Solar power is now cheaper than coal in good locations, and aluminum smelting is an interruptible process (smelters often buy interruptible power to get a better deal), so there's no need for any kind of backup. Solar power and aluminum smelting are a match made in heaven.

Funny, I read that as "somewhere where the environment can't reach you," which would be a fair critique of a robot supply chain on Earth. It's probably easier for robots to setup and maintain solar panels on the moon than to setup wind and solar arrays on earth, where they have to worry about wind loads, mud, corrosion, baby deer running into them, etc. But I don't know if that counterbalances the difficulty of getting started on the moon.

I left ethics as a leading question, because I think it's actually an interesting one. We tend to worry when people start messing up places where there's life or where people might see the view. Can we ignore those concerns in the Sahara Desert? On the moon? In the asteroid belt?

About half of the Earth's land is virtually uninhabited, which means nearly free land; and most of that land has good access to "free" energy (wind and solar power). So why would we have to go to the moon to setup an exponentially growing robot-run supply-chain? Is it ethically better to make rocket fuel and metals on the moon than in Antarctica or the Sahara Desert or northern Canada?

You are basically arguing that human-written stories of this ilk are baseless and AI-written stories will be no better. I agree, but I would say an AI system can probably do this job as well as a person, and for less money. So an AI system could satisfy people who want to read stories about "momentum" or unusual performance, even though those are often just apparent patterns in the noise. You are not in that group of readers and neither am I, but that doesn't mean the AI is ill-suited to this job. And for minor-league teams the job won't get done at all without an AI, because the market is too small to pay real people to do it.

Who said anything about a sample size of one game? AI can "remember" the details of every game, and identify how this game is different from others. Think of it as an "anomaly reporter", which is a lot of what people want from a sports report. That's well within the abilities of an AI. Even momentum is discernible by AI, arguably better than a human. If one team had unusually good performance during a certain span of time, that's "momentum". Sure, humans can spot patterns like this (or sometimes false patterns), but that doesn't mean a computer can't. And are you willing to pay enough for minor-league sports stories to hire the army of human reporters that would be needed to follow every team and judge their performance compared to prior games?

I wouldn't expect to be uplifted by the soaring prose of a minor-league sports report, unless they've singled out an exceptional minor-league game, which is a different goal from this project. For run-of-the-mill sports reporting, doesn't most of the "art" consist of identifying important or unusual details of the game -- which play changed the momentum from one team to another, which player(s) performed better or got more playtime than they usually do, how well or poorly the great player(s) performed? That's all legitimate grounds for AI, and an AI story that reported these details would save you a lot of time vs. reading the play-by-play report yourself.

There might also be strange coincidences, like a bunch of universal constants that just happen to create a stable universe instead of one that collapses or explodes before anything interesting happens. Or seemingly arbitrary rules such as a constant speed of propagation for light.

Part of the problem is that "content providers" want to reserve the option to charge _less_ in poorer markets. For example, a lot of U.S. textbooks are sold in China and India at lower prices, albeit on cheaper paper and maybe with a Chinese cover. There is no legal obstacle to importing these books back to the U.S. and selling them at a lower price than U.S. textbooks. So with DVDs, they introduced a technical obstacle: preventing DVDs sold in one region from being played in players made for another region.

With streaming it may be the same thing -- don't allow a streaming account based in one region to be used in another region. I doubt Netflix offers lower-cost streaming accounts in India, but by now I the video producers have the bug for regional segregation.

Is there anything inherent to these designs that prevents rogue owners from inserting U-238 and producing Pu-239? I haven't been able to find any discussion of this except a note about a French MSR design that uses 50 kg of Thorium-232 and 50 kg of Uranium-238 (not a good sign). If the whole world is going to use thorium breeder reactors, we don't want everyone to be able to kick out the inspectors and switch to bomb-making mode whenever they feel like it.

You are ignoring the proliferation issue -- fast breeders convert uranium to plutonium, so anyone who has one is one big step closer to making nuclear weapons. Breeder reactors cannot be the answer for most of the whole world, so they are not the answer.

Breeder reactors by design can convert unenriched uranium into plutonium, suitable for weapons. Right now, 14 countries own or have access to nuclear weapons. Would you want to limit breeder reactors to them? If so, then what should the rest of the world use for energy?

Breeder reactors are not the answer if they can only be used by nuclear powers, and they're not the answer if they make everyone a nuclear power. There's not much room in between.

I think you're trying to say that mirrors can heat salt with >75% efficiency and heat engines tend to be >90% efficient. The first claim seems vaguely plausible, but the second claim is certainly false. Typical single-cycle power plants have efficiencies around 35%, and combined cycle plants (combustion turbine plus heat-recovery steam generator) have efficiencies up to about 50%.

With a heat engine operating at normal terrestrial temperatures (say 300 K on the cold side and 1000 K on the hot side), the maximum possible efficiency is 70%. To achieve 90% efficiency, the high temperature side would have to be at least 3000 K (half as hot as the sun). I doubt this system is that hot.

I know what logarithmic means, but I was countering the conclusion drawn by huckamania:

If we have seen 0.85 degrees of warming and can expect 3.2 degrees, then clearly it is not true that "most of the heating we should expect to see has already happened." The reports I cited factor in the logarithmic nature of warming, but also account for the fact that it takes a long time to reach the final temperatures after CO2 concentrations are increased. The delayed warming exceeds the logarithmic effect, and we can expect much more warming in the future than we've seen already.

That is precisely why I am not arranging the links in the pretty little table you want. I have given sound evidence for my position, and don't care to jump through hoops putting them into the specific format you requested. Grow up and read the articles.

Anyone who has made the least effort to study the Earth's climate knows that (a) we are dumping a lot of CO2 into the atmosphere, (b) higher CO2 concentrations will cause the planet to warm, and (c) significant warming could cause serious harm. My "if" condition is satisfied: there is substantial evidence that people are causing climate change and that climate change could cause significant harm.

At this point, the burden of proof is on you. If you think we should do nothing about climate change, it's up to you to provide evidence to support your position, which you have not done and cannot do. This obligation is especially strong given the serious risks involved. If our ship is headed toward an iceberg, we must steer away, even if it is inconvenient. If you want the ship to keep sailing ahead instead, it's up to you to prove the iceberg won't damage the ship. As you say, put up or shut up.

Every one of the links in this thread points to an easy-to-read article referencing a mainstream prediction of temperature trends and later evidence showing that the prediction was right. If you have so little interest that you won't read the articles unless the links are spoon fed to you in a particular format, then there's not much point discussing this with you.

The original predictions cited in this thread were published in Science magazine and several IPCC Assessment Reports. Those sources are at the pinnacle of global peer review and debate on climate change: they represent the mainstream view among climate scientists at the time they were published. They are not cherry-picked studies that happened to turn out right.

If you think it would be easy to retrospectively cherry-pick old studies that happened to be right (or wrong), you will enjoy my challenge to you: please give me one reference to a global temperature forecast published in a peer-reviewed source in the past 35 years that significantly overpredicted global warming. You can refer to the article and its rebuttal using an academic citation, a pretty table of URLs, a fun word game, or any other format you choose.

I would also point out that this debate doesn't even center on the right question, which is: given the evidence we have, what action should we take? If there is substantial evidence that people are causing climate change and that climate change could cause significant harm, then the correct policy is to take action to avoid the harm. The only sound argument for inaction would be compelling evidence that harm will not occur. Do you have that evidence? It's not enough to debate the precise magnitude of the risk. You have to show that on balance inaction can be expected to cause less harm than action. Anything else amounts to hoping the problem will not materialize, and as we know, "hope is not a plan."