It may be pathetic; but the 'make america great again unless it sounds like actual work might be involved' attitude is, at least, consistent.

It's possible that the summary is missing an important qualification; but wouldn't it only be possible, even in principle, to conclude that something could or couldn't be a simulation on a specific type of computer rather than in general?

If, say, you were able to demonstrate that you had an actual RNG, not a pseudorandom number generator, you'd know that it isn't being simulated on a turing machine; because those do determinism only. However, in practice, we build computers with what we think are RNGs all the time; because connecting the deterministic finite state machine to a peripheral that's full of thermal noise or radioisotopes or lava lamps or whatever is a totally doable design decision. Were someone in one of our simulations to conclude that non-deterministic behavior falsified the simulation theory they'd simply be wrong; because (it appears) that the reality in which we construct our computers is a little stingy when it comes to things like infinite state storage; but reasonably helpful on high quality entropy.

In the case of this 'non-algorithmic understanding'; it sounds like you may be successfully demonstrating that the simulation would only be viable on a somewhat more exotic machine; but basically just one that has a lookup table attached that it can use to check whether an unprovable statement is true or not. I would not want to be the one stuck building such a device; but it doesn't sound any more exotic than quite a few of the various 'oracle machines' that are supposed, for purposes of theorizing about computability and complexity, to have a black box capable of solving certain classes of problem.

We even interact with a much humbler example of an analogous thing more or less all the time: the reason we bother with storage devices is that there's no way to know what a given series of bits "should" be. Absolutely trivial(assuming sufficient time and RAM) to go through all possibilities for what it might be; but no way to decide between the possibilities. So we suck it up and plug in our flash drive; then copy off the cat picture that we actually want. Essentially a block device is an oracle that answers the otherwise algorithmically impossible question of "what is the state of those n bits?".

I don't say this out of any particular affinity for, or belief in, 'simulationism'; and further acknowledge that the authors may have made a meaningful(but rather narrower) statement by formalizing certain requirements for what a simulator would be required to be capable of; I'm just unclear on how you could make the claim to have disproved simulation, in general, unless you managed to come up with something that could not be implemented as an oracle even in principle, which it doesn't sound like they have.

It does seem to at least suggest the possibility of excluding 'trivial' nesting of simulations: someone simulating us would appear to need hardware that we would not be able to implement under the rules we are provided; just as someone in a deterministic simulation wouldn't be able to implement an RNG, which we at least appear to be able to do(at least, if they are PRNGs, they hide their state somewhere very cryptic); so if there is anyone out there who thinks that it's totally possible that, like, the universe is just big 486s all the way down, man, it would appear that they are on thin ice theoretically, with at least some details suggesting that the simulator need be fundamentally more capable, rather than just bigger, than a system that can be implemented within the simulation; but my impression is that any serious consideration of trivially nested simulations had foundered purely on the size problem among all but the densest rocko's baselisk bros already.

I'm a little unclear on what anyone thought this elaboration was getting them; unless it was purely pessimism about the existence of any sort of untapped channel where cute but relatively crude steganography wouldn't be necessary or could be better-handled by any of the myriad excuses to send bits of encrypted information(altering the agreed-upon portions of encrypted JWTs returned by some auth endpoint or the like).

There's the very specific case of 'warrant canaries', for which there's some US case law around compelled speech vs. compelled nondisclosure that might given them better constitutional coverage vs. just ignoring gag orders; but even that is a matter of some uncertainty; and this sounds like it's both more expansive in terms of what jurisdictions could take a dim view of it and much more overt in just being an obfuscated disclosure.

Clearly if the obfuscation keeps you from getting caught that can work; but as a legal strategy this seems to be a straightforward "just flout the order" that would be relatively simple for any peevish feds to prosecute accordingly; quite possibly even providing a few extras to throw in because doing financial transfers to facilitate crime sometimes counts as an additional issue.

The definitions of 'advance or promote'; and 'equity ideology' are as well. You are basically looking at a situation where you could get hit with a $1.5 million clawback at any time for more or less anything someone at least vaguely connected to the PSF says that someone ends up feeling thin skinned about.

We're not even talking having to do anything: one probably-justified comment about how many people are going to get ICEd on the way to PyCon US this year would, in theory, be readable as falling under Executive Order 2(viii) " the United States is fundamentally racist, sexist, or otherwise discriminatory."

Or, on the even-harder-to-avoid and less inflammatory side; it could just be someone doing vibe statistics about PSF grant recipients (257 groups or individuals last year; so a decent sized sample if the coming year or two aren't wildly lower) and kicking up a fuss on twitter about how they don't seem perfectly demographically matched to the ideal techbro. Wouldn't even need to be terribly plausible or statistically significant, just enough to chum the water a little.

If this were actually just about who gets hired to execute the work specifically funded by the grant the risk would at least be manageable enough to actually treat it as a meaningful choice you are being asked to make, rather than just a sword of Damocles.

The reason for the rules seems like common sense to me. There is a certain distance needed to stop or change lanes when driving at highway distance. If the truck breaks down just over a hill, cars won't see it early enough unless the warning signals are put further back where they can be seen coming up the hill.

I seriously doubt that these rules were just shit someone made up. The NHTSA has so many studies regarding road regulations and guidance. They might be outdated for modern technology, and might be worse than newer alternatives - I don't doubt that hasn't been studied yet - but I would absolutely wager that there were studies done to justify the original numbers.

Furthermore, when congress delegated regulatory power to these agencies they included laws dictating how the rules needed to be determined, specifically so you can't have a bunch of political hacks changing them on a whim. Changes to the regulation need to be justified, and there needs to be comment period to gather any information and concerns that the agency itself might have overlooked, respond to the comments and incorporate any changes as appropriate. I don't want regulators to be able say "this is just some crap" and change rules every four years because they shoot from the hip. That means that changes take 1-3 years depending on how complicated and motivated the agency is, but it is worthwhile to end up with better regulations and avoid being constantly jerked around.

I don't know that MS has been caught doing data transfers specifically(though they'd have to screw it up or have it leaked at a fairly high level to get caught; 'cloud' is basically always opaque on the back end as far as the customer can see); but there have been a couple of instances recently of service getting cancelled. When Trump got into a snit with the ICC cut their chief prosecutor off(Brad Smith mollified more or less nobody with the claim that they didn't cancel service to the ICC, just to the senior official that the feds were upset with, which is probably technically true in the sense of account GUIDs but not usefully true); and the also kicked Unit 8200 out of their cozy custom Azure environment; though apparently with enough notice that they were able to move the data somewhere else.

It seems likely that random European corporations see themselves as lower profile and less vulnerable than the ICC or Israeli military intelligence; but if anyone doing risk assessment for them hasn't at least considered the fact that basically a belligerent old man would just have to decide that they are 'very unfair' tomorrow; or that someone other than greenland needs to be brought into the homeland, and that would potentially be all it takes for your MS EA to just stop talking to you then they aren't doing their jobs very thoroughly.

"Generative" they said...

In the new CEO's expertly crafted I-am-villain statement; he explains: "The past, he added in an email, is “simply not relevant to the new future.”"

I, um, guess it was polite of him to not talk about "the bigger picture" to the people whose eyes he shut down?

Will these ones retain vendor support longer than the last batch?

I suspect that they feel at least incrementally less burned in this case; since, while it wasn't obviously a good idea for a product, it at least goes somewhere: if you can make a phone functional and adequately rigid at that size; it's quite possible that there's a more sensible device size that you can still apply the miniaturized motherboard and whatever mechanical engineering you did for rigidity to; and just fill the rest of the case with battery; and there may be some other cases where the ability to get an entire SoC and supporting components into a particularly tiny area or make a thin component of a larger system quite rigid is handy.

Still doesn't really explain flaying a normal phone until it barely has a normal day's use with a totally fresh battery when you are still going to glue an entire baby spy satellite to one end of it; but some of the actual engineering is probably reusable.
The 'butterfly' keyboards, or the under-mouse charging port, by contrast, went nowhere. They tried and failed at a few iterations of keyboards that committed expensive suicide if you looked at them wrong; then just went back to allocating the extra mm or whatever once Jony was safely out of the picture; and it's not as though putting the port on the bottom rather than the front of the mouse involved any interesting capability development.

Whatever product manager thought that the 'air' would be a big seller deserves to feel bad; but the actual engineering team can probably feel OK about the odds that a future phone will look somewhat air-like if you were to remove the normally shaped case and larger battery.

> And, in that case, why add a new law versus just adding to the CCPA?
It is an amendment to the CCPA. Websites covered by the CCPA are required by law to honor the opt-out, and browsers are required to implement it.

It's stupid; but at least more honest than trying to dress 'grok' up as the master control computer.

In the MS case; it wouldn't be too surprising if that order is also the one that urgency dictates. Neither is totally unavailable on-prem only; or entirely without more-chatty-than-one-would-like behavior; but if your concern is about your dependence and Redmond's potential direct control their groupware stuff is moving faster than their OSes(at least if you have enterprise licenses and someone to handle keeping them quiet) in the direction of pure SaaS.

You'll get some nagging about how Azure Arc is definitely the cool kid's future of glorious hybrid manageability; but your ability to run Windows as though it were 15 years ago is definitely greater than your ability to run Office that way.

I suspect that this won't be the last case we see; as MS has shown comparatively little interest in backing down on the future being azure SaaS, and there's no real equivalent to some steep but temporary discounts for dealing with people who have fundamental privacy and operational control issues; while it's not terribly challenging to find a special discount that makes sticking with the status quo look cheaper than trying to do a migration.

I obviously don't expect better from these sorts of people; but I'm honestly puzzled as to why they would turn the screws so quickly and blatantly despite having gone to all the trouble of a reshuffle and a new lineup and some spiel about being likeable rather than Alexa just being something that you sort of poke at because Prime members were given a free surveillance puck with some offer one time.

Is Panay one of those abhuman lunatics who genuinely thinks that the only objection to relentless advertising is that it isn't "relevant" or "engaging" enough? Does he have a scorpion nature that leads him to knowingly doom his own product just because that's what he is? Is he just a figurehead who got to choose the case plastics colors and smile on stage; but some adtech business unit calls all the shots?

I'd fully expect this sort of thing to betray you; but only after enough of a honeymoon period for people to be pleasantly surprised by the behavior of the launch units so that there is actually enough of an install base to betray.

It sure is a good thing that 'AI' companies are notoriously discerning and selective about their training inputs and not doing something risky like battering on anything with an IP address and an ability to emit text in the desperate search for more; so this should be a purely theoretical concern.

Snark aside, I'd be very curious how viable this would be as an anti-scraper payload. Unlikely to be impossible to counter; but if the objective is mostly to increase their cost and risk when they trespass outside the bounds of robots.txt something that will just look a trifle nonsensical in places to a human but could cause real trouble if folded into a training set seems like it could be quite useful.