site banner
Advanced search parameters (with examples): "author:quadnarca", "domain:reddit.com", "over18:true"

Showing 25 of 107953 results for

domain:academic.oup.com

We do have an asset reporting requirement for any and all crypto. Probably related to our wealth tax.

Buying any goods or services with crypto must be reported for taxation too. The amount used for the purchase would get over 20% gains tax on it.

Not sure what I could spend it on anyhow without risking being discovered, if I were to try to keep it secret. But I'm a perfectly law-abiding citizen. Yup.

No dual purpose vessel can actually load troops onto a fastcraft or RHIB so there is no means to repurpose civ vessels to vomit troops onto Taiwanese shores.

Is this in contradiction of reports/rumors of China bulking up a civilian fleet that can be repurposed for amphibious landings? Was that actually a "backyard furnace"-level boondoggle all along?

They've just grown up with tools that work. They aren't shadetree mechanics because that hasn't been something in their environment. They know other things.

In your opinion what are these other things?

Sysadmin jobs haven't been maintaining systems by hand for the last 20 years (my entire career). A good sysadmin has long used scripts and other tools to help his work, and from there it's easy to move to infrastructure as code.

Good for him. Actually doing it might straighten him out a lot.

Not so arbitrary; selecting for intelligence may have its drawbacks. The infamous paper Natural History of Ashkenazi Intelligence (NHAI) hypothesizes that Europe accidentally selected the Ashkenazi Jews for intelligence for 800 years, and that their population also got a larger proportion of genetic brain disorders compared to human baseline as a result. The implicit warning is that anyone attempting a similar result on purpose will get genetic brain disorders in the resulting population.

Some people try to find scientific refutations for this hypothesis (PDF), instead of outright rejecting it as “eugenics bad” and refusing to do science.

Yeah, apparently his Reddit account is still active. He went on some big mountain climbing trip instead, which is less funny but clearly more sensible.

I think hive mind actually says average IQ is more important than your geniuses for a country level wealth.

My guess it’s because any country can do catch up growth (mostly), but having low IQ makes politics and institutions too crappy.

But they still have a clear path to a much larger GDP just by integrating rural peasants into the larger economy.

The rural population is aging faster than the urban population, is dramatically less well educated, and wages in urban China are rising too quickly to try a repeat of the world's cheapest workshop plan that worked in the 1990s.

There is still some steam left in that boiler, but not incredibly much and, once it's gone, it's gone.

Mainland China (and Taiwan for that matter) is going to be an extremely different place in 50 years. The demographic graying the mainland is going to go through is going to hit their economy like a flood of molasses and the 2070s are going to be exactly when they're in the thick of it.

The PRC essentially has until the 2040s to get something off against Taiwan, afterwards they will be struggling against the worst dependency ratio in world history. They have a good window between about now and 2035 when the American naval procurement cycle is at a nadir and they continue to grow competitive on hardware.

If nothing happens by 2035, nothing is likely to happen ever.

I just started using zyn. It's good but the 6mg is much too strong. I'd suggest the 2mgs or lozenges instead.

6 is enough to give me head and stomach aches.

Eh, I'd argue that sysadmin is maintaining running systems by hand, and DevOps is Infrastructure-as-Code. There is a qualitative difference when you switch from one to the other.

The Chinese military is, to put it bluntly, crap. They can't even plan a (non-artillery razing) takeover of Kinmen 10km from the Chinese mainland, much less cross the turbulent Taiwanese strait into hostile beachheads booby trapped to eternal inviability. All of this is due to the insanely difficult nature of long distance logistics provision that no one pays attention to because the USA has made it look easy.

Chinese amphibious capacity is less than 10 ships, which provides less than 1000 soldiers per trip. These are the only troop insertion capabilities China has available to move boots to the ground at scale, not even factoring armor or logistics. Paratroop insertion is meme level operations now (sorry paras, is true, you boys still slog hard in the shit), Chinese air assault capability is lacking (no heavy lift helicopter capability at all, extremely limited if nonexistent heliborne support capability) and helicopters don't have the range to reach Taiwan anyways. No dual purpose vessel can actually load troops onto a fastcraft or RHIB so there is no means to repurpose civ vessels to vomit troops onto Taiwanese shores.

There is no scenario by which China can force a landing in a contested area, even with unlimited missile strikes turning the western coast of taiwan into a wasteland (its actually really difficult to lay total waste to a city without tube artillery or air supremacy, and modern bomber doctrine doesn't do mass ordinance anymore). The most optimistic is a supremely timed landing with ordinance in-flight adjusted to strike immediate threats as the troops land ashore, a supremely difficult task that no one thinks is possible.

Will Taiwan suffer if China makes the play? Absolutely. Sending thousands of rockets to devastate Taiwan is enough damage, and while China has a crap army the Taiwanese are arguably in a much worse state. There is a wide angle of approach that the West can employ to resupply taiwan from the east, but that depends on either cross pacific journeys (everyone hates that) or securing Philippine-Taiwan shipping, which is its own cpntention against Chinese island 'bases'.

Ultimately, the Chinese military is still unoptimized for actually invading Taiwan, and shows no sign of being ready in the next 8, let alone 3, years. Strangling Taiwan by isolating it diplomatically and physically? That is its most likely plan of action. China is unlikely to entertain notions of contesting the 7th Fleet directly unless a decapitation against the 3rd and 5th can be conducted as well, and that basically is open warfare against the USA. Taiwan is not worth the risk, better to continue eroding US interests by exporting stuff that US manufacturers are unwilling to compete against.

I have a lot of enemies.

Well, my current job has most of the things you describe as being a part of the meaningful startups: the seniors with egos, the bureaucracy, the poorly thought out code bases and dependencies. The main differences are the fact that my job doesn't feel as impactful, but it pays triple to quadruple the salary you're citing for startups.

I didn't read the entire transcript, but I scanned Cohen's testimony, and I couldn't find any instances where he's asked to draw legal conclusions. The only thing approaching that that I could see, as you said in your initial comment, was that he admitted to having plead guilty to certain Federal crimes. The defense never challenged the admissibility of this testimony in general. They filed a motion in limine to prohibit the prosecution from using those pleas as evidence that the underlying crimes were committed, and they won that motion. The evidence of the pleas was admitted so that the jury could evaluate Cohen's credibility, and the judge gave a limiting instruction as soon as they came up. The defense's motion conceded that the plea evidence was admissible for that purpose. They never tried to get the evidence out entirely, and it wasn't in their interest to, either, because without the evidence of the pleas, it would seriously hinder their attempts to discredit Cohen. Given the limited nature of what Cohen actually testified to on direct, the prosecution probably wouldn't have even opposed a defense motion to keep the plea evidence out entirely, since the defense would have had much less to work with.

Beyond that, I don't want to get into too many details, but inadequate jury instructions and insufficiency of evidence are usually long shots when it comes to getting an appeals court to overturn a jury verdict. I argue in another post somewhere that intent (most of the time) doesn't require knowledge that the action is illegal. As for that last bit, it wasn't so much about hiding information from voters as it is hiding expenditures from voters. Laws requiring disclosures were created with the express intent of creating a certain transparency in election-related spending. I was reacting to the commenters here who were saying that Trump was in a kind of Catch-22 because there was no way he could have made the payment without drawing the scrutiny of the FEC. This clearly isn't true; if I were Trump's attorney I would have told him that if he wants to be completely safe he should pay it out of his personal funds and report it as a campaign expense. Alternatively, he could pay it out of his personal funds and not report it because unless it's obvious that sort of thing is rarely punished. Paying it out of campaign funds and reporting it isn't recommended but at least it makes it look like he's on the up and up. What I wouldn't tell him to do is to have a third party make the payments so they can't be traced to him, and then create phony documents to obfuscate the reimbursement.

Yeah I suppose so. Washington seems like a pretty bright guy though--I don't think it would take him long to grasp the concept.

I think there's truth to this, but there are limits. To use your example, if someone who wants to work in tech can't figure out how to plug in an Ethernet cable, then they are in fact lacking skills. Specifically, basic troubleshooting skills. I should be able to hand anyone an Ethernet cable and a computer with a port, and have them figure out how the two fit together. The large majority of one's value as a tech person isn't the specific things they know how to work with, it's their ability to figure out new things when confronted with them. I don't care if someone knows how to set master/slave IDE jumpers, but I do care if they just give up and say "I don't know" when confronted with such a problem.

This is just ricardian equivalence. Chinese food production is inefficiently optimized from a calorie provision perspective because other countries are supplying cheaper agricultural inputs for higher value produce, especially feeder stocks for protein (pork). Were China to be banned from all imported foodstuffs they will pivot (at uncertain pace) to primary staple production and adjust accordingly (potatoes and millet, primarily). Bear in mind that China has sustained its historically enormous population by farming all along the Yangtze and Yellow rivers during non-flood periods, and the Pearl River in the south has always had huge agricultural output.

The Three Gorges dam has stabilized water levels for Chinese agriculture for the foreseeable future barring damstriking, and cessation of groundwater-extraction dependant agriculture can be reversed on relatively short notice. Domestic food production can be spun up at a fairly fast pace, approximately half a year from seed to harvest depending on the municipality. Will it work? Not my specialization but most Chinese seem optimistic that they aren't at risk of starvation even if USA cuts off soybeans entirely or Ukraine ceases all wheat exports.

Combining the above with the point below that Chinese people stock up for 2 months of spare food at a time, and it is reasonable to presume that temporary food insecurity from US export restrictions will not affect China. Total trade blockage would be more impactful, since that would prevent China from easily pivoting to alternative import sources for either grains or critical factors of production.

Which... is really just sysadmin with a fancy name tbh.

I work in tech accidentally. I'm a "little sister." My older brother was very into computers, and I fell into it because working overnights as tech support while I was in law school gave me a lot more time to study than working overnights at a gas station. I knew enough by virtue of being around my brother to be competent (and back then, knowing the difference between SLIP and PPP was enough to get hired). And there was a certain level of trouble shooting and just needing to understand computers that you had to know to use them. So I could swap out cards in my computer as I managed upgrades, I'm not afraid of an IRQ jumper. I've run cables. But I am not a computer person. These are just things I learned by virtue of being around people who lived and breathed this stuff, or because I had to know it in order to use the tools for my specific purposes. And ultimately, because I ended up liking working in tech more than working in law. I'm good at what I do. People who learn I didn't intend to go into tech or that I don't consider myself particularly a computer person are often surprised.

My daughter (college aged) is basically a power user, even growing up with two parents in tech. She hasn't had to do the trouble shooting or the general tech support we had to do, because computers are functional tools now. When something doesn't work, after she turns it off and back on, she's kind of stymied, because things usually "just work" for her. It's the same way a lot of folks are with cars. I grew up with beaters, so there's a level of mechanical trouble shooting I knew that people who grew up with cars that just worked didn't know. Now, because she grew up with parents in tech, she can do basic trouble shooting, she can build simple electronic devices, she knows percussive maintenance can just work. But she has peers (in STEM even) who couldn't figure out how to plug an ethernet cable in (they've never plugged in a phone jack, either...).

I don't think that folks are necessarily less capable, they just don't have the same skill set. It's not required in the way it used to be. In the 80's, if I couldn't build my own PC, I didn't have a PC. Nowadays, building your PC is a niche thing only people who are deeply into (some aspects of) computers do. My daughter and her peers are going to come across as less able, in a lot of ways, than those of us who were in tech in the 90s (or earlier). They aren't. They've just grown up with tools that work. They aren't shadetree mechanics because that hasn't been something in their environment. They know other things. And they obviously have the potential (it's not like we're smarter than them). Nowadays, they have no reason to be able to chant orange white, orange, green white, blue, blue white, green, brown white, brown. So when I'm interviewing (for a junior position), I look for the curiosity, the trouble shooting ability, the engagement. This is particularly challenging because interviewing has become very scripted, at least where I work currently. It's also challenging because there's something about computers that makes a lot of people in it want to be the smartest guy in the room, and they can get really demeaning, really quickly, about someone who doesn't act like they know it all right out of the gate. As if proving themselves superior is more important than finding someone who can do the job. It's not rocket science, even when it is.

I think we often hire the wrong people, both because the candidates show up with less of the knowledge we're looking for, so it's hard to pick out the best ones, and also because the hiring process has become so weird that the ways in which we might have dug for the passion, the enthusiasm, even the basic underlying abilities (do they give up when they don't know something, or do they poke at it again?) aren't allowed. But we focus on, how did this candidate not know the TCP 3 way handshake cold? (Sure, maybe he should have, but it ends up being more important because we aren't allowed to get into how he figured out how to manage pedal feedback, because we'll never learn that he plays guitar in a band for a hobby, and maybe him explaining THAT problem is what demonstrates the trouble shooting and tracing skills that we're looking for, even if he spaced on the 3 way handshake in a moment of stress.)

I don't think that's normally how American law is applied, but admittedly I don't know much about it.

I'm open to being corrected, but as far as I can tell, it is. It's the standard that's laid out in the legislation, and as far as I know there's no case law on the books saying that those particular words don't mean what they appear to mean. The legal podcasts I listen to (Prosecuting Donald Trump, Serious Trouble) have asserted that's the standard that applies (and sure, I'm willing to buy that e.g. Andrew Weissman is biased against Trump, but I don't buy that he's intentionally misrepresenting what the law is). None of the lawyers I've seen arguing against the verdict have raised the definition of "campaign expenditure" as incorrectly applied (e.g. Steve Calabresi argues that campaign finance limits on hush money payments are unconstitutional, but he doesn't dispute that the statute purports to limit them).

I think you're assuming that intent to commit a crime requires knowledge of the criminal nature of the underlying act, when that's not the case (except in limited circumstances). To go back to the burglary example, suppose a thief breaks into a house with the intention of stealing a watch worth $800. The value of the watch isn't in dispute. The burglary statute requires intent to commit a felony, and the larceny statute makes it a felony to steal goods valued over $500. If the defendant is charged with burglary, he won't get the burglary charge dismissed by demonstrating that he genuinely believed that the statute only made it a felony if the item was worth over $1,000, arguing that because of his mistake of law he only intended to commit a misdemeanor and not a felony. To go back to the paper clip example and tie it into the New York statute at issue, suppose it's illegal to buy paperclips, and a junior executive at a company notices that one of his underlings bought paperclips. He doesn't know that this is illegal, but knows that his boss, the CEO, said that it was against company policy to buy them, so he forges documents making it look like the purchase was for something else. He can't argue that he didn't intend to conceal a crime because he didn't know what he was doing was a crime. He intended to conceal the purchase, which happens to be a crime, and he accordingly intended to conceal evidence of a crime; his knowledge of the legality of the underlying activity isn't relevant here.

I mean, ensuring that their entire population has at the bare minimum a subsistence diet of rice porridge and millet is pretty much rule #1 of any Chinese government so there's no way they haven't planned extensively for this, especially considering that every Chinese person I know thinks about food about 5 times as often as your typical white American and has a pantry filled floor to ceiling with non-perishable goods and a freezer that rains down ziploc bags of frozen meat and seafood when you open it.

It'd make sense for Taiwan to blow them up itself.

Indeed, TSMC has a kill-switch.

Would China have the stomach for an extended occupation and suppression if they weren't getting any value out of it?

Um, yes? Remember, this is a totalitarian state that's been putting irredentist messages about Taiwan in all the media and textbooks for literally a lifetime, and a decent chunk of whose current territory is very much an "extended occupation" (Xinjiang and Tibet; literally over 10% of the male population is interned in Xinjiang). They wanted Taiwan before the semiconductor factories were there, and they would get quite a legitimacy bump from holding Taiwan (the primary clause of the devil's bargain the CPC's made with the Chinese population is "we will get back everything that was lost in the Century of Humiliation", and Taiwan was one of those things).

(Also, do remember that occupations are significantly easier when you don't actually care about things like "human rights" and "rule of law". I have zero doubt that if it came down to "exterminate all 24 million Taiwanese and import mainlanders to replace them" or "abandon the occupation of Taiwan" the CPC would do the former, although I suspect their current plan is more along the lines of "kill everyone who was ever part of the DPP, stick the rest in re-education camps".)