Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
PaulDavisThe1st
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
6 ms
·
91.
▲
by
PaulDavisThe1st
3mo ago
"So you're saying there's a chance? It might actually be possible to have 100 bathrooms! I knew it!"
92.
▲
by
PaulDavisThe1st
3mo ago
No, it mentioned 816 out of more than 4500 respondents had net worth over $1M. That's quite different to your claim.
93.
▲
by
PaulDavisThe1st
3mo ago
17% of generally retired people are relying on the retirement checks of their parents ? Really ?
94.
▲
by
PaulDavisThe1st
3mo ago
> $350/month [ 2000 ] > $1500/month [ 2026 ] CPI inflation would suggest that the current cost ought to be about $692. So if it really costs $1500 to cover the things you've mentioned (noticeabley absent: student loans,
95.
▲
by
PaulDavisThe1st
3mo ago
It might, however, be a nice thing to aspire to, no? Economic self-sufficiency doesn't have to imply atomization, I think.
96.
▲
by
PaulDavisThe1st
3mo ago
As noted elsewhere in these comments, it was that long ago before the biggest thing propping up capital investment in the USA were the assets backing union workers' retirement funds. That very much wasn't "people looking for
97.
▲
by
PaulDavisThe1st
3mo ago
> The Web could have maxed out as a distribution system [...] thing is, a few of us (mea culpa, sadly) figured out how to use it, even in 94, to sell stuff. the technology even back then was adequate for this.
98.
▲
by
PaulDavisThe1st
3mo ago
Which (a) the EU can and does regulate (b) the majority of the commission goes to the issuing bank. I hate feeling as if I'm defending Visa or Mastercard, but if we're going to attack them, let's get our storytelling correct
99.
▲
by
PaulDavisThe1st
3mo ago
> People are worried about authoritarianism, yet meekly cede such powers to a corporation? It boggles the mind. The libertarian answer is that corporations cannot threaten your freedom because you always have a choice of whether to deal
100.
▲
by
PaulDavisThe1st
3mo ago
Let me know where I can get PayPal's micropayment rates for the 80%+ of transactions (below US$12) that support ardour.org ... We save almost US$0.25 per $1 transaction because of PayPal.
101.
▲
by
PaulDavisThe1st
3mo ago
It's not the ability to use a maximalist iterator that is of interest here, it's the ability to use or not use any kind of iterator at all.
102.
▲
by
PaulDavisThe1st
3mo ago
But it's not a Python thing. Rust is noted below, and there's also C++. std::container<T> container; for (std::container<T>::iterator i = container.begin(); i != container.end(); ++i) { ...} auto iter = cont
103.
▲
by
PaulDavisThe1st
4mo ago
> They are omitted from searches now but are among the best internet search content I use google search with &udm=14 to remove all AI elements in the results presentation. Searching for "dolosse", the HN item is the 3rd lin
104.
▲
by
PaulDavisThe1st
4mo ago
Atwood is on record as saying that everything that happens in The Handmaid's Tale has happened somewhere in the world within the 50-100 years that preceded her writing it. While Iran may have been an inspiration, it was not the ins
105.
▲
Albuquerque Modernism: Manera Nueva and Steve Baer
(albuquerquemodernism.unm.edu)
1 points
by
PaulDavisThe1st
4mo ago
|
0 comments
106.
▲
by
PaulDavisThe1st
4mo ago
Not just allocation free. It needs to be page-in free too. And critical loops need to be cacheline friendly.
107.
▲
by
PaulDavisThe1st
4mo ago
I don't think it is useful to say that a child "understands how to catch a ball", even though it is something many of us do say quite often. The child knows how to catch the ball, without understanding. Later, the child lea
108.
▲
by
PaulDavisThe1st
4mo ago
I prefer to work the other way around. That is, accept that a lot of human speech (and text) is generated via similar mechanisms to the ones that drive LLMs, but note that there is another kind of behavior - reasoning - which seems to be di
109.
▲
by
PaulDavisThe1st
4mo ago
> Frontier science looks different today. It's rooted in model weights and GPUs. It is flooded with token spend and agentic loops. It blooms in data centers. This seems like handwaving to me. Even the actual frontier science that is
110.
▲
by
PaulDavisThe1st
4mo ago
> But to me it seems obvious that LLMs are not repeating patterns without comprehension and do understand what they are saying; otherwise they would not be capable of doing things they routinely do. So this seems obvious to you, and yet
111.
▲
by
PaulDavisThe1st
4mo ago
There seems to be some confusion between "we can" and "we should" in your comment. Bender (and others) are not discussing the capabilities, but rather (a) the fundamental mechanism(s) (b) the advisability and desirabilit
112.
▲
by
PaulDavisThe1st
4mo ago
You mean since it was invented by Indians, Germans and the French?
113.
▲
by
PaulDavisThe1st
4mo ago
> “Well.” He takes a deep patient breath and looks around. “Y’know, uh, you’re an intelligent guy, and you’re hardworking, and yet, y’know, you didn’t have enough time to get through A Table for Fortune.”
114.
▲
by
PaulDavisThe1st
4mo ago
One of the central points of that video by Monty is that staircase drawings are a lie . It's not that they need to be explained, they need to be avoided.
115.
▲
by
PaulDavisThe1st
4mo ago
No need to add any features to Ardour - it already supports any sample rate that your hardware supports. There are places where (a) double precision (or better) floating point math benefits DSP, but that's nowhere in Ardour (and likely
116.
▲
by
PaulDavisThe1st
4mo ago
Woo!
117.
▲
by
PaulDavisThe1st
4mo ago
For lots and lots of Americans, endless consumerism, driving and junk food are not a part of the dream in any way .
118.
▲
by
PaulDavisThe1st
4mo ago
Note: blind testing is not double blind testing. Scientists evolved double blind for a reason: blind testing doesn't remove bias. Yes, the discussion was "never about analog vs AD". But my point is that I see little point w
119.
▲
by
PaulDavisThe1st
4mo ago
> So the tradeoff is that you lose a small amount of noise floor constantly-- out at the 20th bit So, basically, no better than the best AD converters we already have? My understanding of the fundamental limit to AD performance is that t
120.
▲
by
PaulDavisThe1st
4mo ago
The first diagram in that article is pretty ironic in an HN comment thread about Monty @ Xiph's stuff. Have you never seen his takedown of the "stairstep" drawing?
More ›