They missed their window of opportunity. A few months ago they could have just bribed Trump into giving them blanket immunity forever. Now that Trump is failing in so many ways that even Republicans are starting to balk at bowing to his every whim, Meta is actually going to have to play the game the old fashioned way and figure out how to bribe all of congress to get their immunity. Poor babies. They should have planned ahead. They could have saved themselves a lot of time and money.

In the middle ages Kings were publicly admitting they were marrying children. Though in "civilized" societies they supposedly waited to start the actual intercourse until they girl had her first period, which could range from twelve to fifteen or so. In not so civilized? It was whenever the husband/leader lost patience or got randy.

You are pushing direct lies. France critically depends on the European grid and has in the recent past needed to buy enough electricity to drive up prices all over Europe.

Yes. Trump has no understanding of how reality works. He only can (somewhat, barely staying out of prison) navigate being rich and powerful. He is not even an accomplished liar because he believes the nonsense he claims. But while Trump tries to only give positions to people even dumber and more disconnected than he is, he is failing at that as well and some people in his administration have more insight that he has. They just have no honor or integrity.

I do hope a significant fraction of his supporters (and hence enablers) are recoverable. Because if they are not, they represent a severe long-term problem. And while I think the only path the US has left is into the 2nd world, this will go better and recovery will be more likely and faster if many of the cultists revert to being ordinary people instead of fanatics.

I want Starship human rated so we can load one up with Elon, Zuck, Bezos, maybe Branson and a smattering of others, launch them out past the moon somewhere, and just leave them with surveillance cameras on. Let them experience 24/7 surveillance as their supplies slowly dwindle and they are eventually forced to resort to eating each other to survive. Last one to be eaten wins! What do they win? The chance to starve to death instead of being butchered for Haitian Steaks!

They companies want it because they can use it to consume user data. Usage numbers are bullshit since it's forced into everything people use whether they want it or not. I guarantee you if there were an opt-in instead of an opt-out or an opt-fuck-you-take-it-bitch those usage numbers wouldn't be anywhere near what they are today.

Why don't the billionaires just shake the couch cushions on all their yachts, hand what falls out to the state, and stop whining about it?

Bernie's ideas, at least the way he articulates them, come across as well intentioned. This is 100% the reason none of them ever come to fruition on the national stage. You can not convince the owner class to penalize the owner class, and that's what it would take to make any of Bernie's ideas stick.

As much as I don't want to see him give up the fight he tries to put up, what a waste of a life. Always cutting against the grain and getting bucked back into place for it.

It is not a surprise they overlooked this aspect. They themselves never use data. They already know how things are and if some facts disagree, it is obviously people falsifying data or lying. Hence these complete cretins do not actually understand what the value of real data is. And hence they could not even imagine there are other uses to this real data, because they are convinced all that data is fake anyways.

Incidentally, this little problem is also why they cannot do anything right. Doing things right requires seeing reality. They are not capable of doing that and neither are most of their supporters.

A technology with still zero deployed regular installations, no real prototypes (China has one pebble-bed one which I gather is not what was ordered here and Russia has some sub-reactors on a barge, which you cannot even get approved in the west, both with no public cost figures) not even any real assurance it will even work and they managed to sell three to one customer? That is a great success for marketing and a black day for engineering that now has to deliver on promises that are essentially hot air.

The pilot stepping down could be viewed as practical. On a landable craft, you'd want someone to have good visibility of the ground on away missions in unfamiliar territory, and the Eagle was large enough that the step-down to reach that visibility is a believable component of the craft. It's an altogether well thought out design that I wish more sci-fi shows had mimicked over the decades since.

Incompetence at its finest!

Could, may, etc. does not make for useful predictions. And while resilience to hardware failure sounds nice, the things degraded by ageing are the higher-power elements, like com line drivers. These give you full-chip loss and they are relatively homogenous in effect, i.e. they will take out almost all your hardware in a relatively short time.

On the minus-side, LLM results look less and less valuable every day as actual research (not just gut-feelings and hype) starts to roll in.

In theory, your professor was right. In practice, not so much. The thing is theory requires perfect proofs. Practice only "good enough" ones. (We will ignore that the one-time-pad is mathematically proven secure, because it has little practical relevance...)

So that state of things is that ElGamal has a security proof relying on an unproven assumption that is very likely true. RSA is much weaker on the theory side and current block ciphers or crypto-hashes are even weaker on the proof side.

As to QC, the problem is that effort seems to scale exponentially with qbit numbers and (!) computation length. That means there are not very large sizes of computations they will never scale to. And, for example, doing RSA 40960 (i.e. 10x larger than the larges typically used today) is not that much of a problem. The second problem is that there is no quantum computation in existence that actually conclusively proofs it even works. The theory may well turn out to be a tiny bit inexact and then everything fails. And "quantum insecure" does not matter one bit if the machine for it cannot be built.

As to QC algorithms, block ciphers are safe, hash functions are safe, some other things may be as well. The reason is the compute mechanism cannot really break them (half bit length for block-ciphers, for example, still completely infeasible to attack if that is 100 bits remaining or so).

Unfortunately, the CRA only goes active end of 2027.