Indeed. And, yes, Claude massively over-comments. I have more Claude coding rules about commenting than any other single topic. Though I do wonder if my rules make as much sense in the AI era as when code was all maintained by humans. Most of my rules are about minimizing comments because comments are fragile and tend to get out of date... but Claude actually does do a pretty good job of maintaining the comments. I still try to minimize them, though.

Heh. That's definitely true for me as well, now, not so much in the past. When I was at Google I had a higher caliber of colleagues. My colleagues at the new company are bright, but they're young and inexperienced. But, yeah, if I didn't have Claude to kick ideas around with me in my current position, my code would be much worse than it is.

Nope. Wars used to regularly cause widespread famines, as well as being far more directly bloody. Murdering all the children was for millennia an accepted practice. You should read the book.

Yes, it is

When the norms and the expectations move from considering something cruelty to be funny or enjoyable to merely accepted and then to shameful or hidden -- and even illegal -- those steps are progress.

Related: "Hypocrisy is the homage that vice pays to virtue." While the individual hypocrite may not be better than the person who engages in open vice, and might be worse, the fact that people feel the need to keep their vices secret is a positive indication about society as a whole. Assuming, of course, that the "vice" is actually bad.

That's another area where our society has been improved... we're more tolerant, having realized that many things we considered bad are merely different. We still have progress to make on that front, too, but don't let perfect be the enemy of progress.

That's probably true of many, and it couples well with motivated-disbelief. Confidence that if it happens it won't really be a problem for you makes it easier to just shrug the whole thing off and refuse to think too much about whether your disbelief makes sense.

It's worth mentioning that there's one more position that actually does make sense, even if it's a bit Pollyanna-ish: The belief that science will find a less impactful way to address the problem in the future. The argument is that we shouldn't trouble ourselves now, we should just wait for the new tech that will fix it.

I actually subscribe to a weak form of this view. I think we should be acting now to address climate change, but that we shouldn't do anything too drastic, because technology is going to improve and find better solutions. The world is actually making significant strides toward emissions reduction, mostly in the form of low-emissions electricity production, and not because of a moral obligation but because renewables are cheap! That's the sort of thing that generates real progress, without much pain.

I suspect that atmospheric carbon recapture will always be extremely energy-intensive, but we are on a path to extreme but intermittent energy abundance, and carbon recapture sounds like a great way to spend the extra terawatts when they're available. I think one of the things we're not doing enough of now is research into carbon recapture and sequestration. Reducing emissions can never get us to net-negative CO2, and we need that if we want to actually fix this problem in anything less than a millennium, so cutting emissions is insufficient. The corollary to that is that cutting emissions will likely become unnecessary before we get very close to zero.

So, the conclusion of the weak-form of this techno-optimisim is that we should be working to curb emissions, and we should be directing tax dollars to recapture and sequestration research (and geoengineering, too), but we shouldn't go so far that we reduce economic output.

What I'd really like to see us do is to take a very market-driven approach, facilitated by carbon taxes. Pick a reasonable per-ton price and apply it at the point of extraction, where it's easy to identify and track, so that every downstream use has the carbon tax built in. Fossil fuel consumption that doesn't burn it and release the CO2 can apply for a rebate to recover the carbon taxes on the carbon they didn't emit. Couple that with carbon tariffs which attempt to impute to foreign-made goods the CO2 emitted in their production. Anyone who can prove they're capturing and permanently sequestering tons of carbon should be able to capture that as a refundable tax credit. Planting trees should count, as long as there's a plan to keep that carbon sequestered for several hundred years -- and if the trees burn, the tax liability comes back. Oh, and a small percentage of the tax revenue should be earmarked for climate mitigation research. The rest can just go into the general fund, ideally displacing other taxes, and maybe funding progressive offsets since a carbon tax would be mildly regressive.

There'd be a fair amount of bureaucracy in defining and administering such a tax, especially the tariff part. But it's manageable, I think, in particular because it doesn't have to be perfect, it only has to be good enough that everyone is incentivized to avoid 1-2% of their emissions this year, another 1-2% next year, and so on, and good enough that there's actual money to be made in recapture and sequestration for anyone who can figure it out. With that, we can sit back and let the market solve the sort of problem it's good at solving. We might need to tune the carbon tax rates a bit from time to time, and we'll want to scrutinize the system regularly to identify loopholes to close, but mostly we could just consider it a solution in progress.

And it's also worth remembering that we wage far less war than ever before, and engage in far less of the rest as well. Stephen Pinker's "The Better Angels of Our Nature" documents this very well and I highly recommend it.

Just consider one example: Animal cruelty. Of course some people are still quite cruel to animals, but they're the exception, and this was not historically the case. For example there are historical accounts of a common festival entertainment in medieval France, where cats were put in sacks or baskets or hung from poles and burned alive so their yowling could amuse crowds of festival-goers. Bear-baiting, bull-baiting and cockfighting were other examples. These weren't underground, deviant activities, they were public, family events that whole communities anticipated and attended with great enjoyment.

We're far from perfect, but we're getting better, and not just technologically.

I think that imputes too much evil to them, and not enough stupidity. Apply Hanlon's Razor "Never attribute to malice that which is adequately explained by stupidity".

No, I think the truth is that they think the catastrophe will be good, or they don't believe it. Or both! Humans are very good at cognitive dissonance.

Among the more deeply religious on the right, I think many may accept that climate change is real, but assume that the end will come first -- or even that climate change is the predicted Armageddon that will cleanse the Earth by fire and be associated with the Second Coming of Christ. Among those with the latter view are some would actually like to accelerate the process, because Second Coming is a good thing for the faithful; it ends their struggles and sends them to a life of eternal glory.

Among those who aren't so religious, I think it's mostly motivated reasoning. Having to make large changes in order to prevent a catastrophe is unpleasant, and to some extent it seems crazy to think that puny humans could seriously affect the planet, so it's easier to believe that the climate scientists are wrong, that either climate change isn't happening, or that it won't really be so bad, or that we can't do anything about it anyway. This creates the question of why climate scientists would continue being so vociferously wrong, which leads to a belief that they must be maliciously wrong (the right also needs to apply Hanlon's Razor).

I seriously doubt that anyone is doing a Mr. Burns, rubbing their hands in glee while planning a catastrophe they think they can come out on top of. First, because anyone in a position to do that is already on top, and second because being top dog in a hellish dystopia isn't really a big win.

There are good reasons not to have the government "pick winners", but this isn't a good example. This is a historically-incompetent and utterly-corrupt administration picking winners, and that's guaranteed to go badly. What actually works reasonably well is government grants to the National Science Foundation, and letting the NSF evaluate grant proposals and dole out the money on scientific merit.

There's your problem: Javascript.

I'm actually serious. LLMs in their current incarnation need a lot of guardrails, and I think they do much better with a very strict, statically-typed language.

If I did have to write Javascript and I wanted to use an LLM I would focus hardcore on extremely thorough and extensive unit tests, because that's the only way to provide the necessary guardrails. And I would closely scrutinize all test changes made... or maybe just mark the test code as read-only so the AI can't modify the tests.

Also you should be very picky about architecture, and have the LLM implement in smallish chunks, thoroughly vetting its output each time.

Is this a lot of work? Yes! It's a lot less work than doing it all yourself, but if you just give the AI a high-level problem description and turn it loose, you're probably going to end up with crap. Even worse in a language like JS that encourages crap anyway.

It's high. I have a team of three reviewers, and I think their reviews are kind of thin. They do point out useful improvements, but I think more careful review could find more. That said, I also feel like the overall quality is actually higher.

I've also been coding professionally for a little over 35 years, and AI is a complete game-changer. It's going to take us a while to figure out just how much. I actually wonder if my focus on code quality is pointless. I put a lot of effort into ensuring that code is clean and readable for humans, but will that really matter? My current project is the realization of something I've been thinking about for 15 years... how to build a crypto API that guides users to use it correctly and safely. But will that actually matter, or will users just point an LLM at it that has more knowledge about what's safe and hygienic than I do?

Mostly I don't think about it. I'm building something I've wanted to build for a long time, doing it incredibly quickly, and having a great time at it. I'm going to do this for another 2-3 years, then retire. I expect to leave behind something that I'd have been immensely proud of a decade ago... an elegant design with a very high quality implementation.

But I'm not sure that it will be better in practice than something quick and dirty, because I'm not sure people are going to be writing code at all in a few years, and I don't know if the readability and maintainability characteristics I'm so careful about will even matter to its users.

Fair. In a competitive market those are the same things, but monopoly situations -- especially government-enforced monopolies, like patented drugs -- create curves that behave differently.

To the degree that's true, it's nothing inherent in the processor generations. What's happened is that Moore's Law has slowed dramatically. If there were as much performance difference between a 2026 CPU and a 2020 CPU as there was between a 2006 CPU and a 2000 CPU, we'd still be feeling the need to upgrade regularly.

Consider, for example that between 2000 and 2006, clock speeds tripled, CPUs went from single to dual core, and instructions per clock went up. A 2006 flagship CPU was ~20X faster than a 2000 flagship CPU. From 2020 to 2026 we saw, what, 2X? And most of that gain came from increasing core counts and hybrid big/little cores, which means that most workloads don't realize the full benefit.

When machines are getting an order of magnitude faster every five years, you're going to be upgrading frequently. That hasn't been happening for a while. If we have some major shift in CPU tech that give us 10X faster machines by 2030, the upgrade treadmill will resume.

That should have been km^3, obviously.

Do the math! The Earth is ~12,800 km in diameter, so a LEO shell from 350-450 km up is about 13,300 km in diameter and about 100 km thick. That gives a volume of about 56 billion km^2. If we give every satellite 100 km^3, that means we're limited to about 56 million satellites.

Oh, you want geostationary orbit? That's way, way, way bigger (though possibly subject to Kessler syndrome, unlike LEO).

Basically every space agency does this.

This is a potential concern for high orbits, not so much for where most of the satellites are being deployed.

Space getting "full" is never going to be a constraint on SpaceX's growth, so you should probably look for some different signals.

I absolutely review all of the code, telling the AI to rewrite parts of it, and occasionally doing it myself. I take advantage of the AI to produce not only more code, but higher-quality code (because I will make the AI do refactors that I'd previously have dismissed as not worth the effort). I now get more done in a day than I used to do in a week and, as I said, with higher quality: more/better documentation, cleaner code, more comprehensive test suites, etc.

AI is a huge productivity boost, and it's actually that boost that creates the review and merge bottlenecks. A four-hour merge process isn't a problem when you only produce two merge-ready PRs per week. But I average one merge-ready PR every 2-3 hours.