That's not necessarily enough to implicate "poisons". All sorts of conditions were rare or functionally unheard of until Industrial Revolution periods, because our lives were rather closer to the conditions they evolved for (large amounts of time spent outdoors with a different mix of pathogens than you get in a fairly sterile indoor environment) until that point. Allergies, various autoimmune issues, asthma, and even nearsightedness were rare, less severe, or non-existent in the before times.

For the first three, the working theory ("hygiene hypothesis") is that, because our immune systems evolved to expect a certain level, and certain type, of disease in the environment, to calibrate the body's idea of what constitutes a dangerous pathogen, when we moved from "exposure to lots of digestive system pathogens and parasites" to "almost none of that, just lots of respiratory ailments that spread easily indoors" the immune system went haywire and either identified harmless things as harmful, or failed to even properly identify what constituted "self" vs. "other".

For nearsightedness, that's always existed, but it's gotten worse. Some of that is likely genetic; we developed cheap enough glasses and it became less important to have good eyes to pass on your genes, so the selective pressure to preserve genes for good vision largely vanished. But it's also because our eyes calibrate as children by looking at distant things in good lighting (that is, outdoors, in sunlight). When we moved indoors and stopped working and playing outside as much, the lighting got worse (you may think you use a lot of lights, but that's your eyes fooling you by adjusting to make indoors seem brighter; it's a tiny fraction of what natural sunlight provides), and we mostly looked at nearby things, and whoops, myopia becomes more common and more severe.

And of course, many disorders, like Parkinson's, are largely diseases of old age (typical age of onset for Parkinson's is 60), so in a world with lower life expectancy (yes, a lot of it was child mortality, but some of it was shorter lifespans even for those who survived childhood), the percentage of people who live long enough for Parkinson's to be a meaningful risk would be lower.

To be clear, I'm not arguing that Parkinson's isn't possibly triggered by an environmental toxin. Just that "it was only described recently" as evidence for "it must be caused by a brand new toxin" is overlooking the many, many changes we've made to the environment we live in over the past 200-odd years that cause all sorts of disorders in ways we still don't definitively understand the causation behind, or only very recently found explanations for.

Compressed elliptic curve points is not some crazy secret or complex thing to be making up conspiracy theories about. The nature of elliptic curve cryptography is that knowing the curve and the x coordinate, there are literally only two possible y coordinates, and a single bit signals which of the two possible y coordinates to use, allowing you to reduce, say, 512 bits of key coordinate down to 257 (256 for the x, 1 for the y), at a trivial cost to compute y from x before doing the rest of the math.

The only reason this wasn't done automatically in every implementation of elliptic curve cryptography from the beginning is because some idiot at the patent office issued a patent for point compression (that is, one specific case of selecting one of two options with a single bit), and so for a couple of decades free projects didn't want to risk implementing it for fear of patent trolls. The patents have now expired, so people are doing the sensible thing.

The problem is that the people most harmed by this aren't the morons themselves, but their kids, who have no say in the matter. Sure, in an evolutionary sense, eliminating the children does curtail their influence on the gene pool (depending on how much of their idiocy is heritable), but I'd really prefer it if they took themselves out of the gene pool and let their kids learn from that example.

checks current RAM usage Almost precisely 1 GB from all Firefox tabs put together (seven loaded and an embarrassingly large number of unloaded tabs I'll probably ignore for a few years before cleaning them up by the thousand). uBlock Origin and uMatrix probably reducing the load a bit by blocking all the ads. Chrome is no better at this (and worse at the ad blocking), I really don't get why people are so down on Firefox memory usage.

Umm, carbon dioxide does in fact damage health directly. No, it's not doing anything to us right now outdoors, but indoor levels of CO2, especially with large numbers of people in a poorly ventilated area are substantially higher than outdoor levels. A higher baseline outdoor CO2 makes those indoor levels rise even higher. Health effects begin at around 1000 ppm, which we regularly hit already indoors, and which we could hit outdoors in urban areas by the end of the century. It begins with reduced higher-level cognitive function while exposed to the higher levels (read: we get dumber and dumber as levels rise), and chronic exposure weakens bones, forms kidney stones, and damages the circulatory system (more cardiac arrest, more strokes, etc.). It's why solar radiation management isn't sufficient to solve the problem; if we cool the planet without reducing CO2 levels themselves, we'll all get dumber and sicker no matter how much we do to restore pre-industrial temperatures.

That's only true since 2017. Prior to then, lots of folks itemized. Post-2017, with itemized deductions for regular people limited to the charitable deduction, the (capped) mortgage interest deduction, and the (heavily capped, to punish blue states, and not adjusted for inflation so it's less useful each year) SALT deduction, yeah, basically only singles with new mortgages or those who earn and donate a lot of money get anything from itemizing, but that isn't a permanent and automatic state of affairs. It could happen next year if the One Big Beautiful Bill fails, because all the changes to deductions and the standard deduction were officially temporary, and failing to extend them means the old deduction regime returns.

At the same time he was talking about "undoing some mistakes", he was also talking about gutting the regulations, and regulatory agencies, that fixed those mistakes of the past. A phrase that rings true is "Every regulation is written in blood." No, it's not 100% true, but it's the case for many, many regs. The regulations on food safety, water quality, removing lead from the environment, etc., are all there because people died. When you say "I want to cut regulations (in general)" as opposed to "I oppose these specific regulations as being poorly targeted and in need of improvement", you're saying "It's okay if we kill some people if it saves money." And that's exactly what we're getting.

Capital-L Libertarians lately haven't been all that distinct from Republicans. Sure, they tend to be more marijuana friendly. And they claim to want to cut spending or balance the budget, but so does almost every Republican, and 95% of them, given the right environment, e.g. single party control of all branches of gov't, immediately unbalance the budget further, barely cutting spending, if at all, and handing out even more unsustainable tax cuts. And you'd think, given the whole "personal autonomy as guiding principle" thing in little-l libertarianism, they'd be pro-choice, or at least "we may dislike it personally, but the gov't has no business getting involved". But basically every prominent Libertarian politician and thinker either is vocally anti-choice, or, even if they claim to be pro-choice, glaringly avoids criticizing massive steps towards anti-choice policies at both federal and state level.

I'll take the Libertarian party more seriously the moment they clearly decide that bodily autonomy, the single most basic human right, trumps (rather newly invented; within the last 100 years, most churches tended to go by quickening, not conception, as beginning of life) religious beliefs about non-conscious entities. For now, they're basically NRINO (Not Republican In Name Only).

Gaming via VPN sounds like a lag clusterfuck. You could do it, but for real time games, you'd be at a huge disadvantage.

Have you used it in the last, say, three years? They made massive improvements. I run with uBlock Origin, uMatrix (probably not for most folks, but I'm a paranoid OCD control freak), Greasemonkey with a dozen installed scripts, and Facebook Container, and it's lightning fast. They went through a bad spot 5-10 years ago, but for the last several years it's been as fast as Chrome for me, and unlike Chrome, doesn't constantly try to break ad-blocking extensions.

Except the house still won. The jackpot got this big from many rounds of play with no winner. They were eventually going to pay out, and now that they have, they've locked in the winnings from this round of suckers, perhaps slightly less than they'd have received otherwise, but still profitable for the lottery.

That's a weird way to spell "Humans murdered the shit out of megafauna more and more efficiently as they moved to areas where local megafauna didn't have time to evolve even behavioral defenses against human predation." Climate change moved roughly in sync across the globe, human arrival did not, and guess which one is more tightly correlated with mass extinctions selectively targeting megafauna?

Yeah, that's exactly it. A huge percentage of the kernel level exploits and bluescreens in Windows stemmed from insecure device drivers, because there are far more devices that need drivers than there are programmers capable of writing bulletproof kernel mode C code. The Windows solution was to make it much easier to write user-mode drivers, but they're still in C, so they're still buggy as hell, they just don't bluescreen the box when they go down (and as a bonus, they're easier to debug, which makes it somewhat easier to find and fix bugs when they do go down). I've written code for Windows myself (years ago, for Windows 7/WS2008), and the internal development rules at Microsoft do make it harder to make stupid mistakes with buffer overruns and the like, but it's still C, the moment you start indexing an array all it takes is one mistake to introduce a critical vulnerability.

I think the only reason you see Rust advocated for in this space so often is because, in the systems programming space, there just aren't a lot of options:

• C: Secure entirely by convention and discipline (read: All it takes is one lousy night's sleep and a less than perfect code review to introduce a serious bug).
• C++: Offers more options for opt-in safety, but Linus hates it (for more or less valid reasons), and again, it's opt-in safety, the gain is limited.
• Zig: Even newer than Rust, not stable enough for 1.0 release.
• Swift: Made by Apple, for Apple, don't want to anchor yourself to a language with a major corporation calling all the shots (sure, there open source contributors, no, Apple won't do a thing that hurts Apple products to improve the language for non-Apple product usage). Not sufficiently memory-safe either.
• Go: Finally, memory-safety! Except oops, garbage-collected; can't stop the world in the kernel, no sir.

Whatever you think of Rust, it's got the best memory-safety story of any reasonably mature, not-tied-to-a-single-corporate-overlord language here that gets roughly bare-metal performance and doesn't involve non-deterministic garbage collectors. That's really it. It has other nice-to-have features (e.g. decent standard library; built-in, reasonably well-done package management with a lively package ecosystem), but at kernel level, "memory-safe-by-default systems programming language" is the selling point. Adding interfaces to make it easy for multiple Rust-based device drivers to maintain that safety through a common interface rather than each of them individually wrapping the interface with FFI in a more-or-less safe way doesn't seem wildly unreasonable.

All forms of English are primarily short scale at this point, just like the U.S., even if they were historically long scale. The term "thousand million" still survives as an historical curiosity, but it's synonymous with billion in most usage nowadays. It's Continental Europe (excluding Russia/Turkey) that favors the long scale. Like British/American English, Russian favors short scale, luckily for Google (not that they'd pay either version of a decillion, in either dollars or rubles).