I am trying to figure out how your link possibly supports your assertions. Frankly, your source makes me feel a little dumber for having read it, mainly for the information technology for grade schoolers primer on what a data center is at the start. What I can glean from it is that it seems to be projecting that power usage from data centers will grow by 2.32X by 2030 (linearly, I'll note) So, if it was ~21GW in 2024, then it will grow by 4.62 GW per year. Now, the 11.7 GW noted is just for a single quarter of the year, so that would actually work out to more like 46 GW per year. However, the capacity factor for that is supposed to be around 24%. So, that would be something like a total of 11 GW added per year. 11 GW is clearly more than 4.62 GW, so your own source disagrees with your claim that AI data centers are "on a trajectory to absorb way more than those 11.7GW".

That's even if we assume the four fundamental erroneous assumptions you seem to be taking for granted: First, that somehow solar power is the only renewable power source being added and so it has to stand alone compared to AI power growth; Second, that somehow AI power growth is alone in power demand and that there can't be other factors increasing or, indeed, decreasing power demand; Third that the increase in AI power demand will continue to increase as in the speculative projection in the article you link because, let's face it, LLMs are just the latest bubble; Fourth, that solar (and other renewable) power growth will somehow be flat rather than increasing.

Overall, what you are claiming does not seem to hold up to real world facts or, indeed, your own cited source.

OK. I remember back when a "solar powered flashlight" was a joke too. Back in my childhood. These days though, I own several. I can't tell if you're actually trying to be funny or not, but the simple fact is that the energy required to produce a given amount of light has dropped by about an order of magnitude and the solar cells have become more efficient, and the batteries have become lighter, cheaper, and rechargeable. If I step outside at night I can literally see several gardens lit up by mass produced garden stake lights that charge up during the day and then work through half the night. Those are all cheaply bought at garden centers at the like. Basically, in this day and age, it's laughable to suggest that solar power can''t keep lights on at night because it is easily contradicted by the reality all around us.

It's a question of dying before producing offspring. The article shows that the majority will reach the average age of parenthood in society. That's all it takes.

Heh. My daughter and I were having a conversation about that just last night. She is a mostly a big believer that there is no "great filter" per se. The answer for her to the question of why intelligent aliens have not reached us yet is that intelligent aliens have not reached us yet. Still she had to concede that, if intelligent aliens were simply holding out on reaching out to us, one probable reason, could pretty well be summed up in a specific example. That example was a specific politician whose identity will be left as an exercise to the reader (note that, if you come up with an example that is not the one she was thinking of, it is still probably a correct example, since really it's about how much the politics of our planet suck and what a poor reflection that is on humans as a species).

One interesting thing about it as well is that, even if the question of whether or not there is other life out there is actually truly no, there's a funny thing that happens if the question of "will we ever work out interstellar travel" turns out to be yes. Basically, if we manage to inhabit worlds beyond our solar system and it's a repeatable event, then eventually we get life all over the place and the universe still has billions of years left at least for that life to evolve into all sorts of things. So, basically panspermia.

I mean, one of the houses I grew up in we heated our living room with a fireplace, so... I think the actual concentration may matter some.

Yeah, but that's UV radiation. Most transparent materials are opaque to UVC, the most dangerous UV light. There are even plenty of transparent materials (such as typical window glass) that are opaque to UVB. So that's the most dangerous UV light blocked possibly without even any special filter materials. UVA can be blocked by some materials that are fully transparent to ordinary visible light. Otherwise, there are plenty of coatings that are still largely transparent that can block it. So, you can block basically all the harmful UV light with a roof that is otherwise fully transparent. Not that you necessarily even need your habitat to have a transparent roof in the human habitable areas, but that might be nice for various public areas or even atria in people's private quarters.

In any case, at the approximate 50 km altitude proposed, the total amount of solar irradiance is estimated to be similar to Earth. If it's higher to any significant degree, then habitat areas with transparent roofs just need more light filtering (possibly with transparent solar panels). As it is, they will probably need some sort of shuttering system such as an LCD system or similar to simulate night and maybe seasons for the comfort of the residents since the proposal is to keep the habitat always in sunlight. Too much sunlight can make people overstimulated and manic. For agricultural areas, obviously the light cycle and allowed UV levels would depend on the needs of the particular plants growing in any bay. For solar panels, you would expect coatings that protect against particular UV ranges that might damage them. Possibly the coatings could be fluorescent to convert UV to light usable by the panels, or the panels might be intrinsically fully UV hardy. The materials of the outside of the aerostat habitats would need to be UV hardy or coated with a UV protection layer and they would similarly need acid protection.

Ultimately, EM radiation either UV or visible seems like it would be a non-issue with some basic design precautions. So it is a consideration, but an easily solved one. The same applies for other forms of radiation from the sun. This would be at an altitude where the protection of the atmosphere above provides almost exactly the same protection as that of Earth. On Earth, at sea level, radiation from the sun and cosmic radiation pose basically zero threat (the threat from UV is many, many times greater) barring a massive burst from a supernova. On Venus, at that altitude, the radiation threat would be no greater than being at a high altitude city on Earth.

I just had to remind someone else of this: sarcasm tags are now mandatory on all sarcasm.

Not really. Despite the heat of re-entry, meteorites are generally cold when they hit the ground since they tend to keep cool through ablation and/or sheer thermal mass and the square/cube ratio. Sure, a lot of meteors are going to be sterilized as fiery bolides, but if you have a lump of rock left over after re-entry and it had life in it, it is highly likely to survive.

Right, but the thing about what you're saying is that it means that the $20+ million isn't a real operating cost for the craft. Canceling it therefore does not mean that you really save that money.

So you're saying that the cost of research on the data sent back by MAVEN is misleadingly folded into the "operating costs"?

It is worth noting though that Iridium satellites are LEO satellites. That means a relatively short lifespan compared to the Voyager satellites, not to mention a heck of a lot more that needs to be done to make sure that they don't collide with things, etc.

Yeah, I wondered that myself. Aside from some station keeping, what would/could anyone on the ground actually do with it any more? I mean, maybe there's some storage or lab space where they keep copies of the technology in it as reference materials if anything goes wrong so that they can brainstorm solutions. Maybe they also keep some experts on retainer or something who were involved in it for the same purpose. It is still hard to imagine that could possibly cost over $20 million a year. It seems like numbers like that are often the result of not really understanding the accounting. For example, the $20+ million per year is part of the original project cost amortized over X number of years. So cancelling it before X number of years might seem like money is being saved, except that it's actually money that's already spent. Or the money is part of some larger project and that is its share in the larger project, so cancelling it just means that the rest of the project costs more, etc.

So yeah, that price tag seems like DOGE math to me.

That's a tricky question of etiology. Technically, cancer is normally caused by a an inherited or in vivo genetic mutation which leads to cells multiplying without constraint. That may be a simplification, but it really does seem like it's a pretty valid claim to say that cancer is pretty much always from a cancer-causing gene.

Except of course for all the future generations those children might have gone on to produce without anyone being the wiser. Also that, statistically, the screening done on sperm donors, while it missed this, will produce children with fewer genetic defects on average.