ur-whale

31 ms·

481.
by ur-whale 2y ago
> but they don't fit in. You seem to say this like it's a bad thing?
482.
German flying taxi firm Lilium Files for Insolvency(sec.gov)
13 points by ur-whale 2y ago|1 comments
483.
by ur-whale 2y ago
From the 6k filed today: > The management of the Subsidiaries has informed the Company that they have to file for insolvency under German law and in doing so will apply for self-administration proceedings in Germany.
484.
by ur-whale 2y ago
You might be staying there a long time because you're fuming, so: nope.
485.
by ur-whale 2y ago
> It would require somehow recording an infinite amount of information You're assuming spacetime behaves like the set of reals (something with cardinal ℵ1, if you accept the continuity hypothesis), an object that even if you stay co
486.
by ur-whale 2y ago
The (what very much feels like an) assertion that "If a collision concentrates enough energy in a small enough region, the particles form a black hole" seems very much rabbit out of a hat. That supposes in particular that general
487.
by ur-whale 2y ago
I wish there was a standardized way to let folks who run a website such as this one know how much a casual passer-by viewer enjoys the byproduct of their work.
488.
by ur-whale 2y ago
Yeah, good point. That also makes some of my remarks kind of irrelevant.
489.
by ur-whale 2y ago
Reading the first few paragraphs, I note - heavy dependence on large (Mono) and/or unusual tools (nmake ... in 2024 ... really?) - worse than that, dependence on specific (older) versions of exotic, non-standard things
490.
by ur-whale 2y ago
Doesn't Google already do that with the site:blah.com keyword ?
491.
by ur-whale 2y ago
There's also the other approach: actually reading the content and judging it on its own merit. https://en.wikipedia.org/wiki/Argument_from_authority
492.
by ur-whale 2y ago
Unless I'm missing something, title of the article doesn't really correlate with its conclusion.
493.
by ur-whale 2y ago
> I've started to get back into reading books and long form blogs How do you know the books and blogs you're reading weren't generated by an AI?
494.
by ur-whale 2y ago
Some sort of 3D visualization of this dataset would be very, very nice. This is of course hard ... the distance (3rd number) is not one that is very precise. But still, I always have a hard time picturing what our galaxy looks like when loo
495.
by ur-whale 2y ago
Seems to me the article is claiming a lot of things, but is very light on actual comparisons that matter to you and me, namely: how does one of those fabled AI-designed chop compare to their competition ? For example, how much better are th
496.
by ur-whale 2y ago
Terry Pratchett came to Switzerland to help film a documentary on the topic: https://en.wikipedia.org/wiki/Terry_Pratchett:_Choosing_to_D...
497.
China's Alibaba launches over 100 new open-source AI models(cnbc.com)
2 points by ur-whale 2y ago|0 comments
498.
by ur-whale 2y ago
> Why just actors? So far, they're the only one who signed a big enough check to the politicians.
499.
by ur-whale 2y ago
What I find truly amazing here: an IEEE article where you can just click on the PDF button and actually read the article without hitting you head into a paywall.
500.
by ur-whale 2y ago
> Unfortunately, it means I’d have me. Yeah gearheads have a saying for that: first upgrade the driver then upgrade the car.
501.
by ur-whale 2y ago
SD cards are indeed a really bad deal when it comes to reliability, especially if like me you tend to slam raspis everywhere almost as a reflex: before you know it, you end up with a large-ish fleet of the things in your house. But: Raspis
502.
by ur-whale 2y ago
True. Befunge has real lofty ideals, not just a miser goal of being hard to parse by humans: > Befunge, with the goal of being as difficult to compile as possible
503.
by ur-whale 2y ago
Their stack doesn't seem to include an LLVM BrainFuck backend. This shortcoming is clearly a non-starter for any serious enterprise BF implementation.
504.
by ur-whale 2y ago
> Not useful! https://en.wikipedia.org/wiki/Fortuna_(PRNG) [EDIT]: Also: https://words.filippo.io/dispatches/linux-csprng/
505.
by ur-whale 2y ago
Or you could pump the heavily biased bit stream into a modern cipher and get a - for all intents and purposes - super high quality, completely unbiased random stream with the same bandwidth characteristics as the input. And, as a matter of
506.
by ur-whale 2y ago
Does this work with a guaranteed random bit output bandwidth ? These techniques (eg von neumann) all seem to suffer from that same problem, namely, they crank out uniformly distributed random bits from biased sources, but with no guarantee
507.
by ur-whale 2y ago
> Since the syntax is quite distant from C-like syntax THIS - and none of the other things cited in the article - is the main reason why Haskell is getting no traction in the real world. The syntax is an abomination, and worse, IMO p
508.
by ur-whale 2y ago
Europe is never going to be able to catch up in AI, much less innovate if it keeps cranking out useless and damaging regulation such as this. As Elon pointed out, when regulators crank out an endless stream of laws, they never ever think ab
509.
by ur-whale 2y ago
What would you do differently if you owned a dealership ?
510.
by ur-whale 2y ago
This article is pretty far out, but definitely worth a read. Cryptography is one of these subjects that initially seems interesting in terms of how it works but rather boring in terms of what it does. But then the more you learn about it, t

More ›