0-_-0

88 ms·

931.
by 0-_-0 6y ago
I meant light pollution affects the entire image, while a streak affects only part of it, which you can just remove. A 30 arcsec stripe on an (apparently average) 20 arcmin detector is 2.5% of the total area, so it's not a very large a
932.
by 0-_-0 6y ago
How much does the bloom extend in the image fom the streak? Can you measure it in arcminutes or is it optics dependent? Is the bloom present if you cover up the satellite exactly? (Is it caused by the sensor, or would diffraction around a s
933.
by 0-_-0 6y ago
(can't reply below) If you had a localised shutter covering pattern X for 6 seconds, could you compensate for that by covering the inverse pattern of X for 6 seconds to get the same exposure for all pixels? So if you covered 3 satellit
934.
by 0-_-0 6y ago
What if SpaceX funded research into shutters that cover part of the CCD? Is that a possible solution?
935.
by 0-_-0 6y ago
2% is trivial. They lost 2% of information (as someone pointed out this might be closer to 20%). This means the cost to the astronomer was 2% (or 20%). Calling this "ruining" is an exaggeration. Of course it would ruin a prett
936.
by 0-_-0 6y ago
Light pollution is global, while a satellite streak is localised to a part of an image. These are not the same problem.
937.
by 0-_-0 6y ago
You point to saying that 2% is a small part of an image as a sign of arrogance? Isn't it a small part of an image?
938.
by 0-_-0 6y ago
Can you move the shutter in front of the receiver when you know a satellite will transit?
939.
by 0-_-0 6y ago
It seems to be around 1-2%, which I would say is trivial. This can be further reduced if you use multiple exposures because then only 1 contains the streak. Getting satellite data from SpaceX about satellite positions and turning off the re
940.
by 0-_-0 6y ago
Is it a difficult problem to solve though in software? On its face it seems trivial, and something that should be part of an astronomer's job description. The cost of not having internet satellites seems to be many orders of magnitude
941.
by 0-_-0 6y ago
I don't think SpaceX would break Earth satellites. Isn't it trivial to remove a light streak from a picture? Satellites were a problem before SpaceX and I'm sure it's 1 click in some software astronomers use. The surface
942.
by 0-_-0 6y ago
I think it means "train ride between the 2 furthest possible points", so using train only.
943.
by 0-_-0 6y ago
I think you just described genius . Hope you're using it for good.
944.
by 0-_-0 6y ago
This was already done under socialism in the eastern side of Europe after the Second World War. Jobs had to be created that were unnecessary, or a single job was split between 2-3 people. Keyword is "full employment" if you want t
945.
by 0-_-0 6y ago
When UBI comes up lot of people on HN ferociously debate the merits of UBI vs. negative income tax, but they are mathematically equivalent . For example, a UBI of $1000 and 25% tax is the same as a (possibly negative) income tax of 25% abo
946.
by 0-_-0 6y ago
Oh yes, the Cyclopean usernames of titan blocks and sky-flung monoliths, all dripping with green ooze and sinister with latent horror. Definitely not screen-reader friendly.
947.
by 0-_-0 6y ago
As far as image encoders go, do we really need 2 parameters to describe compression parameters, bitrate and resolution ? Why not just have bitrate, where you can decompress an image to an arbitrary resolution because the encoding is not
948.
by 0-_-0 6y ago
You can also wait until the emulators catch up. I recently played Breath of the Wild in 4K 75 FPS on the PC.
949.
by 0-_-0 6y ago
Only if you call GPT-3 AI. True intelligence would write as well as possible.
950.
by 0-_-0 6y ago
That sounds good in principle, but in reality GPT-3 is pretending to be a human pretending to be an AI, rather than write like "itself", since there's no self involved. Rather, it simply comes up with the most probable cont
951.
by 0-_-0 6y ago
They don't want you to know it can be done with a $1 NFC sticker
952.
by 0-_-0 6y ago
I read a paper that did exactly what you describe but of course I can't find it now...
953.
by 0-_-0 6y ago
I long wanted to see a proof-of-work cryptocurrency that does neural network training instead of burning through hashes. Imagine if 0.5% of the planet's energy consumptions (9 GW) was used for training neural networks instead of mining
954.
by 0-_-0 6y ago
Doesn't your brain derive knowledge through statistical inference? How else? I can imagine the AGI of the future (having perfected knowledge gathering through statistical inference) some day saying: "Humans perform tricks for trea
955.
by 0-_-0 6y ago
Adversiarial training means that they specifically search for answers that the network would misunderstand. If this only leads to a 40 percent loss (the network still answers correctly 50% of the time) I still consider that remarkable. Choo
956.
by 0-_-0 6y ago
Seems like a combination of that and what seems like some algorithm performing temporal denoising on the video going haywire from the quick moving picture.
957.
by 0-_-0 6y ago
I think it's a good example of "correlation does not equal causation". The specific problem I remember reading about is that diet correlates with geography, and geography correlates with mortality through some parasites being
958.
by 0-_-0 6y ago
Is it really true that eating meat causes cancer? The argument is vague but this seems to be an important part of it.
959.
by 0-_-0 6y ago
> Can regulation fix this? What if it was illegal to work more than X hours per week? X=20 for example. Actually, let me answer my own question: Everyone would need to have 2 jobs, and every company would employ twice the number of peopl
960.
by 0-_-0 6y ago
>Why are we not living in a universe with all possible theories? We are! At least if the article's assumptions are correct, which I think they are.

More ›