You don't have to take over a country to get that. The U.S. brain drains countries that aren't democratic just fine. There are a few that we arbitrarily choose not to brain drain or sell to (currently Cuba, Iran, and North Korea, I think), but very few countries are on that list. And Cuba was one of the countries that I was thinking of as a country that has nothing worth taking over, because if it did, then those gains combined with eliminating the risks associated with having a country like that so close to our borders would probably have been enough to tempt someone to overthrow its government long ago.

Then again, Republican war hawks have started thumping their chests about Cuba now, so who knows. It might still happen.

Only if you have something that someone else needs, such as oil. Without that, nobody will bother.

Given how corruption usually works, they probably sold them to some even worse third-world country and deposited the resulting money in a private Swiss bank account.

No, that's actually the only correct interpretation of those words in English. If someone's actions caused the leaking to stop, then a native English speaker would say, "After Half a Decade, the Russian Space Station Segment Leaks Are Fixed" or "... Leaks Have Been Stopped/Fixed". (Or they would avoid passive voice and explicitly say who stopped it.)

Saying that something stopped is different from saying that something was stopped. That helping verb indicates that there is another actor other than the leak who caused the event. Without it, the only correct way to interpret that sentence is that the leak stopped on its own, and that nothing was done to make it stop.

This headline would have caused my newswriting professor's forehead veins to rupture.

I'm not sure how that's possible. Unless Wikipedia is wrong and Google search is equally failing to find a later version, the last version of iOS that the iPhone 5s supports is iOS 12 (2018). The last version of iOS 12 ever released was iOS 12.5.7, which was released in January of 2023.

Are you sure you're not talking about the iPhone 6s?

The last security update for the iPhone 5s was in January of 2023 — three years ago.

Security updates.

At no point did I claim that size doesn't matter. I said thickness doesn't matter, at least within reason. Width and height both matter a LOT, because they dictate how hard it is for you to hold it in your hand and control the phone without having to use both hands. Thickness does not meaningfully affect usability unless you are carrying it in your pocket, and only to a limited extent even then.

What I'm asking for isn't even difficult. You have three dimensions, and they are mostly interchangeable. If you need more board space, just stick a second board behind the first with an interconnect, and you're done, so long as you're not sending high-speed data or something. If you need more battery space, stack them front to back. Use the third dimension instead of being a moron who focuses on making products thinner — something that exactly NONE of your customers are asking for, BTW, and so many design problems become so much simpler.

Now to be fair, the phone I want will *massively* piss off all of the app UI designers who try to cram too damn many buttons into their user interfaces as they try to find ways to scale back down to a 5s-sized screen, but that's somebody else's problem.

Only unlocks 40% of the time, and that's without a mask. 10% with. Requires you to look towards the phone, which makes it impossible to unlock it while you're pulling it out of your pocket before paying attention to it. And the lack of a home button means a much clumsier overall experience. Home becomes swipe up from the bottom, which is hard to do with a case. Double-home becomes a very careful swipe up from the bottom, being careful not to go too far, and you will screw it up two times out of every three.

I'd also be okay with a 5s-sized foldable with a screen on the outside and a home button. Unfold if you want a bigger screen.

Yes, please. And give one version touch ID and a physical home button. I still hate the gesture-based interface even after using it for a long time. It is way too easy to accidentally do things that you did not intend, and way too clunky to do things that you do intend, like swiping to kill an app or turning on the flashlight. The need to touch the screen right at the edge is problematic with cases, which makes the design even worse.

And Face ID still isn't half as reliable for me as Touch ID was, and worse, demands attention, or at least having my face pointed in the general direction of the phone, to work, which means I can't unlock the phone while doing something else. That extra several seconds of latency used to not matter, because I could be doing something else while waiting for the phone to unlock, only occasionally glancing at it to see if it had finished. But with Touch ID, you incur the entire latency penalty by design.

I'd also like it to be smaller — no bigger than the 6s, and ideally closer to the 5s. That way, I can usably operate it one-handed. Even the most recent "Mini" is huge from my perspective, at almost an inch larger in the diagonal direction than the 6s, which was already too big.

What I want is an iPhone 17 Micro that's about the size of an iPhone 5s, but with the same cameras and CPU and maximum storage as the iPhone 17 Pro model has, so that I'm not having to settle for a cut-rate feature set in exchange for a usable form factor. If it has to be thicker for battery life reasons, that's fine. It's going in a case with a holster anyway, and getting clipped to a holster on my belt. Thickness doesn't matter. Features matter. Width matters. Height matters. Usability matters.

Some houses are inaccessible, sure, but there are plenty of disabled-accessible hotels and condos and stuff.

That's probably a third of SF. And before you say "make them one way", that is likely to increase pedestrian deaths.

True. But buses also use those streets. So while you could theoretically do this, you'd have to start by building a proper subway system. The cost of doing so would likely be infeasible. It would probably be orders of magnitude cheaper to build a second-story pedestrian walkway system with stairs and elevators down to the sidewalks in every block. :-)

Additionally, there's the ADA problem. Having large distances between the nearest road and housing or businesses can make cities inaccessible for the elderly and disabled. I suspect that the folks advocating for car-free cities have never tried pushing an elderly parent in a transfer chair around the streets of SF. The real-world fallout from such designs would be devastating, IMO.

Sure it does. You haven't fully thought this through. Traffic flow is all about minimizing latency. Say you have a two-cycle light:

• 30 seconds in direction A
• 30 seconds in direction B

A pedestrian crosses the road in cycle A, preventing the traffic in that direction from turning, which prevents the cars behind from going straight. Those cars wait for 30 seconds in direction A, then 30 seconds in direction B. Latency is 60 seconds.

Worse, nothing prevents this from happening two cycles in a row. Latency can then grow to 120 seconds, 180 seconds, etc. I've sometimes had to wait at the same intersection in SF for as many as three cycles in a row because of pedestrians preventing cars from turning right. When you have to spend five or six minutes at two or three intersections in a row, you could literally walk much faster. The worst case vehicle latency is arbitrarily bad with this design.

Contrast with:

• 30 seconds in direction A (no pedestrians)
• 30 seconds in direction B (no pedestrians)
• 15 seconds in all directions (pedestrians only)

Maximum latency for cars is 45 seconds.

This approach is also roughly break-even for pedestrians, with the worst case being probably 45 seconds of latency for the first (assuming buttons are required) and 60 seconds for the second. But for the second, you get to cross both ways at once, which can save time and energy.

You can also put two 15-second walk cycles in, and your latency for walking gets cut roughly in half, and your latency for driving remains the same in cases where only one cycle gets blocked by pedestrians. And in cases where multiple cycles in a row get blocked by pedestrians, the latency for driving can be arbitrarily better.

Just ban gasoline cars within the city center like Paris is doing (by 2030). Noise and pollution are largely a solved problem at that point, assuming you run street sweepers often enough to clean up the tire dust.

As for space, having to dedicate one parcel every block or two to parking is a small price to pay for being able to get around... unless you mean the streets themselves, in which case I would point out that streets provide natural light for the buildings. They aren't wasted space. They're a health necessity.

Let's go down the list.

• Plumbers: There's nothing particular complicated about most plumbing work. Gluing PVC together is trivial to automate, as it is not significantly more complicated than pick-and-place. For the most part, figuring out the path that pipes should take is also trivial to automate. Cleaning out clogged sewer pipes is already done by machines and is trivial to automate.
  
  This pretty much leaves the sorts of plumbing where wall removal and reconstruction is required, and that's a small percentage of jobs. For the 95% case, plumbers are cheap enough that nobody is willing to spend the money to train the models to do a lot of the work, but there's nothing inherently hard about it, and IMO it is well within the limits of current-generation AI.
• Electricians: Building a robot to change out a light switch or electrical outlet would be relatively trivial with current-generation AI. Much like plumbing, for tasks involving drilling and running new lines, yes, you'd need a robot capable of doing actual remodeling-grade construction work, which probably won't happen in the next five years, and might not happen in the next ten, but it will happen.
• House painters: This is basically Roomba-level automation. The only reason it isn't fully automated is because there are enough laborers willing to work for near minimum wage. And as more people find themselves without jobs, the supply of laborers will only increase, so this will probably not be automated in the foreseeable future. But the only reason it won't be automated is because despite being something that an engineer could work up in an afternoon, the cost of renting the hardware would exceed the cost of hiring a day laborer out in front of Home Depot and paying them cash.
• Checkout clerks have already been 90% replaced by self-check registers, and some stores even use AI and cameras to eliminate checkout entirely. So this isn't just already possible; it's already happening.
• Street sweepers are also trivial once you have self-driving car tech. So like house painting, this will remain non-automated until the cost of retrofitting the self-driving hardware falls below the ten-year cost of paying someone to drive it. I'd give it five to ten years.
• Janitorial jobs are already being replaced by robot vacuum cleaners up to a point. Within a decade, these jobs will probably gone, too, at least in commercial settings.
• Landscaping is already done using a lot of machines. With LIDAR, you could easily design an automated grading machine that adjusts the shape of the land to spec. You could easily design a tree and shrub planting bot, a bot for planting small plants, etc. You could easily design a robot to build walls. None of the physical labor tasks seem out of reach for automation from my perspective. You would still probably want a human being to do the actual design work.
• Software engineering is already being done with AI to a limited degree. Hardware engineering: Trace optimization and other tedious tasks are already being done with AI. So automation in both of these areas has the potential to reduce the number of jobs, though probably not dramatically.

Even construction will be affected, because new construction by robot is likely to be relatively easy, and at some point, tearing a building down and starting over becomes cheaper. Either way, automating new construction massively frees up people to do repairs and remodeling, and there will be way too many people for the amount of work at that point.

That pretty much leaves skilled nursing and education as the last bastions of employment.

Yes, it is.