magic9mushroom
If you're going to downvote me, and nobody's already voiced your objection, please reply and tell me
No bio...
User ID: 1103

It seems likely the Democrats will obtain control of the House in 2026 and probably the whole enchilada in 2028,
I think you're underestimating the capability for current trends to get scrambled; we live in Interesting Times.
Calls for the murder of Elon Musk are frequent and widespread.
For clarification: is this still only randoms on social media, or are there notable figures/organisations taking this stance as well? (Not saying randoms on social media are harmless, but there's a major difference in degree there.)
I will note that while I'm also quite concerned about the situation in the 'States, there are widely-varying threat levels among Western nations. I don't see Australia collapsing even if given a hard shove, for instance. Canada and Europe seem to be somewhere in-between.
If he's old enough that a progress of months is enough to make meaningful differences in his cognition then he was not of sound mind to be president.
I feel obliged to note that this isn't necessarily the way it works. If he had a stroke between then and now, for instance, that's a sudden loss of brain function regardless of what that function was before the stroke. Hell, if he had a stroke before the use of the autopen affecting his motor control, that would explain why he couldn't sign his name without necessarily implying anything about his cognition.
Cognitive decline is not always gradual, and loss of motor control without cognitive decline is a thing (see: Stephen Hawking, who certainly wasn't a vegetable).
A lot of people would rather the government focus on policy than on scandals. They don't care about Biden's dementia or the J6 Committee's conduct; they care about what Trump can do for them in the here and now.
As long as it's clear that blaming China openly will not help with either cleanup or preventing further accidents, there's no point in doing it. If China says (through less-than-public channels) that yeah, we screwed up, we're sorry, let's stop the blame game and instead let's think how to handle the mess, it makes practical sense to play along.
Do want to note here that there is always one way to absolutely prevent further accidents which does absolutely require openly blaming China i.e. starting WWIII. Fucking around with pandemic pathogens in insecure ways is one of the things where that option does have to be considered; consider the death toll that would have resulted if COVID had had the case fatality rate of plague.
I said at the time something along the lines of "I sure hope it's not an actual bioweapon use, because if it is that's casus belli". I stand by that conditional. The actual situation is not quite at that level.
I do think you're being a naïve first-order consequentialist here, though. Lies are never free. Even leaving aside RenOS's good points regarding levers (and sure, Chinese popular opinion has little effect on PRC government actions, but Western popular opinion has an effect on Western government demands which do have an effect on PRC government actions), our institutions going along with this damaged popular trust in them, which is a big deal. The WHO is trusted far less these days because it bent over backwards to play the PRC's tune. Wikipedia had a crisis over this issue which led to a near-complete purge of conservatives via Parable of Lightning, damaging its legitimacy. The US's dissident right grew, raising the likelihood of a civil war there. These are not small costs.
But within reason, I generally lean on the side of privileging the freedom of the (public) artist, regardless of the aesthetic preferences of the public who will be exposed to their work. If it's that important to you, then you should consider becoming an artist too. And if it's not sufficiently important to you, then you are at the mercy of the people to whom it was sufficiently important.
I'm not sure that this is coherent. If the artist has the freedom to put a sculpture of a gory corpse outside my house against my will, because that fits his conception of beauty, then do I not have the freedom to melt down his sculpture with a blowtorch against his will, because that fits my conception of beauty? Am I not also an artist, for making the world around me more beautiful as I see it?
You might say "well, he got approval from the government and you didn't", but since we're presupposing that the public agrees with me, and since this is presumably a democratic government that is supposed to follow the public will, for the government to give him and not me approval is an obvious bug, not intended behaviour.
For 20+ years, in peace time, thousands, possibly tens of thousands of professionals engaged in Nazi/Imperial Japan style extremely unethical human experimentation.
Extremely unethical, yes, but still not quite on the level of Mengele or Ishii.
I think that either of die Linke or AfD would demand even more than the Grüne demands now.
I'm not so sure about that in AfD's case, although it's still a non-starter AIUI because the SPD and Greens would pull out to maintain the cordon.
I accept it is plausible that climate change will cause human extinction within a few decades. The same is plausible for nuclear war, an asteroid impact, a superbug, a super-volcano, renegade AI, et. al.
Nuclear war: no, there's no mechanism that lets you get to extinction. Nuclear winter is literally a hoax, blast/heat/local fallout are too localised, and global fallout's too weak (I ran the numbers on Cold War arsenals and those weren't nearly enough). It could get to #2 on the list of "disasters in history by %humanity killed", but #1 is dubious, let alone X.
Supervolcano: another Yellowstone wouldn't do it (this is known fact; humanity already survived Yellowstone three times before we even tamed dogs). Another Siberian Traps might, admittedly.
Climate change/superbug: not in the normal senses. A normal pandemic can't get everyone because R drops below 1 before #humans reaches 0. The only "superbug" (i.e. infectious agent) that could get actual everyone is a full-blown insect-zombifier-for-humans where victims actively and intelligently attempt to infect others (rabies and toxoplasma are nowhere near precise enough), and that's highly implausible without intelligent design (there's nothing with this level of precision in any mammal, and it's generally thought to get harder with brain size). For climate change to get us would require, well, another Siberian Traps, or a Chicxulub+ impact (another Chicxulub wouldn't do it, due to preppers if nothing else), or some omnicidal maniac deliberately manufacturing and releasing millions of tonnes of fluorocarbons; I'm specifically not including "some idiot blocked out the Sun with a solar shade" because people would notice that and destroy it (with massive casualties, but not X).
Asteroid impact: technically no (at least not without terrorist redirection), but in practice I'll grant this one (we've found all the Earth-crossing asteroids of sufficient size and ruled out collisions, but comets are harder to predict). Low probability, though, particularly given the requirement for a Chicxulub+ one.
AI can do it. Life 2.0 can do it (here I'm thinking of things like a non-digestible alga that doesn't need phosphate and has better-than-RuBisCO photosynthesis, not a pathogen - an independent lifeform that terraforms the planet in ways that are incompatible with human survival, in this case by causing a superglaciation plus total failure of open-air crops). New physics catastrophes and terrorist geoengineering might do it, although I'd be more concerned about those on the scale of centuries rather than decades. And obviously there's the "unknown unknown" term which is unknowable by definition. But AI and Life 2.0 are the known X-risks that scare me. (Obviously there are GCRs that are significantly more likely than any X-risk. Nuclear war's highly likely to occur sometime this century if we don't get X first; I just expect not only people to survive but myself to survive.)
You need another premise to get that conclusion, specifically "light-cone-breaking FTL is impossible".
Nobody knows whether this is true. The "light-cone-breaking FTL = time travel to times before the creation of the time machine = lol where are the time travellers" issue is not a clear no-go, as current understanding of relativity and quantum physics suggests that any attempt to use FTL to build a time machine would fail due to the quantum vacuum misbehaving and collapsing your FTL method at the exact instant time travel becomes possible.
Trump isn't flying blind because the sensemaking institutions he inherited are so corrupted as to be worthless. He is flying blind because he is an unrigorous vibes-based thinker.
Scott, "Planet-Sized Nutshell":
There is an extraordinarily useful pattern of refactored agency in which you view humans as basically actors playing roles determined by their incentives. Anyone who strays even slightly from their role is outcompeted and replaced by an understudy who will do better.
In a sense it's both. The reason Trump is that way is because he's Trump. Why is Trump POTUS, though, rather than losing the Republican nomination back in 2016? Well, because all the sensemaking institutions said "Don't nominate Trump" and the Republican base treated that as an endorsement. "Be a huckster" was no longer a losing strategy to get the Republican nomination, because the Republican base no longer trusted the institutions that normally filtered out hucksters, because those institutions by then also filtered out anyone loyal to the Red Tribe; indeed, the base went so far as to anti-trust the institutions and deliberately do the opposite of what they said.
An attempt at synthesis of your point and FC's (plus some others, like Zvi's notion of the Incorrect Anti-Narrative Contrarian Cluster): Governments in the 90s (or whenever) were Right. SJ governments were Wrong. Trump and Musk seem to be Not Even Wrong.
At greater length: there used to be coherence-producing mechanisms in society that kept everybody on the same page and kept policy making sense. Then when SJ nucleated, it hijacked nearly all those mechanisms and put false stuff into them, leading to obvious falsehoods being promulgated (and multiplying due to the principle of explosion) and policies that didn't work. The Trumpian reaction to SJ realised that it could not restore neutrality in those mechanisms (because SJ did too good a job of installing political commissariats in them) and it was excluded from the Overton Window those mechanisms had produced, so it took a sledgehammer to their credibility and turned much of the right-wing against them. However, it failed to build new coherence-producing mechanisms to replace them*, leaving much of its base and apparently also Trump/Musk without any way of identifying and co-ordinating on truth and on rational planning; they're just saying things. And, well, a government that's working with no map at all is going to do worse at a large number of things than one with a map that loosely resembles but does not match the territory.
Of course, a little buffoonery until Trumpism can sort itself out wouldn't be the end of the world... except that we have a certain drooling dragon at the door watching our every move, which means it might actually be the end of the world as we know it. Oh, well, I live in Bendigo and have most of my prep done**; "I'm clear. Are you?"
@FCfromSSC, thoughts?
*There are rightist intellectuals capable of identifying truth, and there now is an alt-media apparatus capable of dispensing it. The problem is that one isn't plugged into the other - not in the USA, at least, although we seem to be doing a lot better in Oz (the UAP is showing some signs of Trumpian brain rot, but it's not even our biggest alt-right party, so I'm not very worried). A lot of us were counting on Musk and Vance to be lynchpins, but Musk appears to have gone nuts and I haven't heard much about Vance in all this.
**There's a reasonable amount that's best left to "when the war starts" because it's got downsides and/or a use-by date plus isn't panic-buying bait.
Basically, just drop bases in Germany, North Dakota, Beijing, South Africa, Argentina and Tasmania, and give them all Hyper-Wave Decoders (I beeline HWDs immediately after the no-brainer Laser Rifle; I usually deploy the first three bases with Large Radars but wait until HWDs for the last three). This is cash-hungry, but because it lets you recover more UFOs it pays off very fast. Normally I only keep three Interceptors and wait until Avengers to have interception everywhere (because I don't do much interception, preferring ground assaults when possible, and because you need advanced craft to shoot down Battleships anyway), but if I were on the back foot like that and I didn't have Avengers I would spam them (they arrive in four days and Plasma Beams build fast). And yeah, there's a bit of micro with sending Interceptors to where the UFO is going rather than directly at the UFO to avoid long tail chases.
Yes, I've played it. I've played it alot, thank you. You trying to brush off it as 'just a meme' makes me wonder if you've played it.
I've beaten it on Superhuman without active psi and with Cydonia on June 1, and I literally wrote a non-negligible chunk of the wiki.
The hit chances are accurate, at least for "hit" rolls vs. "miss" rolls (rolled misses can still hit at close range, and I think it's possible for rolled hits to miss at extreme range or where there's cover involved, both of these symmetrical between X-Com and aliens). Aliens hit a lot on higher difficulties because their accuracy stats are pretty high. If you think there's "weighting" going on that makes X-Com rolls systematically worse and alien rolls systematically better than the normal formula, you're seeing a pattern that's not actually there, presumably due to negativity bias letting you recall "bad" results better than "good" ones. I guess you're one of the (many) sources of that meme.
There are three ways that the AI "cheats"; it can perform Auto mode reaction shots (you can only use Snap), it "remembers" the position of your units after they leave LoS (most notably allowing psi-attacks on them), and it gets omniscience after turn 20. This isn't one of them.
The remake has an ironman mode that only allows auto-saving, no manual.
I wasn't talking about loading saves; my point was that you can continue from a lost battle without having to just load a save.
The original X-Com was also rather infamous for it's odd, weighted chances of ally fire missing all 99% shots, while aliens were able to snipe characters from across the map.
Have you actually played it? This meme is just from people whining about missing, not a real thing. (Actually, the remake has a worse case of "you missed an alien standing right next to you"; the original makes rolled misses fire randomly within a cone, which will probably still hit if the alien's close enough, whereas the remake forces rolled misses to actually miss.)
And this still didn't stop aliens from sniping your guys as they came down the ramp, as sheer random chance could still fuck you over hard.
That's not random chance; that's you not knowing how to negate it. Smoke grenades give you concealment and thus block reaction fire. In TFTD dye grenades don't work, but in TFTD there's much less of a problem with this anyway (because the Triton has a door and is flush with the seafloor, and because you can open doors without stepping through them in TFTD).
About the only truly-irrecoverable squad wipe in the original (besides Cydonia, of course) is if your main base gets raided early on. This is how I lost my first attempt at TFTD's Superhuman difficulty: a 1 January FBA proc wiped my base while my Triton and main squad were out (because I might be good, but I'm not good enough to beat an Aquatoid Dreadnought crew with three guys), and while I didn't lose immediately from last-base-killed (because I'd placed another at game start) there wasn't enough time to get my second base operational before I lost from poor score.
While you do need substantial amounts of equipment/training to fight Ethereals, by the time they show up you should have a worldwide interception grid, which means that if you're stuck without an Ethereal-ready squad, you can just stop fighting Ethereal missions until you have one (with perhaps an exception for an Ethereal Small/Medium Scout, if you haven't gotten a psionic capture yet; they aren't too bad). You won't lose from poor score, because shootdowns themselves give bucketloads of points and if you're shooting down incoming UFOs the aliens won't score from succeeding at their missions (in particular, because you're shooting down the Terror Ships, they won't be able to create Terror Sites, which means no massive penalty for ignoring/fleeing from them). Quite recoverable, if tedious.
I haven't actually played the remake, to be clear, and that "mostly" was doing some work because I did know about soldiers being less expendable and there being more hardcoded-game-over missions.
EDIT: "Craft is lost" is painful enough that it's usually not worth risking one, but there are cases where I think optimum play is to take substantial chances of one. The most notable is the first Sectoid terror site in UFO, which will be an absolute ball-buster of a mission (this is the only mission in most games of UFO that approaches Cydonia in relative difficulty; early base defences can be worse but are rare) but which is your first chance to get a Sectoid Leader capture. It's worth taking a substantial chance of failure, including craft loss, to get that capture; you don't know when you'll get another chance and you really want psi to at least be in the works when the Ethereals show up in July. In TFTD there aren't many cases where it's both an incredibly hard mission and a chance to get something rare. The first Terror Site is quite hard, and is technically the only guaranteed appearance of Deep Ones, but there's an 80% chance to get Deep Ones in both the February and March Terror Sites so it's probably not worth risking a wipeout; the first Aquatoid land mission (and thus Calcinite corpse chance) isn't quite the horror show of the first Sectoid terror site in UFO (due to Calcinites being a hell of a lot weaker than Cyberdiscs, and due to a lack of need for rank identification/live capture) so while I think it's probably worth risking a craft loss if you have to (to get drills for cracking open Lobster Men), and I have thus risked it, you usually aren't in dire enough straits that it's necessary (at least, not if you're as good at the game as I am, and if you're not playing some kind of challenge run). There are, of course, plenty of hard missions in TFTD, but most of them can be aborted without missing out on anything irreplaceable.
There is a poster here (forgot who, apologies) that uses the metaphor of XCOM frequently. In XCOM, if you die your run is permanently over. So unlike gambling, in XCOM you only want to take odds that ensure victory, or nearly ensure so. You would never consider a “90% chance of winning the engagement” dice roll, because over eleven engagements you’re going to lose permanently.
Uh, that's not how X-Com works, and AFAICT it's mostly not how its bastard remake works either. Actually, one of the reasons the original X-Com winds up on so many "best of all time" lists is precisely that it does allow you to continue from losing battles, and thus has a more realistic war feel because you are not (artificially forced to be) some magic plot-armoured force that wins literally every battle.
More importantly, what possible answer could any of these questions provide from a pro-Ukrainian supporter that would or would not illuminate the prevalence of Russian Propaganda in this forum?
I strongly suspect that some of the reports @FCfromSSC has been getting (he is, after all, a mod) rely on the reasoning that certain positions regarding Ukraine policy, including his own, in and of themselves constitute Russian propaganda and/or are only espoused by Russian propagandists. He is presumably intending to challenge that reasoning by attempting to derive those positions de novo, via the Socratic method.
The Socratic method is kind of notorious for its low success rate, though, with perhaps an exception for a captive audience.
For the benefit of people who don't feel like reading a 90-page document in order to continue the conversation, relevant portions (that aren't censored):
- the CPC ran an online campaign trying to get Canadians to not vote for the Conservative Party
- the CPC tried to "make an example" of a Chinese-Canadian Conservative MP in an unspecified fashion
- the CPC took extreme steps to get Han Dong elected, including bussing in ineligible Chinese students to vote for him in (Liberal Party) preselection on threat of their visas being cancelled
- the CPC interfered in leadership selection of the Conservative Party
- the Conservative Party complained that the team trying to protect the election from interference wasn't taking their concerns seriously
- the intelligence services wanted to brief all MPs regarding foreign interference but the Trudeau government ignored them
- the report accuses the Trudeau government of requesting that intelligence reports be withdrawn due to being too politically sensitive
- foreign interference doesn't appear to have affected the overall outcome of the 2021 election but does seem to have had some more local impact
NB: I am strongly of the opinion that @jkf should have provided such a digest himself.
Probably to sweeten the package it would have to make some painful concessions to China on disputed islands like Scarborough Shoal, but it could potential walk away with robust guarantees of long-term functional autonomy and non-interference, conditional on remaining neutral.
The CPC can be modelled as Darth Vader; due to their narrative that the international order is unfair to China, they consider themselves quite morally justified in "altering the deal" whenever they think the previous deal wasn't ideal for them. Any agreement signed with them can be presumed to be written on water; they'll honour it as long as they're getting something out of it, but if circumstances change they'll renege. Hong Kong's the obvious example.
This is not to say that there aren't situations in which you have a selfish motivation to stay out of it anyway, but trusting their word is a pure mistake.
Do you predict otherwise, or do you simply write them off as too biased to report their own happiness level accurately?
The second one might show up if people get wind of exactly what the study's measuring (albeit on both sides, to some degree) due to wanting to make the result friendly to one's tribe (note the third and final section of this), but if you can keep that under wraps I wouldn't expect it. The first one is more complicated.
The first snag here is that there are some differences between the control group that you're positing and the control group we previously assumed. Detransitioners are generally going to be pretty miserable, I would assume, but detransitioners don't exist in trans-rejection-world because they were never encouraged/allowed to transition in the first place (indeed, this is the most inarguable point in favour of trans-rejection-world) so they should be part of the experimental group, not the control group. Also, we're talking here about potential social contagion effects, so a bunch of the people in trans-rejection-world who would have gone trans in current-world would potentially have looked like "people who never considered transition".
The second snag is with selection bias in your experimental group. Super-trans-friendly communities in current-world are going to attract those that think transition was the best thing that ever happened to them, and repel detransitioners and ex-trans (who tend to wind up places like here; hi, I'm ex-trans although I don't quite count as a detransitioner). They're also, for obvious reasons, not going to include anybody who committed suicide, although that one does apply to the control group as well to an unknown extent.
The third snag is that because of the fairly-recent explosion in transitions, most transsexuals are fairly recent; regret's not always immediate. Whether they're happy now is only part of what we want to know; we care about whether they're happy over their whole lifetimes.
Hence, I think that the experimental and control groups you propose would probably have the result you think, but that that doesn't prove much about the question we actually want answers to. A better experiment would be to take 10,000 people who present with GID, transition half, stop the other half from transitioning, and compare outcomes 40 years later. This one still isn't perfect, but it avoids literal survivorship bias and selection bias, puts detransitioners in the correct group, and follows up. Unfortunately, we can't currently run that better experiment, because the trans movement would consider the experiment an outrage and either get it cancelled, or evacuate the control group to trans-friendlier jurisdictions. It should have been run before transition therapy was allowed to become standard practice, but AFAIK it was not.
Pokemon is such an easy game that it can conceivably be beaten with entirely random inputs, and provably beaten by very-close-to random inputs. It's the ideal case for a video game that a primitive general intelligence would be good at. It does not require reactions or timing, it has very limited controls and interactions, and being incredibly slow and persistent gradually makes the only challenge easier as you inevitably outlevel everything from blundering around in the tall grass for too long. Twitch Plays Pokemon was essentially built on this premise.
Twitch Plays Pokemon only won because of a couple of rule changes that put the lie to this claim.
- The key combination to reset the GameBoy in a way that allowed deleting the save file was disabled.
- During certain segments, the game was set to "Democracy", in which only the most common inputs were actually accepted.
Without #1 they would have "lost" over and over again. Without #2 several lategame dungeons (one of which, Victory Road, is required) are nearly impossible due to rock-pushing and ledge-jumping puzzles where a decent chunk of progress gets wiped out by an errant input. I'm not even sure that you'd win eventually with 100% probability with random inputs (which is ordinarily the case), because I'm not 100% sure whether it's possible to corrupt the game permanently with some of the crazier glitches in RBY (I know it's possible to do arbitrary code execution, just not whether the hardware actually has the capability to ruin the cartridge permanently).
From what you mention, I think the problem is likely that the "knowledge base" isn't actually working nearly as well as a human's memory, so Claude is effectively amnesic; "when experience is not retained, as among savages, infancy is perpetual. Those who cannot remember the past are condemned to repeat it". This is known to be fixable (if probably requiring at least fine-tuning).
So yes, human agency involves a bunch of different capabilities and you need all of them at some level to be able to function (I don't think a human anterograde amnesiac would do very well at this task either, though one that had already learned how to take notes effectively might muddle through). This particular AI skimped on one of them, because it's not as useful to the task it was designed for, and that's crippling its capabilities at this task to levels well below its underlying reasoning ability.
Take out social stigma, and I think transition, on balance, is a very positive experience for the average person who tries it.
There are lots of ways to re-invent yourself. Most of those ways don't sterilise you, don't frequently render you permanently unattractive, don't have side-effects lists including suicide (yes, suicide; tampering with sex hormones can do that, which is why we try to avoid doing it unless necessary), and/or don't leave you dependent on pharmaceuticals for the rest of your life. Admittedly, there are quite a few that can do one of those (tattoos, joining a cult, wild orgies without protection), but not many that do all four.
I think it's a massive reach to take "real outcomes are maybe a smidgen better than refusing transition i.e. utterly awful" and then assume that if we sent the transphobes to the corn field it'd be actively good. Transition belongs in the "cost" column, not the "benefit" column.
Unless I'm missing something, we're talking about advocacy of assassination, not ITG threats of assassination. If you write a post telling people to do X, there's a possibility that people might read your post and do X. This is unfortunately true even if you weren't being serious. Hence, it is not unreasonable to be concerned about a high volume of people advocating X when X is "assassinating a high-ranked government official".
It is also not unreasonable to be concerned about the possibility of the culture war boiling over into mass violence in the USA, regardless of one's side in that war (if any). I'm thus not seeing the relevance of the tu quoque; I'm trying to get a better read on P(Boogaloo), not trying to score points.
More options
Context Copy link