396 karma · joined October 16, 2022
That is not to say it's impossible that we, uniquely, possess the consciousness gene. Possession by demons is technically possible, too, and equally silly.
Good types do work for you, so that you don't need to write the code to do that work. Correctness is a side effect, because being wrong would take more work.
The overwhelming majority of problems do not fit OO, and using OOP on them makes a mess. Similarly, FP.
Continue reading
Subscribe today
with a 12-week introductory offerPolitically, it has been very useful to convince underclass voters that they are middle or even, often, upper class, and get them voting against measures that would benefit the underclass including themselves. Trump supporters helped institute policies that massively benefit the actual upper class, at their own expense.
Old code can be modernized incrementally, e.g. replacing uses of naked pointers with smart pointers and then deleting destructor code, or replacing C-style for-loops with modern range-style for-loops or calls to standard algorithms, passing a lambda. Many class declarations can be eliminated by using lambdas inline. A sweep throughout a big codebase just modernizing one kind of thing may take only a few days. This typically results in deleting a great deal of code, and all the bugs in the deleted code evaporate.
Deleting masses of code always feels great.
I am ready for the successor, Risc-6, already. It can retain most details of RISC-V while fixing its more glaring deficiencies. It is not a mistake to optimize an ISA for use in undergraduate courses, but probably is one to use identically the same ISA industrially.
I don't know if the worst deficiency is the need to fuse up to 5 consecutive instructions to get useful operations per clock cycle to a competitive level.
While fixing serious problems, numerous smaller incidentals could be patched, such as having chosen the architectural representation of "true" to be 1 instead of -1.
It was well known exactly when the Japanese planned to start the Pacific War, though intercepts did not say where. Many in the Navy were sure it would be at Pearl Harbor. Notably, the militarily useless battleships were there, but not the aircraft carriers.
The Soviets killed up to 15 million of their own before the war even started. (This number may be an overestimate because most locales had been required to overreport their own population over the past century, such that when an accurate census was conducted, the new number lacked both those killed and also the overreports.) How many million Ukrainians were deliberately starved before the war? 2 million? 4 million? (We may compare this to Churchill personally arranging to starve 3 million in and near what is now Bangladesh, around that time.)
Neither Germans nor Soviets saw any reason to bother feeding or sheltering their POWs. Soviet POWs were treated as deserters and sent to the GULag on their return, along with many thousands of Cossacks who had escaped well before the war, and were rounded up and sent back.
We need not doubt that there was a great deal of ineptitude in the Soviet officer corps, which had been mostly exterminated before the war and then repopulated according to purely political criteria. And, we need not doubt extreme disregard for the lives of conscripts, both Soviet and many, many others shanghaied from occupied territory.
None of this is to minimize the heroism of the people sent to fight against the Nazis, subject to attacks both from their own government and from a foreign invader. Fairly early in the war (1940?) the NKVD was ordered to abandon its quota of deserters to be shot from the rear for not advancing eagerly enough.
Negative leap seconds are even worse than regular leap seconds.
Fixing UTC just means not announcing any more leap seconds. Not doing a harmful thing anymore is always the best fix.
Google's 24-hour smear makes leap seconds, bad enough already, ever so much worse.
We could fix the problem just by never announcing another one.
If it happens, it will be much earlier than you hope, as millions, then tens of millions, then hundreds of millions force their way across borders, and fascist governments ride that into power and then start a global thermonuclear war. The refugees will only be fleeing crop failure and increasingly murderous kleptocratic government.
It is already starting. Lots of governments are flirting with dictatorship, and many have got there already. US and Brazil have drawn back a hairsbreadth from the brink, but for how long?
Irrational is contrary to efficient.
But of course bitcoin is technically negative-sum, because miners don't pay in, but do cash out. And, some fraction of coin is lost and so not cashed out. Whoever can't cash out donated their stake.
When it gets where it has been more or less constantly falling for long enough, and everybody wants out, it will fall at an increasing rate, and everybody still holding donated their stake, too.
You need a battery that tolerates many many cycles, and being charged at 100x its usual discharge rate. Maybe a molten CaSb battery, from Ambri.
I guess if all the cars can be wired to take in power, you just need a short stretch of 3rd rail at each charge point, say 60 m, not 1600. If each car has its own battery and drive motors, they don't need to be wired together. Relative charge rate goes to 10,000x, but the grid load stays the same.
A fuel cell driving a heat pump is an interesting notion. The "waste heat" from the fuel cell is not wasted, here. If half the contained energy drives the heat pump at, say, 3x, you get 1.5x vs just burning it.
The extra expense of the fuel cell and heat pump seem hard to justify unless you have other uses for them. This is to say that hydrogen for home heating is unlikely to be important.
LH2 is very attractive as aircraft fuel.
But I take your point: for other uses, the production efficiency counts more. Conversion to raw heat is pretty good, losing only what it takes to split the H2 and the O2, and then whatever of that heat you fail to direct to the end use, e.g. the steam that rises past the sides of your saucepan.
In many more places, crops are failing.
People necessarily leaving these places will need to go where other people already are.
If the electricity could be delivered directly to the train, the energy used would be cheaper, but installing (and replacing "shrinkage" of) many thousands of miles of "third rail" would cost a lot.
The hydrogen is not yet cheaper than the kerosene, but costs on that side are falling fast.
NRGV ("Energy Vault") was trading at a $2.4B market cap less than a year ago, and are now at $0.4B, still radically overpriced. Their product was an obvious scam before, then they pivoted to another obvious scam. The only way their stock could have any legitimate value would be if they still had $0.4B in cash, and could be prevented from squandering it. All perfectly legal, apparently.
Same for fusion startups. (It is just barely possible that Helion could have value.) Kyoto Fusioneering builds equipment to use in fusion startups' sham demos, and provides a conduit for dumb VC money into the pockets of startup founders and VCs who invest in it. All perfectly legal, apparently.