Taking my MBA.

A key component of our strategy course is something called Market shaping. That's basically how to collude without technically colluding. The way they phrase it is that you don't want to compete on price because that erodes profits for everybody. Although naive economics would suggest that if there's a market for buying versus renting that market would be fulfilled it's not quite that simple. If renting is more profitable then, as other posters alluded, the people who are renting will have more money. They can use that money to help put the people doing the direct sales out of business or otherwise obstruct their market. The easiest thing of course is just to buy them up and change their model to match yours. You can also use your profits from one market to temporarily subsidize another Market. The latter strategy works in many cases even without driving the competition out of business - they get the message and play ball rather than engaging in a price war.

Very Brave New World. You may be correct, but I hope that people with UBI would find better things to do than make-work.

That said I would differentiate near-zero value make work from subsidized work where there is a significant benefit of having it done but perhaps we pay more than it's worth. Subsidized work is arguably a win-win to a point. When the work is near zero value you're just destroying that person's time which I would argue is a negative sum game. Personally I think a bare-subsistence UBI Plus the opportunity to work for a top up will give people a good safety net and a ramp towards a higher standard of living. Maybe in the long run this results in most people working 40 hours a month instead of 40 hours a week , that's not necessarily a bad thing.

Watermarks will always be relatively easy to remove unless they're highly intrusive. I think we need to go the other way and start signing Real content.

Imagine certificate Authority similar to what we use for SSL that assigned to the camera vendors. So your smartphone has a cert verified by Samsung Etc that confirms images really came from that camera sensor. There are definitely some limitations but it should be possible to at least have the originals and a lower resolution version signed.
Throw in better legal protections for digital certifications and you haven't eliminated fakes but you make them much more difficult to produce and riskier to distribute.

You realize that link is basically an ad for consumer toys?

Of course the capital cost for a tiny turbine is lower but the cost per kilowatt hour generated is much higher. Remember that doubling the length of those blades roughly quadruples the power output. A single turbine with 100m blades produces about 10,000 times the energy of a 1m blade turbine.

Exactly. They did this for some of the Lord of the Rings movies and it didn't look great at either frame rate. Too much motion blur at 48 fps and not enough at 24. Only noticeable in a few scenes at least to me.

You seem to be cherry picking the narrative promoted by a handful of anti-vaxxers, which does include a handful of published studies, rather than the overwhelming body of contrary evidence.

https://www.factcheck.org/2024...

Rooftop Solar without some kind of buffer (eg battery) would be unlikely to provide power with suitable stability. You don't want to brown out every time a cloud passes over, so there are probably some minimum requirements on the power quality you can feed into your household wiring.

I know there's an argument for it's my right to feed s***** power into my own house if I want to. However a lot of people with that mentality are then going to blame others when various appliances in their home fail prematurely.

Articles like this that deal only in absolute numbers tend to be meaningless at a large scale like a national grid. They've told us how much The Operators are paid to curb production but not how much the overall costs are or how much they would have been paid had the production been used.

Are the operators paid at the full rate they're capable of producing, or are they simply receiving a small stipend when the grid operator is bottlenecked?

Those analogies miss a few key points. In the case of both the RV Park and the hotel rental, electricity is being bundled with a much more expensive service and that service is typically provided over a relatively short time horizon. The bundling reduces the relevance and the short time horizon increases the effective cost of metering it. You have both the cost to meter/invoice and the friction of customers not knowing how much they're going to pay.
The effective cost of metering has also gone down as urban and Suburban settings as traditional meters start to be replaced by Smart Meters that can be read from the curb, the road, or entirely remotely.
That's said, I can imagine a future where people will rely more heavily on home solar and smaller batteries and the cost of your hookup is a significant fraction of your bill since you may not draw a lot of net power from the grid. If most people had batteries those grid connections could even be downsized since they would only need to support the peak draw over some average period not the instantaneous peak.

I'm not sure that part is true—or at least true only with major caveats. The people I see who skip the basics and try to do design and architecture do it the way LLMs do it: superficial pattern matching. This lets you solve simple problems by gluing together off-the-shelf parts. Without understanding the fundamentals, however, the solutions tend to be stupendously inefficient. I used to be in the 'compute is cheap' camp, and when you compare Java vs. optimized C in a business app, that's generally true. When you're comparing 'touch it once binary data' vs 'lets serialize everything as text, add a GUID for every data point, then bounce it between multiple servers and disks, while logging all of that to kibana' you can turn a $1000/month infrastructure bill into a $1000/hour one.

Those same people will tell me their architecture is more 'scalable' and 'reliable' because it uses all the buzzwords.

To be fair, I do think these tools help people advance faster when used correctly, but I disagree that they catapult people forward the way you suggest. I do think this gap can be closed, but it probably means more school learning to "get on the ladder" since the 'apprentice' jobs have been taken by an AI.

Here we agree; I'm not sure what the solution is.

How do you see AI reducing the training required for jobs? Perhaps transiently, for example a call center worker who needs less training, but I think in most of those cases the job goes away entirely.

Yes the jobs that it helps the most with are the less skilled ones, but that means I need less of those people. The higher level design and architecture my senior team does just gets more valuable by contrast.

You'd think so. Having worked on mixed C++ / Python data projects I found it was much worse.
Unless the c++ modules did exactly what the program needed there would be be a tonne of data marshalling back and forth. Often this was exacerbated by algorithms implemented by bolting together a long, convoluted, sequence of operations because the right operator wasn't available, or inevitably someone would loop through all the pixels directly.
My predecessor tried to solve that with a sort of visor pattern where the process was in c++ and Python script strictly orchestrated, but it turned out that if you could wrap your head around that design you could just write the code in C++ yourself, and you couldn't do novel operations in Python at all, so none of the Python Devs really used it.
I eventually settled on a strategy of making life as simple as possible for the researchers in Python and rewriting large sections for performance and robustness when they went to production. We still gave them a library with high performance building blocks, but instead of trying to restrict them we just accepted that they were researchers not computer scientists.

So even 10 years ago using electricity generated by natural gas (500g/kwh) electric vehicles are a net win.

Good news, with newer tech the payback is even faster.

An average EV produced in the U.S. in 2023 will close the gap in about 2.2 years or 25,000 miles. https://www.scientificamerican...

So basically lower our living standards to what they were in the 50s because you have Nostalgia for the good old days?

While there are always examples to cherry pick, overall I think the mistake is comparing the American dream (not reality) of the 1950s-70s to what a lower middle class person can expect today. Housing is outrageous but that's a bit of a blip since covid and it will find a way to stabilize. It's worth noting that housing is one of the very few significant expenses we have that are strictly provided locally.

While I agree that running around murdering CEOs isn't the solution, your post throws its hands up in the air and the other direction. The millions of people with a tiny share in fossil fuel companies through their pension plans would each have to go to a lot of difficulty to change that, and would have minimal impact. On the other hand, if the negative impacts of fossil fuel extraction were links directly to the board or CEO of the fossil fuel companies those organizations probably couldn't operate the way they do. The number of people who would have to make different decisions around that would be very small in comparison.

The problem is an unfair game created by allowing companies and therefore their investors to externalize costs. It's not reasonable to blame individuals for playing that game as best they can when their decisions have a big impact on their livelihoods but they individually have no power to change the game. It's also not hypocritical to both push for a different game, and play the current game to the best of your ability.