Just for the fun of it, I looked up some of the earliest stories about fusion on Slashdot. Take this one from 2002: U.S. to Rejoin the ITER Fusion Project. It reports, "The USA left the ITER consortinum in 1999 when it bulked at the 10 Billion dollar price tag." Here we are 24 years later: ITER still hasn't been completed. The official estimate of the total construction cost is $22 billion, but that estimate is considered totally unrealistic. In 2018 the US DOE estimated the actual construction cost at $65 billion.

Then there's this one from 2002: British Researchers Say Fusion Is Close. It mentions "a leading scientist saying that Fusion power is 'within reach' in the next decade, with commercial plants to follow within another 10 or so years." I think that's still about the time frame people are predicting?

Here's the very earliest one I found, from April 1999. It talks about "the pittance of money that is being put into the research, versus the known benefits of making advances in this."

It's nice that some things never change.

Shell shock was something much more specific. It referred to the condition soldiers developed after experiencing intense bombardment. We now recognize it as a special case of PTSD, which is a much broader category. PTSD can happen to anyone who's been through a traumatic experience: victims of violent crimes, patients who have recovered from severe illnesses, people in abusive relationships, etc. Most of them have nothing to do with being shelled.

For most applications, biofuels are a bad idea. Long distance plane flights are possibly an exception. Batteries are unlikely to have enough density for many years, if ever, so using a chemical fuel is essential. If you want to decarbonize long distance air travel, finding an alternative to fossil fuel is the only option. It remains to be seen whether that will end up being biofuel or a synthetic fuel made directly from CO2. Both are possible.

You're quite the pessimist! But seriously, AI has "never delivered anything besides some small improvements"? I guess you count ChatGPT as only a "small improvement" over ELIZA? Stable Diffusion is only a "small improvement" over procedural image generators from the 90s? Neural machine translation is only a small improvement over the rule based translators we had before? AlphaGo crushing the world champion is only a small improvement over the programs we had before that played as mid-level amateurs? AlphaFold is only a small improvement over the structure prediction programs we had before? Neural computer vision systems are only a small improvement over what we had before?

Seriously?

If you really believe those things, you need to get acquainted with the modern world. The advances made by AI are huge, dramatic, pervasive, and changing everything.

Everything about that statement is wrong. They started from a database of 150,000 known materials. They came up with 2.2 million new materials that they predicted could possibly exist, but not necessarily be stable. Of those, they predicted that 380,000 should be stable. Only about 700 of them have actually been verified experimentally. That means the number of known stable materials has grown by about 0.5%.

Wired has a better article about it.

They released their predictions under a license that forbids commercial use. I don't know what would happen if they tried to enforce that. You can't patent a material, only a use of the material. Copyright isn't designed for this sort of thing. They can't claim trade secret, since they published it. So it's not clear they have any legal basis to restrict it.

I think the main competitor is hydrocarbon fuels, not batteries. Methane's weight density isn't quite as good as hydrogen's, but it's not bad, and it's so much easier to work with. It's easy to store. You don't need super-refrigeration or very high pressure. It also is easy to make from renewable energy, and the efficiency is only slightly lower than making hydrogen.

Think about what they're saying. The largest state has the highest fraction of people moving from somewhere else in the same state (which is the largest state, so what did you expect?).

The next largest source of people is the next largest state (Texas).

The third largest source of people is the fourth largest state (New York).

Seriously, what did you expect? The bigger a state is, the bigger a source of people moving anywhere it's likely to be. Then everyone takes that obvious fact and tries to spin it into support for whatever their favorite political narrative is. No. This doesn't support your political narrative. It's just the trivial fact that people moving in California are more likely to come from places with a lot of people than places with few people.

A lot of code today is really inefficient because there's no reason to optimize it. But not game engines. That's one place where people still care about performance. A lot of work goes into squeezing out every last fps.

I think a bigger difference is that there's a lot less cheating today. Old school demos are all about cheating. You try to do simple things that look like you're doing something complicated. Like pseudo 3D, or warping images in simple ways, or modifying a color lookup table to animate the whole screen at once. Games today don't need to cheat. They can do real 3D. They can use physically accurate lighting. They can do actual raytracing instead of faking it. It looks better and it's a lot easier for game developers. You just tell the engine what you want and you get it. You don't need to figure out how to fake every effect. It takes a lot more computing power, but modern computers can afford it.

Oracle doesn't have a lot to do with it anymore. Most of the community has moved to OpenJDK. Java on Android is based on OpenJDK. Oracle can't do anything about it.

Kotlin isn't Java. It's a multiplatform language with multiple backends. You can compile it to Java bytecode to get interop with Java libraries, or you can compile it to a native application, or compile it to WebAssembly to run in a browser and get interop with Javascript libraries. It's come a long way from its starting point as just a Java alternative.

A quantum computer can execute the algorithm in polynomial time. A classical computer emulating a quantum computer takes exponential time.

In what way doesn't the form factor compare? This seems like a good design to me. Monitors last a lot longer than computers. Why replace your monitor every time you replace your computer? Separating them saves money and reduces waste. And it's not like it's taking a lot more desk space. The mini is a really small box.

There's a bit of an irony here, with the article quoting someone describing the Cruise CEO as "very Silicon Valley". Waymo is an actual Silicon Valley company (started by Google), but they've taken things slowly and carefully, almost to a fault. Safety really has been their top priority. Cruise is part of a traditional car company (GM) that ought to have safety deeply embedded in their culture. Instead they sacrificed safety in the rush to get a product out as fast as possible. The roles seem to have switched, with the traditional company more "Silicon Valley" than the actual Silicon Valley one.

Even better, leave it up to each game. Let them query whether an unofficial controller is connected. Developers of competitive games will probably choose to block it. Developers of casual or social oriented games will probably choose not too. There are lots of online games where cheating isn't a concern.

I don't think it really does that. Notice the carefully chosen wording.

All they're saying is they won't use it to show you ads. It doesn't say they won't still collect every bit of data they can about it. It doesn't say they won't use it in other ways, like selling it to data brokers. You aren't getting more privacy. You're just getting fewer ads.