"Kirsch says to stop treating reading as civic medicine. "It would be better to describe reading not as a public duty but as a private pleasure, sometimes even a vice,""
Well, I guess I'd start with telling him to stop trying to "trick" people - even kids - into doing what he wants?

It's a routine fault of progressives AND evangelical conservatives: this inherent sense of moral certainty, and the instinctive justification that "pretty much anything goes because I'm doing it to HELP you".

Reading (or more specifically, the desire to read, as there are tons of people with impairments that get in the way of literally reading a book) I'd say is symptomatic of intelligence. So what we really want are people who value intelligence, who value reason. Reading will more or less automatically follow.

I'll be honest, I don't think 'reading' alone is inherently magical. Reading the sports page, or some fantasy smut about milking male minotaurs - they may both be enjoyable, but neither is going to make someone the kind of constructive, reasoning citizen we NEED in our Republic.
Well, that any democracy needs, not just ours.

Of course, then we get back to the 'certainty'. Reasoning adults need to be able to hold in their heads a fundamental RESPECT for the other person's ideas. Even if they don't agree.

So here's the funny bit for me. Kirsch suggests that we trick kids into thinking reading is scandalous, a vice. Is that really what he wants? What if they read actually-scandalous texts (according to Kirsch's orthodoxy) like something by Charlie Kirk? The Art of the Deal? Would he be as intrinsically delighted with "people reading" then?

That makes apparently 1 person in earth that believes the BBC is "too slanted to the right" lol .
Then again, as we've seen over the past 5 years, to the new rage-addled Left Marx himself would have come off as far too right wing.

Iâ(TM)m curious whether it will be viable to extract HBM. BGA packages are absolutely doable; but HBM seems potentially more challenging.

Which AI?

I asked Claude, and it gave me a much more reasonable figure (200g of bacteria in a 70kg person), and provided a link to the relevant research.

No disagreement here. Asus is a totally logical party to be raising retail prices because component prices are going up. I was specifically responding to the claim that 'they' would find 'some other convenient excuse to raise prices, just because they can'; for which Asus seems like a much poorer candidate.

As a basically commodity vendor I'd assume that Asus is very poorly suited to try to mask component prices, because their margins just aren't all that exciting; but as a basically commodity vendor they are in an equally poor position to 'because we can' any price increase that their competitors don't have basically the same reason for also making; since you can just buy from someone else unless ROG RGB LED compatibility is just that critical to you.

It will be interesting to see how that shakes out. Given the downright absurd debt levels and negative margins it seems unlikely that this gear will avoid the fire sale forever; but it's a lot more specialized compared to what came out of .com bust era equivalents. Even 'normal' GPU compute servers are aimed at relatively specific purposes(and all but the low end ones that are just ~8 PCIe GPUs in a row are even fairly topologically quirky, bunch of nvlinked GPUs and a PCIe topology that depends on GPUs RDMA-ing directly to ethernet or infiniband NICs without keeping the CPU in the loop); and the 'rack scale' or hyperscaler custom stuff is quite likely to be a liquid cooled monster with custom dimensions and atypical rack weights, 'modular' mostly for service purposes, not just pets.com's xeon pizzaboxes that can be unracked and auctioned off in an afternoon.

I don't doubt that people will come up with something; if the discount is steep enough all sorts of retrofits or technically-inefficient uses become viable; and there will presumably be some post-crash inventory in things like HBM and GDDR that was contractually locked up but not actually soldered down yet that will be easier to find a new home for; but it will be a lot more involved than the mostly fairly conventional stuff from the .com bubble period.

Asus seems like a pretty weak candidate for just pulling off a price increase for fabricated reasons. They aren't a small shop; but I can't think of a single product of theirs that doesn't have a fairly ready substitute from another PC OEM or component manufacturer. It's possible that some of their stuff has a reputation that could support at least a modest price premium, I definitely don't know what their reputation is across all areas; but there are more or less direct cross-vendor substitutes that they would need to worry about if they tried to bump prices.

Or, crazy idea I know, maybe wealthy people aren't universally - or even mostly - stuck up self obsessed narcissists, and that's just your own seething envy and sense of self justification?

If I had to choose between Walmart vs a store that literally filters out the poorest at the door, I know where I'd rather work or shop.

At a magic show, everything you saw was real. I mean, it had to be, it actually happened. But what you thought you saw happen isn't what you actually saw, it's how your brain interpreted what you saw. Eg., you saw the magician slide the cup far enough off the back of the table so the ball under it dropped out into the basket hidden back there, but you didn't notice because it happened so fast and he'd directed your attention somewhere else. But the guy who works eye-in-the-sky for a casino, he noticed because he's used to catching things that happen quickly and he's always on guard for subjects trying to direct attention away from what they're doing that he needs to be paying attention to.

I'd put it another way: the "you have to pay to even GET IN THE DOOR" keeps out the rifraff.

"...describes the stores as spaces of "cooperation, courtesy, and grown-ups mostly acting like grown-ups." Shoppers follow unwritten rules: move along, don't block the way, step aside to check your phone. Checkout lines form orderly queues. ..."

Note the entire article is about how civilized an experience it is, and how "weird" that is; that Atlantic tries to paint it as "cult like" shows how utterly fucked up our Intelligentsia has become.

And you could have achieved exactly the same thing at lower compute cost with a stateful firewall. NAT didn't save you from worms, the stateful firewall that NAT requires in order to work did. But you can have the firewall without the NAT, and the result is simpler, more efficient, easier to manage and more flexible.

So have your router run a firewall that denies inbound connections be default, the same way NAT does. This is a side effect of NAT, but can be done better and more easily by a simple firewall.

Then don't run a web server in your bedroom. And maybe have a firewall that blocks inbound connections by default (which is a side effect of NAT, but absolutely does not require NAT).

But many of us would like to run servers from home.

Silver and gold aren't like fiat, they're inherently inferior to fiat currency, for exactly the reason you cite.

PPH is deeply incorrect when he says that fiat currencies are not backed by assets. Fiat currencies are backed by legally-enforceable promises to repay debts. Every dollar created is balanced by the simultaneous creation of an obligation by someone to do some sort of productive work to generate value that is used to retire that obligation (at which point the dollar is destroyed).

Precious metals have a small utility value coupled with a large speculative value. Fiat currency is also a mixture of utility and speculation, but the mixture depends on the probability that the borrower on the loan contract that created the dollar will repay the debt. Since nearly all debt is repaid, dollars are mostly real utility. Further, the probability of default is offset by the fact that borrowers that don't default repay more than 100% of the debt, because of interest (though interest must also offset the discount rate, i.e. currency devaluation, i.e. inflation). So the precise mixture of utility and speculation behind a dollar is hard to nail down with precision, but the speculative part is very small, usually indistinguishable from zero.

Fiat currency evolved rather than being designed, but it's hard to see how a more perfect system could have been designed. Not only does fiat currency have greater intrinsic value than precious metals or similar physical exchange media, it naturally expands and contracts the money supply as needed, which facilitates rapid economic growth in good times and prevents devastating deflation in bad times.

Cryptocurrencies are among the worst possible forms of currency. They have negative intrinsic value (they cost money to produce but have no intrinsic utility), have very limited ability to expand money supply when needed (they could adjust the mining success probability, but there are strong disincentives to do so) and cannot contract the money supply when needed. Further, they have very high transaction costs, which leads to them being treated as assets rather than currencies. The fact that they're very hard to regulate does, however, make them great for crime of all sorts.

No, fiat currency is asset-backed. The assets in question are legally-enforceable contracts to repay loans. Fiat currency is precisely a legally-enforced commitment to produce something of value in the future, bundled up into an exchangeable medium.