Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
trevyn
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
86 ms
·
841.
▲
by
trevyn
8y ago
They’ve both checked out long ago, particularly Sergey.
842.
▲
by
trevyn
8y ago
You may be interested in György Buzsáki’s “Rhythms of the Brain”: https://www.amazon.com/Rhythms-Brain-Gyorgy-Buzsaki/dp/01998... “...self-emerged oscillatory timing is the brain's fundamental organizer of ne
843.
▲
by
trevyn
8y ago
Ah, thanks! PDF link: http://nano.caltech.edu/publications/papers/Matheny-2019-Sci... Looks like they did simulate it, and the experimental results differed from the simulation due to number of harmonic terms used
844.
▲
by
trevyn
8y ago
Note that absorption and metabolization rates can affect the safety profile; for example, this is why acetaminophen is generally safe at lower doses, but quickly becomes hepatotoxic at higher doses, because different metabolic pathways come
845.
▲
by
trevyn
8y ago
Sci-hub doesn’t seem to have it, but Google shows the full paper mentions something about simulation. It looks like it’s only 8 oscillators and they actually built a physical implementation — why couldn’t this all have been done in simulati
846.
▲
by
trevyn
8y ago
If the camera is streaming or recording video, then it is likely to be significantly warmer than ambient; a thermal camera (they’re not super-expensive these days) should easily reveal hotspots.
847.
▲
by
trevyn
8y ago
> It's exactly the same as prescription pharmaceutical pills Note that meth amphetamine is rarely if ever prescribed — the (non-meth)amphetamines commonly prescribed for ADHD have different safety profiles as well. (And, of course
848.
▲
by
trevyn
8y ago
> But without copyrights, it is unclear how one could make a living as an author, a publisher, a musician, or a filmmaker. IMO, the obsolescence of the perceived “need to make a living” is the Inevitable here.
849.
▲
by
trevyn
8y ago
I doubt they ignore DMCA requests.
850.
▲
by
trevyn
8y ago
I dunno, Sony has never really been very good at software. Look at the PS4 OS — it actually repeatedly warns you not to unplug the machine without “powering down” from the software first. Imagine if your phone didn’t have a reliable filesys
851.
▲
by
trevyn
8y ago
What do you believe now?
852.
▲
by
trevyn
8y ago
> when confronted with someone who clearly isn't skilled but wins anyway That’s what keeps the fish around!
853.
▲
by
trevyn
8y ago
> will probably die of hunger It looks like you’re in the US. You’d have to try quite hard to die of malnutrition here.
854.
▲
by
trevyn
8y ago
IMO, we already have the mental-disease equivalent of germ theory, it’s just far more complex and not yet well-understood by enough people to have broad social impact. “The germ theory was proposed by Girolamo Fracastoro in 1546, and expa
855.
▲
by
trevyn
8y ago
“...the western world has been puzzled by how the vast Pacific could have been navigated relying only on the stars.” Strap your undesirables to rafts and push ‘em out to sea? I don’t see why there has to be intentional “navigation” to end u
856.
▲
by
trevyn
8y ago
Reuters as well.
857.
▲
by
trevyn
8y ago
> I often times see people confuse East Asian aesthetics with looking 'amateur' to American eyes. I’m guessing that both are driven strongly by economics, so it makes sense that there is some perceptual overlap.
858.
▲
by
trevyn
8y ago
Is it possible to empirically measure such credibility?
859.
▲
Jailbreaking the Simulation with George Hotz [video]
(youtube.com)
2 points
by
trevyn
8y ago
|
0 comments
860.
▲
by
trevyn
8y ago
And you have to check a box that says "I agree that my model will only be used for applications that follow Google's AI Principles." Wow.
861.
▲
by
trevyn
8y ago
https://en.wikipedia.org/wiki/Message_passing#Message_passin...
862.
▲
by
trevyn
8y ago
Check the annual out-of-pocket maximum on your plan. It’s been capped at about $9k since Obamacare.
863.
▲
by
trevyn
8y ago
> I don't think I'm a high profile enough target that somebody would bother sifting through everything I say in front of my laptop for something interesting. The threat model is closer to automated transcription, global histor
864.
▲
by
trevyn
8y ago
That’s a really interesting hypothesis. Breaking off the nose is probably easier to do and maybe was a sufficient “this is not a real person” mental input to damp the response.
865.
▲
by
trevyn
8y ago
It works fine for cars too as long as everyone has equal disregard! See: Southeast Asia.
866.
▲
by
trevyn
8y ago
And the 8-minute above-ground version: https://m.youtube.com/watch?v=WJYOMFayruw
867.
▲
by
trevyn
8y ago
Well, how cheap we talking? :-) https://www.reuters.com/article/us-india-liquor/deaths-mount...
868.
▲
by
trevyn
8y ago
Wait, this is easy — have candidates interview each other , and extend offers to every pair that would hire each other.
869.
▲
by
trevyn
8y ago
The guidelines for safe aviation have way lower Kolmogorov complexity than the guidelines for safe “everything that software can do”-iation. It’s actually close to a subset when you think about it.
870.
▲
by
trevyn
8y ago
Can we do airborne ground-penetrating radar yet?
More ›