Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
ryandamm
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
9 ms
·
61.
▲
by
ryandamm
2y ago
Yeah I think this might actually be backwards; the Soviet Union had explicit production targets, and (iirc) in like the 70s they actually had more industrial output than the US. Remember the "missile gap," when the US was concerne
62.
▲
by
ryandamm
2y ago
I don't think my point was straw-manning the article. The article directly conflates economic output with manufacturing output, by using manufacturing signals (like power consumption!) as a corollary for some hidden, Real Economy. That
63.
▲
by
ryandamm
2y ago
I think America in its best expression is an amalgamation of many different backgrounds and attitudes. Even differing opinions on the functioning of democracy, though I personally think some opinions denigrating democracy are a bit out of b
64.
▲
by
ryandamm
2y ago
Economics experiments are actually a thing. You can have natural experiments, or you can uncover patterns in data, or you can try controlled experiments (in microeconomics), though that sometimes just feels like applied psychology. I dunno,
65.
▲
by
ryandamm
2y ago
That's what PPP is meant to measure.
66.
▲
by
ryandamm
2y ago
Also, "I become richer just because I think I'm richer" was actually a thing Donald Trump once said in a deposition: "My net worth fluctuates, and it goes up and down with markets and with attitudes and with feelings, ev
67.
▲
by
ryandamm
2y ago
That's a fair counterexample. In that case, you could argue that "true GDP" didn't change at all, merely our measurement of it had error. But pretty soon you ask 'well, what is the real signal?' And it's t
68.
▲
by
ryandamm
2y ago
If you can truly come up with an objective measure of wealth, I believe there might be a (quasi) Nobel Prize in economics for you. The reason economics is hard, in part, is that there isn't some core firmament you can rest on; there is
69.
▲
by
ryandamm
2y ago
I have a lot of respect for economists, I know several. They are among the most rational people you shall ever meet. And they are very aware of how incomplete their tools are. But they do a lot with such tools that is worth knowing.
70.
▲
by
ryandamm
2y ago
Two quick thoughts… one, because every transaction in theory happens because the purchasing party values the good or service more than the money, every transaction does — in theory — create wealth, velocity of money is sort of measuring a p
71.
▲
by
ryandamm
2y ago
Arguing that manufacturing is strategic is not the same thing as manufacturing being “all that matters in an economy.” There’s a logical fallacy in your rebuttal.
72.
▲
by
ryandamm
2y ago
China has major structural weaknesses in its economy that are currently overshadowed by the US cr*pping its own and its allies’ pants. Specifically: there is not much of a home grown consumer economy; they have driven growth too heavily on
73.
▲
by
ryandamm
2y ago
I prefer my economic takes from economists, or people who actually study these things. -there is more to an economy than manufacturing. Using manufacturing metrics to rate an economy is like scoring a soccer game based on how far your playe
74.
▲
by
ryandamm
2y ago
I saw this at a DSO* presentation ("Darpa's Darpa") back in ~2018. Hope they've gotten some early traction, that was some of the least wild crap that was being presented. My company was pitching holographic cameras, and
75.
▲
by
ryandamm
2y ago
FYI, my colleagues in the past have used Metashape, if you're in the market for an OTS solution. There is also Epic's RealityCapture; I have not used either one, but as far as I can tell capture density and quality matter a lot. F
76.
▲
by
ryandamm
2y ago
That’s going to depend a whole lot on their budget, your capabilities, and the final requirements — is this being used for visuals, or metrology? I have suggestions for getting visual fidelity but if you’re looking for an explicit 3D model
77.
▲
by
ryandamm
2y ago
You could try generating per-view depth maps, going to a point cloud and meshing from there. (I suspect splats may reduce your accuracy as an intermediate.) I’m not aware of a fully-baked workflow for that — though it may exist. The first s
78.
▲
by
ryandamm
2y ago
Walgreen’s was closing stores anyway and used the pandemic shoplifting as an excuse… but it was never the actual reason. Crime is at historical lows.
79.
▲
by
ryandamm
2y ago
Except for leverage... but the general point remains that the upside is capped and the downside in theory is not.
80.
▲
by
ryandamm
2y ago
I'm hardly an expert, but when I switched to lead-free solder (for the same reason, had a young kid around), the only change that helped me was soldering at a significantly higher temperature. The lead free stuff still doesn't beh
81.
▲
by
ryandamm
2y ago
Red has two cameras that cost less than $10k.
82.
▲
by
ryandamm
2y ago
I was told by a Microsoft researcher that it will unlock currently unsolvable problems in chemistry, weather modeling, materials science, fusion and plasma physics, drug development… the list went on but it was really long. Most advances he
83.
▲
by
ryandamm
2y ago
I don’t think you get much iron in your food from seasoned cast iron; the polymerized oil is an impermeable barrier. We have a lucky fish and it has what looks like rust all over the surface because the metal is directly exposed to the cook
84.
▲
by
ryandamm
2y ago
That’s not a fair comparison, The New Yorker has always had a different relationship with its writers. A freelancer who writes for The New Yorker is likely a highly respected journalist/author/other luminary. Their staff writers a
85.
▲
by
ryandamm
2y ago
No, that still violates thermodynamics. https://en.m.wikipedia.org/wiki/Maxwell's_demon
86.
▲
by
ryandamm
2y ago
… which is usually mediated by electric field interactions, involving virtual photons. Right?
87.
▲
by
ryandamm
2y ago
I think the barrier to commercialization is the lack of demonstrated economic value to having push button splats. There's no shortage of small teams wiring together open source splats / NeRF / whatever papers; there's a
88.
▲
by
ryandamm
2y ago
This is spot-on. Note that the aperture on the camera was quite large, I want to say on the order of 100mm? They sourced really exotic hardware for that cinema camera. They also had the "Immerge," which was a ~1m diameter, hexagon
89.
▲
by
ryandamm
2y ago
It's more expressive than mesh-and-texture, which is the default: it can represent things with poorly defined edges or transparency. Also, because it's solved via GD, you tend to get better fidelity than traditional photogrammetry
90.
▲
by
ryandamm
2y ago
It renders pretty fast, actually. The knock on splats is that they create huge file sizes for video data.
More ›