You could also use this to make it look like you're making eye contact when you're, you know, actually looking at the other person you're on a video call's eyes in their image. Normally there's the obvious problem of "look at camera to appear engaged, or look at person as normal if really engaged".

I do believe that the original paper can be summed up as "Our static time of use pricing needs to be updated because the grid behavior has changed", with elaboration on exactly how and why and a focus on the needs of a particular load type. Having a static pricing curve (that you update relatively rarely) reduces the costs to adapt to the grid load, as you can just schedule and use timers rather than connected smart devices. Dynamic pricing is only applicable to a smaller subset of users because it requires more communication, and even then for a large shift like swapping nighttime loads to daytime it is useful to communicate "yes, it's likely to be roughly this price-to-day curve for the foreseeable future", so large loads know not to just ride the price hump out.

Well, in the case of the covid vaccine it doesn't include any of the machinery this includes to do a DNA edit and just has the code for a spike protein, and in this case, it's got a whole payload of machinery attempting to do a permanent modification to a patch of the patient's tissue in their liver, so yes, it's rather different situations. What we don't know is stuff like whether the modified liver cells will be dividing and persisting or whether they'll be replaced by fresh, unmodified cells off a stem line, that kind of thing.

It's only "similar" in that they both use RNA and a delivery mechanism, it differs in every other detail, and the "with RNA" has barely more information than tacking on "with a computer".

There's no germ line editing here, so generations aren't particularly any more involved than with any other source of mutation in somatic cells.

You aren't going to pass on changes from editing some of the cells in your liver, and the likelihood of this reaching gametes in a functional form is extremely low. It'll actually be harder to apply this technique anywhere _except_ your liver, given that your liver is where this kind of thing goes to get removed, but in this case that's where the target is already.

it's pretty easy to fix: rotate the telescope around it's central axis, rotate the image back, and compare. The diffraction patterns move with the mirror, the real stuff out there doesn't. So you do the math and cancel it out.

It'd take something out there rotating at exactly the rate we happen to pick for the telescope to have a mistake from it.

OTOH, it takes additional observations, so we'll be seeing plenty of less-processed images with the artifacts too.

Long term it's not a big deal to have fewer people. On the scale of a human lifespan or two, it's a pretty big deal, because it shifts the population pretty heavily to older, so you've got much fewer people to actually work relative to how many people need work done. That might cause quite a lot of suffering and potentially rapid collapse, although we might get lucky and outpace people leaving the workforce with automation.

That's pretty much what's looming: starving in your 70s because there's nobody able-bodied enough left to farm. Depends on the exact population curves, of course.

Generally there's some mechanism whereby you can "delegate" your tokens to allow someone else to validate, with the expectation that they pay you some of the profit from validating as interest, so it's less about having a lot of money and more about having a lot of trust represented by people choosing to delegate to you.

I don't think it'll particularly stabilize or destabilize the currency much, apart from breaking (thankfully) the connection to "how much power and how many GPUs can you buy with that". That was already not an even exchange, else it wouldn't have made any financial sense to go into mining.

In terms of attacks, you can generally detect whether a validator is misbehaving, and a misbehaving validator gets actively penalized; I don't know whether ethereum is going to use slashing, but it's not uncommon to allow another node to _confiscate_ the deposit a validator made if they can prove misbehavior - for instance, if the validator signed two different versions of the same block, the way they would if attempting to double spend. In general you also need more than one validator to make an attack feasible, so it's not so much "buy a lot", it's "buy more than half of all ethereum in existence", and then you can keep an attack going for a little while, before other nodes on the network confiscate all of that ether and you've lost more money than Jeff Bezos' net worth.

Power draw reduction is _also_ cooling requirement reduction, too. Ultra-cold is almost free if nothing is producing heat, you're just dealing with leakage through your insulators. The cost of keeping it there is probably pretty small with 80x less energy to move.

They mimicked, to a very high degree, a very real device the which is very much designed to cause a seizure in a diagnostic setting (to verify the diagnosis and the like), and the mimic proves to have the same functionality as the device. Blinking lights that happen to be similar would be poor form, but dropping functional and dangerous to some people medical equipment into your game as a prop is not quite the same.

Pretty sure that's 60% of original; it's hard to tell judge what it ought to be at given unknown damage potentially internal to the cable. Not keeping up with the coating maintenance (due to funds shortage) would not be a surprising root cause for the loss of the telescope.

Essentially, it's not that kind of cable; they were static heavy steel cables mounted in a socket at each end. There was no provision for lowering it through the dish (and it'd have just torn through the dish and smushed the instruments under the platform). Certainly would be nice to see any replacement have that kind of "descend to safe position" capability, though.

With the accumulated damage they observed on the cables, it probably wasn't an original problem with the steel. Corrosion and wear happen and can reduce strength considerably. And there was probably a minor earthquake involved in this one.

it's only a telescope, though. Aricebo was our most powerful planetary radar, and FAST doesn't have the weight capacity for the transmitters to make it into one.

Gold is, in fact, a worse conductor than copper, but the non-reactive surface will readily cold-weld to other gold surfaces in atmosphere (most metals will do this in a vacuum). This combined with it being not bad as a conductor (third, I think, behind silver and then copper) makes it quite good for connector contacts, as the welded spots conduct much better than the surface of most metals.

I've had nothing but a good, solid experience with dyn.com, but they're certainly not the cheapest - a name is $15, DNS service to use it is $35. Dunno if there's a charge to delegate the name to a different DNS provider. They're still good at having subdomains pointing to not-so-stable IP addresses, though, and you can but are not required to set up your entries as you wish up to just a count limit (75 records per domain). Haven't tried their email forwarding from the domain stuff.