Even the summary said the match was only 93% probable. That's a LONG way from any reasonable certainty. (Out of 1000 people you'd have 70 matches. If you don't do other steps of elimination, what is the size of the population you're screening?)

Perhaps not, but they are designed to *attempt* to only target the things they are aimed at.

I don't think that's a well defined term. Historically you can only figure out which actions were war crimes after one side has won, and decided to set up trials. The rest of the time different sides pick different actions.

No. They say "sinister".

Batteries are not a sufficient answer here. But when you add in a grid connection they work. The weather will be different in different places, so some places it will be windy, some places it will be rainy, and other places the sky will be clear. I think this works even in Britain, and that's North of Ontario.

Well, and FWIW, there has been a huge increase in the number of
"security fixes" Debian has been downloading recently. I assume the same is true of other Linux distros and probably for Apple, though I don't think those are made public. Perhaps the BSDs haven't seen a large uptick.

Once I see it doing what I want, I'll be interested. But in order to do that, it needs to be able to understand the action of the fiction I'm asking it to read. E.g. if it's reading "Alice in Wonderland" it needs to use a proper voice for Alice talking to herself as she falls down the rabbit hole. (And for the editorial comments interspersed.)

O,. and it needs to be able to do this without an internet connection.

While your points are good, this looks like negligence to me. So, yeah, our "justice" system needs reform, but that facility should be made to change it's practices. (Not that I'm sure this decision will cause them to do that.)

The problem is that TMI was NOT a really bad disaster. It was frightening, and could have been bad. Actually, though, it was quite contained. But the more reactors you have, the greater the chance of a really bad disaster. If one of these is small enough to be thrown around by a tornado, you have new possibilities of a really bad disaster. (I didn't check. I assume it's too heavy. But perhaps it could be broken open.)

OTOH, smaller reactors have a smaller "worst possible case". I'm not really convinced that they're a good idea, but I could be. (If it's intended for use in space, what happens if there's a catastrophic launch explosion? That might be a good use, if they can handle the heat, but I'd prefer that the fuel be mined on the moon.) Low probability accidents that are possible WILL eventually happen. (Or at least one should plan that way.)

It's a new model, so the only way to really know whether it's safe or not is to try. But modelling gives VERY useful indications.

WHY is it wonderful news? The reports I've seen say that small reactors are less efficient than larger reactors and have increased chance of dangerous results (though the size of the danger is smaller).

Just asserting that it *is* wonderful gives me no reason to believe you.

I "sort of" agree. But my estimate of this is low enough (cost-including-pollution vs benefits) that I don't think it should be subsidized including being granted favorable legal treatment. In fact I'd be in favor of eliminating bankruptcy protections for any damage it causes.

I didn't mean to claim that I though trade secrets in code were a good thing, merely that they were a reason a company might not want to open their source code. Patents are not such a reason. (Also, I don't believe software patents should be valid. Copyright, yes. Trademarks, yes. Trade secrets, yes. But software should not be patentable. [And copyrights should have a much more limited duration.])

You left out "trade secrets". That would be plausible if the software were sufficiently innovative.

That's "tool using", not "tool making", but I wouldn't be surprised if dogs could to simple tool making. A cow was recently documented making a backscratcher.