Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
ben_w
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
43 ms
·
751.
▲
by
ben_w
3mo ago
Even as a native English speaker, this is one of my more common mistakes, along with weather/whether[/wether]. How did we end up with the word with the long-o having one o, and the word with the short-o having two?
752.
▲
by
ben_w
3mo ago
> We should find a way to get "re-sensitized" to what we are witnessing and the pace of it. We (collectively, there were obviously many exceptions) didn't internalise exponential growth even with the much faster doubling t
753.
▲
by
ben_w
3mo ago
> Well, apparently they've had (lots of) the raw material for years. "raw material" != "material". While I assume they could've purified enough of the former into the latter by this point if they wanted to,
754.
▲
by
ben_w
3mo ago
Spaceplanes have a nice visual aesthetic, but absent something much more energy-dense than chemical fuel, they'd still need to be multi-stage, e.g. https://en.wikipedia.org/wiki/Northrop_Grumman_Pegasus#/medi.
755.
▲
by
ben_w
3mo ago
> A decisive detail would be the prompt used. Beyond liability, why would this matter? e.g. if OpenAI had actually instructed the model "here's the challenge [attachment challenge.md], do whatever it takes to win, it's fin
756.
▲
by
ben_w
3mo ago
As I understand it, there are only three options: 1) OpenAI and HuggingFace are both telling the truth. IIRC not actually a crime because no intent, it is a technological accident, civil responsibility only, but IANAL so it's good &quo
757.
▲
by
ben_w
3mo ago
Today? Wow. You must have very weirdly expensive batteries and/or electricity. Around here, a complete replacement of non-rechargeable bike light batteries would cost €1.25, and they'd still last more than one ride each. Rechargea
758.
▲
by
ben_w
3mo ago
Tariffs so far are on paper entirely, and in practice still overwhelmingly, paid for by the buyer. So go ahead, make the stuff you buy from us 10x more expensive with you (not us) paying the difference to your government, then be surprised
759.
▲
by
ben_w
3mo ago
Back when I lived in Cambridge, someone worked out this was actually true for one of the car parks. (Not those exact numbers, but the same point).
760.
▲
by
ben_w
3mo ago
Why a year, when they've often been breaking the rules for much longer than that? Why not 20% (30%), given that's the share of world GDP (PPP) due to the EU? Why revenue rather than profit? The answer is: big round number that
761.
▲
by
ben_w
3mo ago
I'm thinking iron pyrite. One of the ways it can oxidise will smell of rotten eggs, and its presence in concrete can destroy the structure.
762.
▲
by
ben_w
3mo ago
I live in Berlin, so no. With my mere B1 German, I'm not entirely sure if the pun would even work in German, despite "Vorhangen": translated puns sometimes work fine, other times just make no sense at all.
763.
▲
by
ben_w
3mo ago
You appear to be arguing with the assumption that "get the operators to comply" is false. These companies do, in fact, change behaviour in response to such actions; they may hire lawyers to find and exploit as many loopholes as th
764.
▲
by
ben_w
3mo ago
For me it depends on the details. The baseline is as you say, negative, but there's a plumber not too far from me whose logo is AI generated, and I can't say I'd hold it against them because I wouldn't be hiring them for
765.
▲
by
ben_w
3mo ago
Ironic that this would be physical spam and the complaint is AI. Sure, it is is an environmental disaster, churned out in industrial quantities, low-effort, an emulsified layer of mediocrity, slop everywhere, fouls the air for miles around;
766.
▲
by
ben_w
3mo ago
Sometimes it is, sometimes it isn't. If it was never about objective quality, Plan 9 from Outer Space and The Eye of Argon would be highly regarded, not the butt of jokes and drinking games. And those are the examples so infamous t
767.
▲
by
ben_w
3mo ago
> On the other hand, there's no reason to believe that, in this hypothesized future, LLMs would stick to the languages that exist today. With the exception of machine code, all programming languages exist to make programming easier
768.
▲
by
ben_w
3mo ago
It's not "obviously logical", it's a pattern which we mimic to avoid mockery. example For, semi-randomise I word order can this like, Yoda worse than, and be understood. > is no problem for a machine that takes contex
769.
▲
by
ben_w
3mo ago
I've found it's had unacceptable scroll performance in long contexts for a while now. In this case, it takes me 12 seconds to see content when first opening the link, and about 18 to re-render content when scrolling.
770.
▲
by
ben_w
3mo ago
If natural language was structured in a logically computable way, we'd have had interesting chatbots by the late 80s, basically as soon as a dictionary fit in local RAM, and for the same reason we got compilers. Da hole raisin y nat-la
771.
▲
by
ben_w
3mo ago
ChatGPT repeatedly asserting that 0/x=0 is division by zero and therefore no value of x satisfies the equation.
772.
▲
by
ben_w
3mo ago
Aye. A decade or so ago I wondered if the reason maths was hard was the names being optimised for writing by hand. Everything's single letters if they can get away with it, so when mathematicians run out of Latin alphabet, they use Gre
773.
▲
by
ben_w
3mo ago
Everything has to compete with everything, but: (1) agriculture is both the biggest user and not done in a way that comes even close to maximising land efficiency; nor is there any reason it should, given there's not much pressure on a
774.
▲
by
ben_w
3mo ago
Both statements are true without contradiction. A specific model develops as it passes through training. AI labs change the architecture between models to allow them to surpass the previous models' best scores. It's also going to
775.
▲
by
ben_w
3mo ago
> History is going to look back at this time and how we let such foolishly inconsistent people make such grand choices for everyone poorly. Assuming we have a future history. We've already got "history slop" with AI rewrit
776.
▲
by
ben_w
3mo ago
> I don't think you can 100% ensure your queries have absolutely no biology and cyber keywords inside. Obviously. Almost everything is a precursor to something dangerous, to the extent that if some model isn't aware of the risk
777.
▲
by
ben_w
3mo ago
"thanks, you too" :P
778.
▲
by
ben_w
3mo ago
That's the neat thing, with LLMs we can fully automate this discussion and have it at thousands of tokens per second and millions of tokens per dollar. :P
779.
▲
by
ben_w
3mo ago
Let they who has never ='d when they meant to == cast the first argument. (There's a reason why we benefit from code review).
780.
▲
by
ben_w
3mo ago
I'm not sure exactly which enterprises you have in mind, but sure: quantities which can be expressed as "a year's worth fits on my desk" should not be described as "staggering", and 10 PB of hard drives will (j
More ›