Even if such a counteroffer were made, would European politicians be willing or able to accept it? My sense is that at least in Germany, print media is uniformly transatlanticist in outlook, and so to accept a Chinese offer over US objections would put you at the receiving end of a protracted storm of bad press. Few politicians inside the mainstream Overton window could weather this, since the sort of blanket press skepticism that would be required for an individual voter to dismiss a consensus of reporting from all the papers has itself been completely contained by association with out-of-window parties and movements.
Your estimate of engagement area is off by three orders of magnitude. Existing systems are designed to bring down drones from several kilometers away. This is actually really easy to do, and gets easier the closer you get.
I think we still may be thinking of different types of drones - there are the long-range plane-type ones like Russia's Shaheds/Lancets/Orlans and Ukraine's jury-rigged single-engine aircraft and some dedicated designs the names of which I don't remember, and then there are the low-flying helicopter types ranging from Ukraine's Baba Yaga to modded off-the-shelf FPV ones. The latter can easily fly between trees, buildings or stacks of containers; I don't see how you can engage them in a much larger area from a single point because in a busy industrial area there is simply no point near ground level that has line of sight of that much space. (You could of course place it in an elevated area and aim down, but then you are aiming towards the ground and I'm not sure what you would have to pay people to work in an industrial area covered by such a contraption.)
Then they got into a war where EW matters, and the truth became more complex. There are effective EW systems on both sides of the conflict, which can and do suppress drone activity, this is not some hypothetical annoyance when effectively deployed
I see an abundance of FPV drone video streams from both sides where the drone actually flies into a vehicle with EW equipment. This usually plays out as some noise in the video stream that gets worse as the drone gets closer, but the target is hit all the same. I'm sure there are cases where the interference results in failure, but cases where it does not are not one-offs.
I suspect they could not afford the war it would start... Even if they could, they nevertheless choose not to.
Since the whole "Russian economy collapse in a month" fiasco, I'd take Western predictions about the financial capabilities of its adversaries with a lot of salt...
Ukrainian benefactors are providing satellite capabilities to track Russian ships and communicate with drone boats
That's a fair point, but what do we know about Chinese satellite capabilities? Russia's legacy kit is one thing, but I'd imagine China to actually be quite good at something of type "get a lot of good cameras and radios into orbit fast".
Prototypes enjoyed significant software assistance from western companies
Interesting if true (but again, is that really the bottleneck for an adversary like China or even Iran?).
Russian navy vessels make the state of US navy vessels seem palatable by comparison - maintenance and armaments are very likely heavily degraded across the entire fleet to begin with
That's only really relevant for the scenario where naval drones attack ships, no? In the autonomous drone carrier scenario, they would not even get close to capital US surface ships.
an autonomous submarine drone carrier unloading a swarm on Manhattan
Would the "submarine" element really be necessary? Do you think the naval tech gap between Russia and the US is so big that Russia can't track surface craft of the size of Ukraine's drone boats in the open sea but the US reliably can?
your country is not a postage stamp investing heavily in modern missile defense systems to repel an endless stream of homemade rockets from the doghouse next door
You lost me with the metaphors in this passage.
The time to prevent drone attacks is before the first drone ever takes flight.
They've evidently failed with that in the case of Iran + proxies (and yet they are still not in an official state of war against either). What do you think would happen if, say, China did a drone-swarm warning shot against the US, say in the context of US saber-rattling against a blockade of Taiwan intended to break its resistance? It's hard to predict, but I could see a drone attack that manages to largely avoid human casualties failing to elicit the Pearl Harbor response and instead making public opinion lean towards "yeah, we don't need this war".
Despite my substantial disagreements with the wokeists, I'm willing to fund them if they can act as a counterbalance to a complete takeover by utilitarian techbroism.
Are they effective as a counterbalance, though? I'm yet to see a "utilitarian techbroist" excess that the SJ movement could not be made to acquiesce to in return for assurances that it will be wielded in the interest of their political goals, and in fact what seems to happen is that every time such assurances are made (ex: any social media opinion-management tech) the would-be counterbalance becomes another hard obstacle to overcome if you want to put brakes on the technology (ex: the "freeze peach" meme, fielded by SJ in defense of Big Tech getting to do what they wanted to do anyway).
In general, it seems pretty counterproductive to hope for an incumbent ideological movement to rein in an incumbent technological one. As long as the concerns of temporal and spiritual power are orthogonal, they are naturally complementary to each other; what you are proposing is akin to wanting to check a medieval absolute monarchy by investing in the priesthood.
Drone defense is surprisingly straightforward, provided you're a real first world country.
This sounds like copium to be able to say that it doesn't mean much that Russia and Ukraine are not managing to pull it off reliably.
Russia has widely been considered superior to the US in EW, and yet both Russia and Ukraine are now in a place where all their EW measures are at best a minor annoyance to each other's drone activity and the only things they can jam reliably are stodgy known-frequency systems like GPS and Starlink. Shotguns on turrets sounds appealing, but I haven't seen evidence that it works reliably in a realistic settings - physics get in the way of any sufficiently heavy cannon rotating to track a fast-moving close target, an additional drone coming in from a different angle costs much less (and eats much less manufacturing line capacity, before you start talking about GDP gaps) than an additional turret, and with anything more advanced than Shaheds the drones can come in low/sneak around terrain in such a way that just firing a shotgun at them is bound to cause collateral damage. Then, of course, a modern country's functioning depends on the safety of more than a few "high-value" locations - a Factorio gamer faction like China could easily afford paralyzing a city by sending one quadcopter equipped with a grenade and a frequency-hopping transponder to each gas station and perhaps even each of those small plastic roadside electric/telecom switchboxes. In terms of larger infrastructure, a container port occupies tens of square kilometres, while a putative scifi shotgun turret against low-flying drones in such an environment could perhaps cover a 0.01km² area.
Of course, cheap by military standards is still ludicrously expensive by infrastructure standards, and there's a few orders of magnitude more critical infrastructure targets than military targets, so there's not really a scalable solution to this problem that involves grounding or destroying drones just before they strike infrastructure targets. The actual scalable solution is to license and regulate drone ownership, and use early warning systems built on top of existing surveillance capitalism to track and crack down on anyone whose purchasing habits start to look like the incredibly obvious signs of building a drone fleet, not to mention the equally obvious signs of building a ton of explosives to attach to those drones.
There may be a reason why the NJ drones are reported to come in from the sea, too. Ukraine has demonstrated the unreasonable effectiveness of jetski-sized drone boats. Cartels have already DIYed similar craft. It wouldn't take much inventiveness to replace the explosive payload of one of those designs with 4 quadcopters to be launched at inland targets when the boat gets close enough to shore.
Anyone with the capability to overcome regulation and surveillance and still pose a credible threat (cartels, China, Russia, maybe Iran) faces the risk of starting a war with their actions - and if this risk isn't enough deterrence, we've got bigger problems.
Well, the thing is - speculation about the game theory of an actual direct US-versus-adversary conflict and how the ability to wreak more non-nuclear chaos on the US mainland may impact the game tree aside, the goals and ambitions of the US still go well beyond defending its own territory, even if this is a hard sell to funders and the voting public sometimes. The problem the US currently faces with drones is not just that it may not be able to defend its own territory; it's also that there is no technology platform it could even hypothetically provide to Ukraine, Taiwan, Israel, Australia or any other ally that is not quite under the US nuclear umbrella or even the US dead fresh-faced college kids umbrella to save them from the threat of drones, and this is a glaring limitation to its ability to project power. If the US just wills it, you could be made safe from traditional air attack, obtain arbitrary amounts of firepower, or sub-1h delay high-resolution satellite imagery and RF emissions data for any point on earth; but apparently (and so whether you get those blessings is merely a matter of being willing to pay up/sufficient alignment with its objectives); but it turns out that if you are suffering a death by a thousand drone cuts all across your territory, this is beyond the Fairy Godmother's powers to prevent.
There's probably a significant element of mass hysteria adding noise to the body of reports, but I found this theory from Reddit (tl;dr: US government sockpuppet theatre to create momentum/public or internal support to marshal significant funds for a drone defense moonshot) plausible. Considering the way the Ukraine war is being conducted and the Chinese drone swarm videos that are being circulated, it would be shocking if US military planners were not currently running around desperate for a way to get the funding bodies to acknowledge the scale of the problem without projecting weakness to the outside.
As in my own favoured theory for the "tictac video" UFO case before, there could also be an element of flexing the US military's own capabilities to likely adversaries. "See, our ship-launched drones confuse and overwhelm even the rest of our military and civilian law enforcement. Do you think you would have a chance?"
I'm sorry, but Western society was founded on Greek paganism. Christianity almost destroyed it once, and we only squeezed through after a thousand-year rut by deciding that Christianity as it was is uncool and remaking it in a different image. In fact, we to then continued to tweak away at it further to great effect for some 500 years more. No wonder that, having been left with such a strong cultural memory of this serving us well, we would eventually slip up and remake it again in a way that is bad without even realising how we screwed up.
As a parallel comment already stated, there is not actually anything confusing about instances of people not enjoying something that they claim is enjoyable. There are people who overeat, people who are fed up with doing daily quests in live-action games, people who drive for 8 hours to go to an amusement park and spend their whole time there bored, people who align their whole lives to enter a profession and then are miserable from the realities of it, and even people who go halfways across the world to visit a city and break down from disappointment; yet, nobody generally takes those things as evidence that food/games/amusement parks/professional commitment/international travel need to be made exceptional and/or taboo again.
The Rationalists had a phrase I admired, "I notice I am confused". They realized confusion was useful, and an opportunity to seek deeper understanding. For those who hold that sex is harmless entertainment, this seems like it ought to be confusing. Why did she expect to enjoy doing this, and why didn't she enjoy it?
This sounds like you are confused (about the lack of visible confusion from "those who hold that sex is harmless entertainment"). What do you think could be possible reasons for this lack?
Every time someone goes viral having a bad experience from sex, conservatives like clockwork parachute in to run victory laps wearing their best told-you-so face, as if the case for the sexual revolution rested on an argument that nobody can ever possibly not enjoy sex. Have you ever seen this persuade someone - not in the way where someone already agreeing pretends to be persuaded, or someone who is in the process of joining your tribe taking notes on the ideological package they are supposed to download, but someone who actually isn't with you and isn't about to be performing a very specific update to their worldview regarding whether the sexual revolution was a good thing?
It's not like norm violations can only be answered by either mirroring the norm violation or by doing nothing at all. To have any chance of maintaining norms of mutual disarmament, such as the former uneasy liberal "colour-blindness" consensus, one needs to find a way to punish norm violators in a way that signals that one is still interested in maintaining the norm, and that the punishment can be withdrawn if the norm violation stops. In this concrete case, this would amount to punishing left-wing racial interest groups while taking care to not be directly seen as benefitting right-wing ones.
For example, I don't know, commit to always refuse to hire/admit enthusiastic racial spoils beneficiaries (DEI committee members, people who lived in racial program houses in college, ...) if alternative candidates of the same race who refuse/renounce them are available? (Keep the actual underlying quota to be legally safe or at least force courts to opine on whether you are allowed to specifically discriminate against the incumbent political group.)
-
What if causation is structured differently? Like, a preexisting proclivity for temperance leading to (1) preference for smaller breasts and (2) better life outcomes.
-
Does the lower class thing hold up in places where there is no correlation between class and obesity rates? Most people like looks that they grew up with.
-
"Those who disagree are just falsifying preferences" is a too emotionally satisfying theory to not be suspect. Also, even if it is shaped by social pressure, in what sense is a sufficiently internalised preference less valid? Nobody likes IPAs or jazz in a state of nature, either.
-
I would guess that IQ correlates positively with a lot of sexual preferences that are not reproductively advantageous. Furries are known to be something like the Jews of IT, too.
I don't see this being usable fuel, at least. The shooting of an insurance company CEO codes edgy left comparably hard as "transport-oriented teens" Facebook meme groups that pivoted to violent fantasies towards landlords. Any Cathedral cleric with the slightest social spidey sense (which is something that the job actually requires) will see that "the CEO shooter was actually a rightish rat" is much more likely to make right-wing rationalism attractive to edgy left-wing youths than it is to galvanise further opposition to right-wing rationalism.
More likely that the 3D-printed weapon angle will be played up significantly.
...and then the "left woke" can point at your "good tribal politics" to argue that white supremacism is a real threat and galvanise support for non-white racial supremacism. This feeling of inevitability ("the alternative... other tribes simply take from you... don't bother... doing anything about it") surely is what toxoplasmosis feels like from the inside.
Well, I don't intend to make the case that what the shooter did was in fact positive-EV with respect to your or my value function here. But given your prediction, there are possible value functions that he could have that would make that outcome positive-sum for him - he would just need to have the right mixture of concavity (weighing outcomes of the form "one individual denied prohibitively expensive life-saving surgery" much more than one million times "moderate misery due to your finances being wrecked by high insurance premiums") and aversion to surprise/betrayal (someone who thought they were covered being unexpectedly denied being a much bigger deal than someone not being able to afford insurance at all with higher premiums). The former is basically just left-wing care-foundation ethics, and the latter is common.
The would-be reformer will avoid the industry entirely now.
Surely depends on what kind of reform they planned. There are reform plans that do not entail becoming a successful insurance CEO with a target painted on your back. You might think that any such plans are reforms for the worse, but the shooter may think otherwise (e.g. see above).
I'm inching closer to the pitfalls I just described myself by going there, but executing such a "scapegoat" has effects that go beyond "the scapegoat is now dead", and therefore "saying that any or all problems are but-for causally due to the existence of the evil CEO is facile and wrong" and "it is a good intervention to unalive the CEO" are compatible viewpoints.
Someone could believe that this should be done pour encourager les autres, and the intervention will leave every actor that is involved in this complex system second-guessing how far they can go until some Luigi chooses to throw a koopa shell at them, and perhaps choose to exercise restraint; they could believe, in an Assangian way, that just the necessity of hiring more and fancier security details will raise transaction costs for similar enterprises in a way that will make them just a bit less sustainable; they could even anticipate the approximate shape of the popular reaction that we actually got and conclude that the society-wide soul searching and increased salience of the system that the scapegoat figureheaded alone will cause beneficial changes as a side effect, outweighing any negative ones. What impact do you figure this event had on the probability that a startup pitching "reform medical insurance" gets funded a month from now?
It is unusual for kids to be more attached to peers than parents.
Small-n and possible bias due to typical-minding, but this was not at all what I observed in my environment growing up (and I would expect the schools I went to to be biased for some measure of well-adjustedness if anything).
You might scoff, but in the sixties a survey of high schoolers found that, if all their friends wanted them to join a club but one of their parents said no, most would not join the club.
What was the exact survey question/setup? Did it come with a guarantee that if you join, your parents will never find out? Otherwise, this would have been confounded by fear of consequences. (Many people are not confident that they can maintain a lie in front of their parents, which could be internalized like "I wouldn't want to live with the guilt".)
Do you not figure some Mottizens could believe that what he did was an optimal action while in their normal state of mind? There are reasons to assume that related positions, if they are present, would be underdefended in the discourse here: after all, a defense would involve a lot of needle-threading to get dinged by mods and the community for fedposting, nor trip any possible actual fed-operated web crawlers that may have been deployed for early threat identification. (Internet lore certainly includes tales of excessive post-election Twitter posturing resulting in unexpected Secret Service home visits.)
I remember when this forum discussed Cuties, and there was no shortage of right-wingers who aired out their righteous seething fury about it.
People who fear their neighbour are wont to beg for the elites to protect them from him.
There's no visible spike in ngrams around 2003; in fact it starts growing around 2000 and grinds to a halt around 2006. You could maybe argue that the later growth was due to the Obama administration's use against Bush as you posit, though it seems that the graph starts growing a bit late for that. A search for opinions on its usage in the wild before Russia/Ukraine mostly brings up references to Australia using it with respect to China (example). I also tried to search for the phrase filtering for things up to 1/1/2014, and the top results that were dated correctly are about the ICC ("only for Africa and thugs like Putin"), a paper that flat out only has the phrase "American liberal hegemonic order" in its non-paywalled part, and a Clinton speech referring to the US as the cornerstone of "rules-based international order".
I think the problem is not so much how the term "rules-based international order" is used with respect to the US, but that the term is basically not used with respect to the US at all, except by a few low-agreeableness cranks who can't read the room. German newspapers can barely publish a single article about what Russia or China do in their neighbourhood without referring to the "rules-based order" or adjacent terminology such as "war of aggression" ("Angriffskrieg", a German favourite), but either of those things is hardly ever mentioned when discussion the wars of the US or Israel at all. The other "rules" are evidently also only for Africa and thugs like Putin.
If someone creates a set of rules for you to follow (and the ones that created them were American institutions, which many of these articles are not shy to brag about) but does not need to follow the rules themselves, to follow the rules is to accept their sovereignty over you.
de facto
Did you mean to say ipso facto (by itself/automatically)? De facto is used more like "essentially" or "more or less", which, while it works here, seems a bit unidiomatic.
If every right-wing win is de facto evidence of Russian election interference and every right-wing loss is a victory for democracy, the rules-based world order is screwed.
Why? Right-wing movements in many places have come to oppose American globalist hegemony recently, and "rules-based world order" has always been politics speak for "obey the American globalist hegemony". If the people internalise what you said and consider anti-$thing wins to be tantamount to [enemy action] and pro-$thing wins to be tantamount to [cluster of positive affect], this seems pretty good for $thing.
I think Sweden is may be an outlier in terms of the hassles they put up for international transactions. I had to go through what sounds like the same process so that I could make a SWIFT transfer to pay a bill to a German company; the bank (Nordea) representative told me that they have to do this due to some new regulation concerning international money laundering. I have not encountered anything like it between other EU countries (I get around a fair amount), Eurozone or not, even in recent years.
There is no way such a rule would be backed by the Western countries unless they could get reassurance that it would never be applied against them in the court of public opinion. A quick search reveals USAID spent $63 billion in 2024, and this is not counting other allied financial moves like the EU's offer of more than 10% of Moldova's GDP as a loan to help the pro-EU candidate. (Imagine the pandemonium if Georgescu's campaign had a promise of a $35B loan from Russia attached to it if they leave NATO!) Then, there is the circumstance that Western culture is "Universal Culture". The West out-spends and out-memes its adversaries on its periphery regularly; losing one rare match is not a good reason to throw a tantrum and quit the game.
Without looking it up, how many people do you think are shot by police in the US, a country of 330,000,000?
Unfortunately looked this up already in the context of the argument earlier, I think it was 600.
Without looking it up, how many times do you think police engage in "police coming in like an infantry platoon clearing an enemy building"?
I would guess significantly more - if we make it something well-defined like SWAT dispatches, perhaps on the order of 100k? Is that data collected anywhere or is it another thing where you could only find local data and not everywhere due to how fragmented the police force is?
As you said, Europeans and Americans have very different perspectives. Americans would characterize this as state ownership of children and very authoritarian.
I think there's a general theme that relative to Europeans, Americans are more concerned with impositions by the state but much less concerned with impositions by non-state actors, even though from the perspective of an average citizen the two might not be readily distinguishable as lofty authorities. As a caricature, we figure that an American would get very upset by the government banning him from soapboxing for some political position, but would see nothing wrong with it if a corporation bought up all roads and public squares in his city and instituted a ban against voicing the same position on company property (along with a host of other house rules). Moreover, if someone then proposed to force the company to surrender roads or parks to the public hand, or circumscribed its right to enforce rules of its choosing on it, the American might be up in arms about that being an intrusion upon the company's free speech.
When Europeans call America authoritarian, it comes off as preposterous to us. Putting people in jail for mild social criticism is nuts and authoritarian and has nothing to do with "libel and slander."
Right, and putting people in jail for 25 years to life for all sorts of one-off transgressions comes off as nuts and authoritarian to us, as to million-dollar fines and jail terms for software piracy (...).
I'm getting the sense that you switch between taking my words overly literally in some cases and loosely reinterpreting it in a way that is convenient for your argument in others. To rehash:
-
I think that searches, SWAT calls and other similar "they come to your house because they think there is a threat in it to neutralise" situations in particular are a scenario in which I would feel much less safe around US police than around European police.
-
Data supports that US police in general are much more likely to injure and kill those they interact with than European police
-
Personal experiences support that US police are more hostile and less helpful than their European counterparts. This is in their interactions with me as a Caucasian academic with naive good-kid vibes; who knows what they would do if they were responding to a SWAT call or following a lead from someone in the computer security "industry" I know.
-
I grant that there are reasons they turned out like that, but I see no evidence that they are not like that to everyone, i.e. that the hostility is precisely targeted at the uniquely American problem elements. There are more YouTube bodycam videos of American police roughing up harmless-looking white kids than total incidents of German police doing that.
It was an example. As I said in response to your other post, in every easily-delineated scenario where there is evidence that can be compared, US police look worse. To try to rebut this by dismissing each easily-delineated scenario as an irrelevant small sample seems like a god-of-the-gaps argument to me - "surely in some other domain that we just so don't happen to have good data on, US police are actually nicer and more professional than European police! What, they're also hostile and violent in this one? Guess they must be nicer and more professional in one of the many others!"
I said that I think that 2% do, but based on the data I could find the only thing that can be proved is that there is an upper bound of about 25% (about 600 total killings, of which an unknown percentage happened during federal searches, vs. ~2000 federal searches).
- Prev
- Next
According to what metric? GDP per capita? I'd consider that stronger evidence for the proposition that GDP does not measure anything useful than that Japan actually underperforms according to any sensible measure of performance. It is more surprising that people still have this blind faith in economic metrics considering how official inflation figures have gone viral for being comically disconnected from easily observable real price changes in recent years.
It's a blatant exploit, which seems designed to inflate the score of systems implementing a particular economic culture, that GDP includes the "market value" of illiquid and socially constructed line items which can be created out of thin air at no material cost. If the US were reduced to an irradiated reservation occupying half of Wyoming after a future war against China, its economists would probably still find a way to award themselves the superior GDP medal by having the two surviving lawyers on the territory mutually bill each other a few hundred trillions for consultations.
More options
Context Copy link