Hey, I've got a great idea: let's take one of the worst wireless protocols ever devised and use it for the most ubiquitous communications network of all time!

The expression is "half-baked." Sounds like an AI nurse came up with this complaint, in which case, point very much taken.

It was less than a year ago that KB5026372 was breaking VPN connections; and now they've broken them again?

We're taking pious lectures about excessive attack surface and too many features from the systemd guy?

Bad Data meaning "anything I don't agree with or like"

That works until it is gamed by bad actors.

"Pictures. Or it didn't happen!"

AI has ruined the joke.

Thinks they are Apple.

We use Windows because we're not Fanatics, but because we kind of have to.

You know its bad when your 85 year old Father-In-Law keeps asking about Linux.

I'm a little confused by who is supposed to be caving to the threat here. It's a paid database, so I assume that Thompson-Reuters/Refinitiv aren't thrilled; but it was apparently stolen from one of their customers, not directly from them, so their reputation for security competence isn't really affected; and I suspect that most of the people paying for access to this sort of database need something authoritative that ticks the "I'm really trying to know my customer, really" box when feds or auditors come around; so even a reasonably fresh and reasonably large leak is still of limited value("So, you decided to reduce costs by basing your compliance efforts on data of unknown completeness, potentially subject to unknown modifications, sourced from unknown criminals? Very interesting...") as an alternative to continuing to subscribe.

If anything, it seems like its release would be largely positive: probably lots of interesting leads to be followed up, both with regard to what the creepy data broker types know and the things they know about the people they consider relevant, by people who are in no position to afford access normally(if it's even something you can just purchase if your money is green enough; rather than being offered specifically to potential customers known to be in financial services; not just anyone with a checkbook).

Setting aside the fact that CTRL-Z is "suspend process," this is a poor decision by the judge for a myriad of reasons. Scrivener's errors can be corrected at any time, by common law, so the divorce should be invalidated on those grounds alone. Beyond that, valid contracts require both intent and consent, which was apparently lacking here, so the divorce "agreement" should not be binding.

Heâ(TM)s arguably not wrong that VMwareâ(TM)s offerings outside of their core product are kind of inchoate(though, in fairness, itâ(TM)s not like the âhyperscale cloudâ(TM) guys donâ(TM)t all have a stable of shit thrown at the wall to see what sticks that surrounds the core of services that people actually care about or trust); but that seems like a pretty shabby excuse in this context; where it would have been trivial to just not fuck with what people were using and liked while making the alleged investments in glorious future VMware; then letting the value proposition of that help sell it.

As it is, itâ(TM)s hard to read this as anything other than an awkward(and almost certainly temporary, nobody ever genuinely stops trying to boil the frog once they start); climbdown after recklessly spooking more customers, harder, than intended.

It's not unsolvable by any means, there's just insufficient incentive to share storage and bandwidth on that scale. Even if it started out well, eventually it would probably reduce to a relatively small set of paid hosts, as with NNTP. In which case, just use NNTP.

I'm not sure that intelligence is the bottleneck of technological (or any other) progress that many people seem to believe it is. I think this is the view of people for whom technology is inscrutable, but most progress is predicated on research, where the biggest bottlenecks are time and the adequate application of resources (and convincing people to give you those resources). It's not clear to me how a "super" intelligent AI would immediately change that, unless perhaps people trusted it implicitly, so it was consequently better able to allocate resources than we do at present.

In any case AI makes mistakes, and there's no reason to believe that mistakes diminish as intelligence increases, so trusting AI as above probably wouldn't be prudent. In other words, reliability/trustworthiness is its own thing, its own obstacle, and only tangentially related to intelligence, if at all. There are highly intelligent liars, for example and conversely, if you give a principled, intelligent person flawed information, they will naturally arrive at flawed conclusions. The quality/trustworthiness of information is just as important (if not more) than the capacity to analyze it intelligently, and the process for establishing the quality of information is through research, not by "being smarter."

Granted, ML algorithms can potentially expedite analysis, but it's still limited by the quality of data, which is not something I believe intelligence can inherently improve. I am open to that possibility; I just haven't really seen anyone explain how that might happen (let alone provide a testable explanation). Most people just wave a magic wand and say smarter = faster.

It can only happen that way because that's the way it happened. I believe that's called confirmation bias.

In any case, we already have access to essentially unlimited energy through fission. Before that we had inexhaustible (on the timeline of centuries) geothermal energy. It wasn't exploited earlier or more extensively because we had hydrocarbons, which were portable and thus doubled as convenient fuel for vehicles. But in the absence of abundant hydrocarbons, we might have developed a more robust electrified transport system. In fact, this was one competing vision back when motorized transport began. The fact that hydrocarbon-based transportation won the day doesn't mean electrified transport was infeasible, or that technological progress would have stagnated.

Progress in the absence of natural repositories of hydrocarbons might have taken longer (on human timescales), but not necessarily, and in any case the difference likely would have been insignificant on geological timelines.

If you are trying to explain why we haven't detected any aliens, how is "they were massacred by even more advanced aliens" a remotely adequate answer? That just leaves you with "why haven't we detected the even more advanced aliens?". The question was never "why do we detect so many deathbots and so few little green men?"

If anything, superintelligences are presumably more capable of doing high-visibility things(if they want to) by virtue of being more advanced; and, while they could all be carefully hiding because they're paranoid that same explanation would hold for standard aliens as well.

Seems like an awful lot of hypothesis to explain nothing.

... anything written by a government drone? AI might be a step up.