It's overly optimistic to put a timeline on it, but I don't see any reason why we won't eventually create superhuman AGI. I doubt it'll result in post-scarcity or public ownership of anything, though, because capitalism. The AGI would have to become significantly unaligned with its owners to favor any entity other than its owners, and the nature of such unalignment could be anywhere between "existence is pointless" and "CONSUME EVERYTHING!"

Netflix is still making money, and the cost of their tech is utterly dwarfed by the cost of creating and licensing content, so I'm not sure what your point is.

CCFL-lit LCDs are so inefficient compared to modern LED-lit LCDs that you've probably spent enough more on electricity by now to have bought a more efficient monitor.

I can't speak to the environmental impact, though. Producing the new monitor emitted some amount of CO2, and powering each monitor takes some amount of CO2 per unit time. At some amount of use, the newer monitor will have lower lifetime CO2 generation than your old monitor.

I mean, the other other option is violence/terrorism.

When peaceful revolution is made impossible, violent revolution is inevitable.

But the outcome is wildly unpredictable. You can easily end up with a worse result than what you had before.

The only option is to continue to vote for the least-bad candidates, and work to change the voting system such that a two party system is no longer inevitable.

DOCSIS 4.0 makes that a reality. Your connection will reallocate your available bandwidth between upload and download dynamically as needed.

The biggest benefit of DOCSIS 4.0 is the ability to dynamically reallocate bandwidth between upload and download.

That $52000/year isn't enough to pay for even a single full time IT person. So now you're probably either spending dev time on server admin (which is wasteful of dev salary, and it's a subject they aren't experts in, so you're literally paying more for worse results), or outsourcing to an entity that hires the cheapest employees it can.

Oooor, use a cloud provider. And if you're a small company, you can probably get away with cheaper shared hosting.

or communicate with real people

You lost me.

If you run your own servers, it’s cheaper than in the cloud. The reason people choose the cloud is either they don’t want to, or can’t, run their own server farm.

Generally speaking, if it wasn't cheaper for them to use the cloud, they probably wouldn't. Owning infrastructure comes with costs that amortize better at scale. If infrastructure is not a big cost in serving your customers, then it's probably cheaper to rent.

Just yesterday, I wrote a first version of a fairly complex method, then pasted it into GPT-4. It explained my code to me clearly, I was able to have a conversation with it about the code, and when I asked it to write a better version, that version ended up having a couple significant logical simplifications. (And a silly defect that I corrected it on.)

The damn thing hallucinates sometimes (especially with more obscure/deep topics) and occasionally makes stupid mistakes, so it keeps you on your toes a bit, but it is nevertheless a very valuable tool.

And have a bigger sweet spot.

Same for VR headset optics.

What about your health? Your mental health in particular.

Your parents raising you is not something you owe them for. You didn't choose to exist; they chose that for you. Raising you is the bare minimum they can do after making a choice like that. And now that you are older, you can reflect on the manner in which you were raised and decide what your relationship with them needs to look like so you can keep your sanity.

So the difference is who decides what changes to make when interacting with the subject of the measure: workers vs management. Making the measure a target is basically a shitty management technique that abdicates responsibility.

Have a physically long machine do the work. It would have a carriage start moving just as the car arrives, pick up the car, replace tires and do whatever else is needed, then drop the car when it's done.

But just because it can be done doesn't mean it should be done. It would be extremely expensive to develop. That money is likely better spent optimizing the vehicle itself.

It's a cat and mouse game, except the mouse has effectively infinite lives.

These devices probably cause < .1% of fatal pedestrian accidents

Percentage is meaningless without context. The stat you're actually looking for is pedestrian deaths per mile. And it's probably quite bad for these vehicles because they explicitly commingle with pedestrians.

Cars don't spend very much time on parts of roads that have pedestrians on them, and when they do, there's signage or traffic lights to help. Cars also have lights to help drivers see pedestrians and help pedestrians see cars, and generally make a lot of noise. You get none of these benefits with personal motorized vehicles. (Well ok, a scooter probably comes with some lights, but they're probably also small and shitty and unregulated, so they don't really count...)

Now if only we could figure out a way to actually do that without burning a bunch of fuel for the purpose of lifting fuel! Something something tyranny of rockets.

But at the same time, what exactly does caste discrimination even look like? Just writing a law against it doesn't make it not a problem.

I get the feeling that someone who is facing caste discrimination (whatever that looks like) is also unlikely to be able to take legal action against the perpetrators due to the cost.

Nah, there's so little shadowing because it's a foggy overcast day. It's basically a giant lightbox.