefr 2y ago
> fn/n where n is parity of fn was confusing for me at first, but then I started to like the idea, esp. in dynamic languages it's helpful to have at least some sense of how things work by just looking at them. To be clear, this
613.
by derefr 2y ago
> It's generally my view that from a bird's eye view LSMs are simpler than on-disk Btrees, largely because they are more composable and a lot more of their data is immutable. I dunno about this. On-disk B-trees can actually b
614.
by derefr 2y ago
To me, it’s always been strange that we speak of “node editors” as the concept here, when the defining abstraction is the data structure of a node graph representing a dataflow. Specifically: • with the graph having an associated schema tha
615.
by derefr 2y ago
> I acknowledge and appreciate Anthropic's addition to the corpus of scraped data, but that data (both input and output) is still ultimately from others; if it did not exist, there would be no product. There’s a Ship of Theseus thin
616.
by derefr 2y ago
> My banks as for a password in that iframe. Not done type of one-time-code. Maybe there's an option for this that a bank can choose when "enabling" 3D Secure? Seems stupid, though. Never experienced it personally. > If
617.
by derefr 2y ago
No, (Visa) 3D Secure is challenge-response. It pops the iframe on behalf of your bank, sure, but the iframe itself is operated on a white-labelled basis by Visa (with your bank's insignia put there because Visa knows the bank logos
618.
by derefr 2y ago
Forgot to mention, OVH makes the current topology and traffic saturation of their network switches public: http://weathermap.ovh.net/ (I think they mostly do this so that customers can see and verify that any DC-level pee
619.
by derefr 2y ago
> is a meter attached Probably not by default, but if your usage starts to saturate their network switches they’ll add one, to figure out who’s disrupting everyone else’s QoS.
620.
by derefr 2y ago
That’s the opposite of what the GP poster meant to imply. They meant that you don’t hear about the ones that use Signal because they don’t get caught.
621.
by derefr 2y ago
> core AWS infrastructure is still heavily dependent on Perl scripts mashed together 15+ years ago What part of the infrastructure? The control-plane logic that triggers when the dashboard/CLI/CloudFormation request modificatio
622.
by derefr 2y ago
> Your code will get refactored or rewritten. Docs will be superseded, then lost in some CMS migration. Before long, it will be as if you have never worked there. The exception is if you build a fundamental component of the system, and
623.
by derefr 2y ago
I don't disagree that the variant hypothesis is wrong ; I just disagree that it was one covered by their writing. The author was talking about "calling out to a chess engine" — and then explained why that'd be fragile a
624.
by derefr 2y ago
> Many, many people suggested that there must be some special case in gpt-3.5-turbo-instruct that recognizes chess notation and calls out to an external chess engine. Not that I think there's anything inherently unreasonable about a
625.
by derefr 2y ago
> Anthropic means "human," and if a company must remind me it is made of/by/for humans Why do you think that that's their intended reading? I had assumed the name was implying "we're going to be an AGI
626.
by derefr 2y ago
They didn't really have a chameleon. See "Experimental setup" in the linked paper [emphasis mine]: > After fabrication of the meta-optic, we account for fabrication error by performing a PSF calibration step. This is accom
627.
by derefr 2y ago
Okay? I already said "I accept the answering-questions part." Doctors not having to manually probe any signs of heart/lung problems each time they're in the room, would mean that they get either get more time to test
628.
by derefr 2y ago
Well, that's what this thread is fundamentally about / the point that the GGP poster was making. Coding challenges should measure your ability to quickly and efficiently dive into a big unknown codebase to fix something — beca
629.
by derefr 2y ago
Real. Engineers who don't think they have this problem, are engineers who see their dependencies and the lower layers of their stack as ossified black boxes they "can't" touch, rather than something they can reach into a
630.
by derefr 2y ago
> Some people get it in their heads that their commit history needs to be "clean". It doesn't... If by "clean" you mean "readable" (in the literate-programming sense), then I'd be a strong advocate

More ›