I'm looking for an NLP program that can match pronouns in a sentence to the noun they refer to. I don't need anything else, just that. Does anyone know if there's one available that's not too expensive?
I knew Ukraine wasn't about to lose any major cities, but after hearing bad news about Ukraine for awhile, my confidence that they would take back territory in the next year (on net) fell to 60%. Now it's back to ~88%. [edit: err... make that 80%]
Why? the last two weeks of tweets & retweets by @TrentTelenko, who predicted both that Russia would invade #Ukraine and fail to take over the whole country, and who knows tons of military trivia. I don't know if Trent is holding back any info that reduces Ukraine's chances, but he offers quite a barrage of good signs.
Here I focus on facts I didn't know before, but at the end I'll have some bonus facts that are also instructive.
*Edit* (June 10): very bad news today lowers my confidence all the way down to 60%[1].
- Larger quantities of artillery shells are being sent to Ukraine than I thought: https://twitter.com/Osinttechnical/status/1530038089799946240 Edit: somehow I misinterpreted the number "208,000 155mm rounds allocated for Ukraine" from the U.S.; the number felt bigger than it is, though still higher than I knew about before. Google/Bing wouldn't tell me how many shells Russia is using, but I seem to recall somebody claiming Russia uses 30,000 shells per day and Denys Davydov saying Russia fires 20 shells for every shell fired by Ukraine. It should be noted though that much of Russia's artillery is imprecise and slow to configure (basically 'Uragan and Smerch' is lousy, see https://twitter.com/noclador/status/1531012132975910912, and by contrast I've seen some impressive videos of precise Ukrainian artillery). It seems like the Russians have just dumped a lot of their shells indiscriminately on cities until Ukraine has "nothing left to defend", and the spread-out shelling patterns on Ukrainian trenches also demonstrate low precision. As Small Wars Journal noted in 2020 (discussing 2015), "Ukrainians claimed that for every salvo they fired, they received 10 to 15 salvos in return". So it's not entirely clear that Ukraine will get as much gear as they need, and I have to suspect I'm missing some info that is favorable to Russia. On the other hand Russia doesn't seem to use its artillery effectively and I don't think I weighted partisan/guerilla fighting enough. Overall I'm reducing my confidence to 80%.
- Early signs Russia is running low on equipment seem to have been confirmed by more recent signs (the partial retaking of Severodonetsk and Zelensky's visit to Lysychansk) that Ukrainians in Donbas are not on the verge of collapse: https://twitter.com/PhillipsPOBrien/status/1530828743832440832 (that's in addition to Russia's deployment of old T-62s and Ukraine knowing about this in advance, as if you needed more confirmation of their good intel https://twitter.com/ChrisO_wiki/status/1529517640053661699) BTW: Russia has lost far more tanks than heavy artillery, but artillery is their weapon of choice. However, when RU runs short on tanks and BMPs, I expect it'll be easier for Ukrainian footsoldiers to advance despite that artillery, especially as their numbers increase.
- *Before* Ukraine's apparent offensive north of Kherson stalled out, @TrentTelenko judged it to be a "shaping operation" rather than a real offensive, which would make the "stall" something to be expected. https://twitter.com/TrentTelenko/status/1531814609908383744
- Russian losses should mainly be infantry, and their losses are so large that Trent thinks they've "hit a wall"—their infantry is nearly gone after accounting for both dead and wounded. Supporting evidence: RU deployed 25 BTGs to Severodonetsk and is still failing there. https://twitter.com/TrentTelenko/status/1533491373604802561
There is one thing I don't get about Trent: he evidently takes the Ukrainian claim for Russian troops KIA (30,500 dead) at face value, to reach a conclusion that Russia has taken 79,000 casualties with 80% being infantry. I'm not sure why one would trust that 30K number, but on the flip side he uses a low ratio of wounded to killed on the RU side, as compared to what is typical in wars, so maybe it balances out, almost.
[1] June 10: the reported casualty rate has risen to ~150 Ukrainian soldiers killed per day. This is probably higher than Russia's casualty rate and certainly higher than the rate of losses for Russian nationals. Since Ukraine is mostly in a defensive posture, this should not be happening. In fact, I've decided that even though I earn minimum wage (and can't get a tax deduction for Ukrainian charities), I will donate $1000 today. https://www.theguardian.com/world/2022/jun/10/ukraine-casualty-rate-russia-war-tipping-point ... June 30: I suspect Ukraine's high casualty rate is related to the continuing artillery imbalance, in which Russia just has far more shells with which to bomb Ukraine. I can only hope that Ukraine's rapid failure around Lysychansk is more about Ukraine trying to reduce casualties than about Ukrainian forces "collapsing".
As you may or may not be aware, there's currently a market on Manifold for whether anyone will take you up on your ban appeals process. As prediction markets tend to do, this creates an incentive for someone to get banned from your blog.
You're one of my favorite writers, so I'd prefer to not actually say anything inappropriate in your comments section, and would appreciate it if you just banned me without that being necessary. :)
In a sense, Michael Shellenberger getting 95000 votes (with only "50%" of ballots in) is a triumph. Jesus only needed twelve, Shellenberger has literally thousands of times more support.
In a more practical sense, fourth place is the second loser. And 3% of the vote is a country mile away from the 15% obtained by the second-place finisher.
Thomas Carlyle, who in life was extremely dismissive of Spirtualism, was made to go through his paces in the afterlife, where he had completely changed his views - according to this book, "What Spiritualism Really Is, By Thomas Carlyle In the Spirit-World, And Through the Impressional Brain of Dr. WM. J. Bryan" written by a doctor in 1919.
Apart from this example of how you can be made to say anything once you're dead, it's why the starry-eyed ideals around AI and Transhumanism don't impress me that much. Spirtualism, too, denied it was a religion; it was nothing like those musty, hoary old relics. You didn't have to take anything on faith. Spiritualism could be examined and tested, it produced results, and reproducible results. Eminent and respectable people like doctors and scientists backed it. This wasn't pie-in-the-sky based on texts from thousands of years ago, you too could go to a séance today and have independent confirmation of contact with the dead.
So faint demurrals of "but we're nothing like a religion, we never demand you take anything on faith!" are not that convincing to me about the latest New Scientific Wonderment.
(Belated) Congratulations to Scott on this title, which was apparently fully decodable (once I found the referenced piece) without reading the article:
"Current Affairs' Marxist Critique Of Toddler Show Blippi Isn't Marxist Enough"
(My reading: the anti-industrialism and the white cringe strike me as fundamentally at odds with Marxist thought, suggesting that Robinson is just name-dropping Marx for leftist cred rather than actually being Marxist.)
Has anyone here read Erik Hoel's The Revelations? I found Kierk Soren to be a really interesting character. I'd like book recommendations with main characters like Kierk Soren.
One more thing, any advise and book recommendations for a 23 year old that's about to embark on his intellectual journey? Thanks.
I have a question about where the reasonable deductions that can be made about how AI inputs are related to outputs ends and the "black box" of artificial intelligence begins. I often see posts about AI where the author sort of resigns themself to believing that AI is an impenetrable black box while also making reasonable deductions about why it is behaving in a certain way. I would love some clarity/links to papers about why the consensus is that most of what is happening is opaque and not something that can be understood through the development of the field. I would also love links to any papers or books about like meta-AI studies. I don't mean Meta the company but rather the AI equivalent to meta-physics.
My wife was recently diagnosed with gestational diabetes and put on a low glycemic index diet. As far as I can tell the major risk of gestational diabetes is high birth weight but the literature all seems to suggest that high birth weight is a positive predictor of outcomes such as IQ all the way up to about 12lbs (our baby is only predicted to be at ~66th percentile of birth weight which is far below that). Is there a reason why you should want to treat the gestational diabetes and reduce the birth weight or is this just a way to make the obstetrician's life easier during the delivery?
I wonder how we changed psychologically in the last hundred years or so due to mass empathy, caused by mass media. Before radio, TV and internet, the only emotions we came into contact with were our own and those of the people around us. Of course there were books and group gatherings, but nothing like now, where you get bombarded with other people's - real and fictional - emotions. And those generate empathy; a scary movie wouldn't be scary without empathy. So all day I 'feel' other people's emotions, from my friends on Facebook to fictional drug dealers in the Ozarks. Does anybody know if there has been anything written about how that changed us?
Say that I'm a woman on the curvier side. I'm happy with my breasts and hips but would like to slim down my waist a bit once or twice a month during parties and formal events. It looks like shapewear/corsets would be the thing I need, but it's a jungle out there. There's plenty of low-cost, low-quality (I assume) trash. The most vocal corset people seem to be BDSM cultists, not average gals that want to look a bit thinner sometimes. Thus me asking here.
Does anyone have a good beginners guide to shapewear? What are some quality brands that I can trust? Should I just bite the bullet and get a >$400 corset (money is not an issue anyway) and try to avoid getting sucked in by the vortex of "tight lacing" and "waist training" that seem to make up 95% of online corset discussion?
For the last two years, I've been doubting myself and spending a ridiculous amount of time finding the most optimal career path for me. My conclusion is that all roads lead to Software Engineering.
I plan to work on AI safety in the long term, but first, I need to start my career and focus on earning enough financial freedom.
The fastest route to getting a well-paid job in the IT field is learning Full Stack Web Development, so I'm doing that through The Odin Project.
About me: 123 IQ score (Raven's 2). Asperger's. 99th percentile for Extraversion, 98th Openness, 31st Conscientiousness (understandmyself.com). English level C1.
Who’s tried Kalshi, the new prediction market using dollars? It looks like they devoted the time and effort to get regulatory approval. I’m enjoying it somewhat, although I find the set of actual markets a bit dull (e.g., the closing value of the S&P 500, inflation, Biden approval ratings).
Can we get a SlateStarCodex deep dive into this? Collab or otherwise? I wanna see the data for the top 10/20/40/80 instead of just top 20. Maybe genre breakdowns?
Perhaps this is not the community for it. But if you are feeling like something is missing in your life, I recently converted from atheism to Catholicism, and I am happy to answer questions if anyone is interested in learning more about it.
Not trying to be a spoilsport, but wouldn't it be more effective to make a ACX goodreads group or something instead of this new wheel?
Currently reading Ubik which I think is fun, before that Gomorrah which is one of those books I got stuck on since it's just very depressing and even though it's not hard to read or very long it took me a month, and before that The Call: Inside the Global Saudi Religious Project by Krithika Varagur which I thought was very interesting and maybe worthy of a post by Scott.
Gladio has become such a meme but I've been very very slowly working on something about Dutch Gladio over the last three years (meaning reading some books about it and trying to get like 2 official documents) and The Call was really refreshing in 'oh yeah of course this is constantly happening everywhere!'
How strong is the evidence for the efficacy of EMDR? My wife's counselor suggested it to her, and from reading the description it seems...uh...dubious.
I’m looking to generate some conceptual trolley problems for a project I’m working on. For example: A runaway trolley will obliterate either Sexism or Racism. Which track to do you put it on? Etc.
Give me ideas of things this trolley might obliterate: good, bad or mixed. No living beings as trolley targets, please. Extra points for the absurd and/or thought-provoking. Thanks!
Someone I know with a background in computational linguistics and a degree in social work needs a job quickly. She is German and wants to stay in the US. Where should I tell her to look? I feel like someone in EA would have some ideas for this person.
Any medical students/ physicians here? I was thinking of starting a Whatsapp group where we could discuss studies and/or the practicalities of the doctor life.
If so, text me at 6098240815899 without the 8s. :)
I wrote the comment in #2 about the DSL test being incorrect. I want to point out that I don't think this statistical issue is actually particularly relevant to the debate at hand. I think basically everyone agrees that there is a strong statistical effect here, and the question of "how strong exactly" is mostly moot.
Scott's point was that there is an effect, and that effect is caused by coinfection with Strongyloides worms. Ivmmeta/Alexandros's believes it has a separate and distinct effect on covid itself. Alexandros/ivmmeta seem to have done some work to try to statistically distinguish these things by looking at infection progression and other metrics, an approach that seems superficially plausible to me. That is, plausible that you could separate these things in this way, in principle.
It seems to me like that's where the locus of the debate should be. It is still important to correct the statistical issue just as a matter of good hygiene, and Scott's use of the t-test was definitely quite inappropriate there, so that's good. I just want to point out that no matter what the outcome of this issue is, it's mostly immaterial to the broader question re: Ivermectin/COVID.
Logic/Bayes question: Suppose there is an open question about an underlying fact in a certain domain, and I have a model of how things work which predicts that an observation O is moderately more likely given hypothesis H1 than H2. You have a model which predicts the opposite. We have made observation O, update our conditional probabilities in opposite directions, and then learn that H2 is in fact correct. I should now take your underlying model a bit more seriously than my own, because you said that the observation was evidence for what turned out to be the correct hypothesis while I said the opposite.
This is a sort of meta-Bayesian reasoning, which can help us favor accurate models over inaccurate ones. But obviously it's going to produce updates strictly smaller than the update within the favored model - if your model says something is barely evidence at all, then the confirmation is barely evidence of your model.
Now, some questions: first, can we quantify these meta-Bayesian updates, and if so, how messy is the formula? Second, can we construct any scenario in which we would go up another level and have such a concept as meaningful meta-meta-Bayesian updates?
Is this summary of the US vs. European collegiate systems correct? The European college academic experience is much more focused on one's specific major, and does not feature a lot of required classes outside of it. The US college academic experience involves a broader, not deeper education, with required extracurriculars in classes outside of your major. I.e. if you study Chemistry, Comp Sci or Literature in Europe you mostly just take classes in those fields, whereas in the US you're required to take a number of extracurriculars, to be a 'broader' thinker. Thus, American students are not studying their primary subject as intensely. Do I have that right?
(This isn't even getting into medical or law school, which in the US is another program after one's Bachelor, and in Europe I believe rarely or never is).
If I do have that right- are there any advantages to the US system? I've heard many times, from many different people, that European students in a given subject are basically always ahead of American ones at the same point, given the deeper/broader dichotomy. So- why does the US structure its educational system this way? America is famously the world's #1 technological, financial, military, economic, scientific and cultural power at the moment, so I find it a bit difficult to believe the US academic system is markedly worse than Europe's. Do American college students somehow gain something substantive by being forced to take classes outside their major? (Also, how does the rest of the developed world like Canada, Japan, South Korea, Taiwan, Israel etc. do it?)
Edit to include: also, in theory, a college could offer this more intensive/less broad education program here in the US, right? America has tons of colleges & federalism- kind of surprising to see that at least a few colleges don't offer the European system instead, right?
The widely used term 'culture war' is a translation of Kulturkampf, a 19th-century political dispute between German chancellor Bismarck and the Catholic Church. But now we use it without any memory of that, as far as I can tell.
It uses 'culture' more narrowly and 'war' more broadly than I'm comfortable with. The narrow use of 'culture' obscures the multitude of values and habits on which partisan rivals tend to agree (cf. Anthropology 101). The broad use of 'war' trivializes the death and destruction in actual war.
The first fairly thorough alternative I've come up with is 'disputes over identity, respect, education, bodily autonomy, and diversity'. For short, you get 'DIREBAD' or 'IREBAD disputes'. Memorable, but also somewhat childish.
For those of you who have an arts and humanities streak but who've gone into business/science how do you keep up with your more artistic/humanistic pursuits? I find myself regretting not keeping up with my more artistic/humanistic interests. But the opportunities seem to require either more commitment than a working person can give or to be extremely amateur to the point many people have little to no experience. Does anyone have experience with striking that balance? Anyone have advice?
PS: I guess this is technically a request. But I don't think it degrades the discourse too much for an open thread.
I saw a headline about monkey pox being seen in Pennsylvania. Pennsylvania is quite a large state, and I don't think a case of disease here is all that meaningful.
What would be a better size/system for announcing what's going on, good or bad.
Chunk the world by population? Watersheds? Transportation networks? Different systems for different topics?
Monkey pox has since been seen in Philadelphia-- not good news, but a little more meaningful.
Why the so-called hard problem of consciousness is nonsense
I'd been thinking of writing a post on the subject – and maybe I will. But now I want to post bits of a post written by Keith Frankish, https://www.keithfrankish.com/2022/05/like-a-rainbow/. His 2nd and 3rd paragraphs:
"Here’s one: rainbows. Rainbows are real, aren’t they? You can see them with your own eyes — though you have to be in the right position, with the sun behind you. You can point them out to other people — provided they take up a similar position to you. Heck, you can even photograph them."
"But what exactly is it that’s real? It seems as if there’s an actual gauzy, multi-coloured arc stretching across the sky and curving down to meet the ground at a point to which you could walk. Our ancestors may have thought rainbows were like that. We know better, of course. There’s no real coloured arc up there. Nor are there any specific physical features arranged arcwise — the rainbow’s “atmospheric correlates”, as it were. There are just water droplets evenly distributed throughout the air and reflecting sunlight in such a way that from your vantage point there appears to be a multi-coloured arc."
That's it, really. An analogy to be sure, but a good one. He continues:
"When I reflect on my own experience, it seems to me that my consciousness is an inner world, where the world around me is rendered in private mental qualities — “qualia” — for my benefit alone. But there isn’t such a world. Neuroscience finds nothing like it in the brain, nor even anything isomorphic to it. Rather, it finds complex trains of neural activity proceeding in parallel and triggering a host of reactions — physiological, psychological, and behavioural. My sense of having a rich qualia-filled inner world is an impression created by all these processes, but the processes themselves are as different from the supposed inner world as a moisture-infused mass of air is from a colourful aerial arc."
That is, it's like rainbows. Perfectly real, but not in the most obvious way.
It's like, when Chalmers posited the "hard problem" he asks us to imagine a complex circuit diagram that accounts for everything neuroscience has to say about consciousness. That's the "easy problem," solved!
"Now," he asks, "see consciousness anywhere in that diagram?"
"No", we reply.
"That's the hard problem," and adds another box to the diagram. "Find that box, and you've solved the hard problem," says he.
No. That's just make-work for philosophers, like digging ditches and then filling them back in.
Lots of interesting points, first person I've seen accurately portray my feelings about the aliens in public health. Speaking of injecting malevolence into health decisions.
What about mandates? I was terminated from a remote position because my organization decided that they wanted to force the decision on me, due to this I am denied any unemployment benefits that I have paid into.
If your goal was to make these things appear malevolent (justifiably or not) wouldn't you use them as a cudgel to purge newfound undesirables from society?
It seems to be the case that the bar to involuntarily confine someone experiencing mental illness (whether in a hospital or jail) is extremely low -- as soon as they do something moderately disruptive in public any cop can make this happen if they feel like it.
But the bar to involuntarily treat someone with antipsychotics is way higher, as far as I can tell. If someone says "no", then you have to go get a court order, and it's often only allowed for people who are already committed in state hospitals.
Involuntary outpatient commitment theoretically exists but seems not to happen much and also seems to be institutionally complicated. Certainly more complicated than "arrest -> charges -> no bail -> indefinite jail stay while refusing treatment".
Obviously it would be better for people to enjoy freedom and have their symptoms treated enough to function, vs. having severe symptoms while locked in a jail or hospital. So I wonder why the rules on this seem to be backwards, and how it happened.
The book review thing is awesome, and something i expect to see more of: communities using computers to communicate the community's values, and individuals using computers to help them make choices more in line with their own values.
https://francisquinn.substack.com/p/newcombs-paradox-and-metaphysics?s=w
I'm looking for an NLP program that can match pronouns in a sentence to the noun they refer to. I don't need anything else, just that. Does anyone know if there's one available that's not too expensive?
I knew Ukraine wasn't about to lose any major cities, but after hearing bad news about Ukraine for awhile, my confidence that they would take back territory in the next year (on net) fell to 60%. Now it's back to ~88%. [edit: err... make that 80%]
Why? the last two weeks of tweets & retweets by @TrentTelenko, who predicted both that Russia would invade #Ukraine and fail to take over the whole country, and who knows tons of military trivia. I don't know if Trent is holding back any info that reduces Ukraine's chances, but he offers quite a barrage of good signs.
Here I focus on facts I didn't know before, but at the end I'll have some bonus facts that are also instructive.
*Edit* (June 10): very bad news today lowers my confidence all the way down to 60%[1].
- Key M270 and M142 MLRS systems (which can threaten Russian supply lines to Izyum) are being sent to Ukraine (I heard this already but it's good to have confirmation, plus Trent says it's a "game changer" https://twitter.com/TrentTelenko/status/1531822945911328768): https://twitter.com/noclador/status/1531012132975910912
-- Ukraine is even getting quality stuff from Germany: https://twitter.com/TrentTelenko/status/1533576265847955456 (Germany still drags its heels though https://twitter.com/visegrad24/status/1534615136467550209)
-- "80 Russian artillery pieces have been destroyed with less than 12 Caesar SPH in a few days" https://twitter.com/TrentTelenko/status/1534291040357830656
-- Spain to supply 40 KMW Leopard 2A4 MBTs: https://twitter.com/TrentTelenko/status/1534341808687308800
. -- Update: Ukraine WON'T get enough ammo for their new long-range GMLRSs: https://twitter.com/TrentTelenko/status/1537131753441214465
- Larger quantities of artillery shells are being sent to Ukraine than I thought: https://twitter.com/Osinttechnical/status/1530038089799946240 Edit: somehow I misinterpreted the number "208,000 155mm rounds allocated for Ukraine" from the U.S.; the number felt bigger than it is, though still higher than I knew about before. Google/Bing wouldn't tell me how many shells Russia is using, but I seem to recall somebody claiming Russia uses 30,000 shells per day and Denys Davydov saying Russia fires 20 shells for every shell fired by Ukraine. It should be noted though that much of Russia's artillery is imprecise and slow to configure (basically 'Uragan and Smerch' is lousy, see https://twitter.com/noclador/status/1531012132975910912, and by contrast I've seen some impressive videos of precise Ukrainian artillery). It seems like the Russians have just dumped a lot of their shells indiscriminately on cities until Ukraine has "nothing left to defend", and the spread-out shelling patterns on Ukrainian trenches also demonstrate low precision. As Small Wars Journal noted in 2020 (discussing 2015), "Ukrainians claimed that for every salvo they fired, they received 10 to 15 salvos in return". So it's not entirely clear that Ukraine will get as much gear as they need, and I have to suspect I'm missing some info that is favorable to Russia. On the other hand Russia doesn't seem to use its artillery effectively and I don't think I weighted partisan/guerilla fighting enough. Overall I'm reducing my confidence to 80%.
- Early signs Russia is running low on equipment seem to have been confirmed by more recent signs (the partial retaking of Severodonetsk and Zelensky's visit to Lysychansk) that Ukrainians in Donbas are not on the verge of collapse: https://twitter.com/PhillipsPOBrien/status/1530828743832440832 (that's in addition to Russia's deployment of old T-62s and Ukraine knowing about this in advance, as if you needed more confirmation of their good intel https://twitter.com/ChrisO_wiki/status/1529517640053661699) BTW: Russia has lost far more tanks than heavy artillery, but artillery is their weapon of choice. However, when RU runs short on tanks and BMPs, I expect it'll be easier for Ukrainian footsoldiers to advance despite that artillery, especially as their numbers increase.
- There are indications Russian casualty rates are higher than Ukraine while Ukraine mobilizes faster: https://twitter.com/TrentTelenko/status/1532823278699663360
- Russian artillery is manpower-heavy, lowering Russian efficiency: https://twitter.com/noclador/status/1528024733983424512 (see also https://twitter.com/TrentTelenko/status/1530567969889300482)
- Unconfirmed good news for Ukraine https://twitter.com/TrentTelenko/status/1530027464990277650
- Unconfirmed advanced cancer and assassination plot for Putin: https://twitter.com/Archer83Able/status/1532367395964309504
- Apparent Russian tank losses are confirmed accurate: https://twitter.com/partizan_oleg/status/1526199389764874240 (a loss rate dramatically higher than that of Ukraine https://twitter.com/Lee__Drake/status/1529870039167315969)
- *Before* Ukraine's apparent offensive north of Kherson stalled out, @TrentTelenko judged it to be a "shaping operation" rather than a real offensive, which would make the "stall" something to be expected. https://twitter.com/TrentTelenko/status/1531814609908383744
- Russian losses should mainly be infantry, and their losses are so large that Trent thinks they've "hit a wall"—their infantry is nearly gone after accounting for both dead and wounded. Supporting evidence: RU deployed 25 BTGs to Severodonetsk and is still failing there. https://twitter.com/TrentTelenko/status/1533491373604802561
-- Related tweets on Severodonetsk: (1) UA tactical advantage: height https://twitter.com/Osinttechnical/status/1533249952809463809) (2) LNR is out of troops: https://twitter.com/TrentTelenko/status/1533556977674117122
-- Related 🧵: 'I think the" Culminating Point" for Russian offensive operations in Ukraine is almost upon us.' https://twitter.com/TrentTelenko/status/1534291035924340736
-- Related: "Ghost Troop" corruption means Russia had less soldiers than it thought https://twitter.com/TrentTelenko/status/1534700521633857536
-- Followup 🧵: https://twitter.com/TrentTelenko/status/1534700516843864065
- And some minor evidence.
-- Ukrainians are more motivated and produce new hardware that Russia possibly wouldn't (https://twitter.com/TrentTelenko/status/1533516137992007681 https://twitter.com/UAWeapons/status/1532816735996346370)
-- Russian 122mm MLRS destroyed close to the front (https://twitter.com/TrentTelenko/status/1533521893969612800)
-- Ukraine still attacking snake island (https://twitter.com/raging545/status/1534248225632358401)
- And a little counterevidence: https://twitter.com/TrentTelenko/status/1534412330330636289
- "Butcher of Syria" General Dvornikov seems to have been fired, suggesting that Putin was unhappy with Russia's performance and is trying a useless "Hail Mary" https://euroweeklynews.com/2022/06/05/putin-has-relieved-general-alexandr-dvornikov-as-commander-in-ukraine/
Plus here's some stuff I already knew:
- Russia still loses Generals ocasionally: https://twitter.com/RALee85/status/1533518405042589697 https://twitter.com/anders_aslund/status/1533523990312099845
- More bad tactics from Russia https://twitter.com/mdmitri91/status/1532689797269045248
- Some pro-Russian milbloggers are changing their tune https://twitter.com/mdmitri91/status/1533505067881533441
Bonuses:
- a long 🧵 on the history of Russian political and strategic goals in Ukraine: https://twitter.com/John_A_Ridge/status/1529262545521020928
- partisan/guerilla warfare is another point in Ukraine's favor, e.g. https://twitter.com/cliffordlevy/status/1533934971722817542
- the most important bonus: powerpoint videos by Perun: https://www.youtube.com/channel/UCC3ehuUksTyQ7bbjGntmx3Q/featured
- Forecasts from understandingwar.org have been pretty consistently optimistic for Ukraine, and pretty consistently correct too
- Fun fact: Russia is happy to starve third-worlders to death if it hurts Ukraine. Let's hope Ukraine wins quickly! https://twitter.com/vcdgf555/status/1533557914954178560
- Fun fact: visual representation of estimated density of Russian BTGs (via uawardata.com): https://twitter.com/HN_Schlottman/status/1533917743375859712
- Anyone know where to get English versions of these POW videos? The ones English ones I have seen have been fantastically entertaining. https://www.youtube.com/c/VolodymyrZolkin/videos
There is one thing I don't get about Trent: he evidently takes the Ukrainian claim for Russian troops KIA (30,500 dead) at face value, to reach a conclusion that Russia has taken 79,000 casualties with 80% being infantry. I'm not sure why one would trust that 30K number, but on the flip side he uses a low ratio of wounded to killed on the RU side, as compared to what is typical in wars, so maybe it balances out, almost.
[1] June 10: the reported casualty rate has risen to ~150 Ukrainian soldiers killed per day. This is probably higher than Russia's casualty rate and certainly higher than the rate of losses for Russian nationals. Since Ukraine is mostly in a defensive posture, this should not be happening. In fact, I've decided that even though I earn minimum wage (and can't get a tax deduction for Ukrainian charities), I will donate $1000 today. https://www.theguardian.com/world/2022/jun/10/ukraine-casualty-rate-russia-war-tipping-point ... June 30: I suspect Ukraine's high casualty rate is related to the continuing artillery imbalance, in which Russia just has far more shells with which to bomb Ukraine. I can only hope that Ukraine's rapid failure around Lysychansk is more about Ukraine trying to reduce casualties than about Ukrainian forces "collapsing".
What is "the AK47 of paintball guns" and what is "the AR15 of paintball guns"?
As you may or may not be aware, there's currently a market on Manifold for whether anyone will take you up on your ban appeals process. As prediction markets tend to do, this creates an incentive for someone to get banned from your blog.
You're one of my favorite writers, so I'd prefer to not actually say anything inappropriate in your comments section, and would appreciate it if you just banned me without that being necessary. :)
https://manifold.markets/EnopoletusHarding/will-so-much-as-a-single-banned-pos
In a sense, Michael Shellenberger getting 95000 votes (with only "50%" of ballots in) is a triumph. Jesus only needed twelve, Shellenberger has literally thousands of times more support.
In a more practical sense, fourth place is the second loser. And 3% of the vote is a country mile away from the 15% obtained by the second-place finisher.
We've had a discussion before about spiritualism, so here is an article from the Public Domain Review:
https://publicdomainreview.org/collection/what-spiritualism-really-is
Thomas Carlyle, who in life was extremely dismissive of Spirtualism, was made to go through his paces in the afterlife, where he had completely changed his views - according to this book, "What Spiritualism Really Is, By Thomas Carlyle In the Spirit-World, And Through the Impressional Brain of Dr. WM. J. Bryan" written by a doctor in 1919.
Apart from this example of how you can be made to say anything once you're dead, it's why the starry-eyed ideals around AI and Transhumanism don't impress me that much. Spirtualism, too, denied it was a religion; it was nothing like those musty, hoary old relics. You didn't have to take anything on faith. Spiritualism could be examined and tested, it produced results, and reproducible results. Eminent and respectable people like doctors and scientists backed it. This wasn't pie-in-the-sky based on texts from thousands of years ago, you too could go to a séance today and have independent confirmation of contact with the dead.
So faint demurrals of "but we're nothing like a religion, we never demand you take anything on faith!" are not that convincing to me about the latest New Scientific Wonderment.
Why don’t you ask Gideon Meyerowitz-Katz which test is best to do on the Ivermectin data then do that test?
(Belated) Congratulations to Scott on this title, which was apparently fully decodable (once I found the referenced piece) without reading the article:
"Current Affairs' Marxist Critique Of Toddler Show Blippi Isn't Marxist Enough"
(My reading: the anti-industrialism and the white cringe strike me as fundamentally at odds with Marxist thought, suggesting that Robinson is just name-dropping Marx for leftist cred rather than actually being Marxist.)
What's the current status of abortions in Texas? Are abortions still being performed? Are bounties being paid out?
> the studies aren’t homogenous
Not homogenous, "from the same origin", or not homogeneous, "of the same kind"?
After a decade of being glasses-free after LASIK, I need glasses again (I can still see 20/20, but I need prism correction and blue-light blocking).
It is strange that I have forgotten things like "how do I keep glasses clean".
How do I keep glasses clean?
Scott, where did you get your characterization of Shellenberger?
From his Rogan clip?
That's not the last word, but neither is it an error.
Hello everyone.
Has anyone here read Erik Hoel's The Revelations? I found Kierk Soren to be a really interesting character. I'd like book recommendations with main characters like Kierk Soren.
One more thing, any advise and book recommendations for a 23 year old that's about to embark on his intellectual journey? Thanks.
I have a question about where the reasonable deductions that can be made about how AI inputs are related to outputs ends and the "black box" of artificial intelligence begins. I often see posts about AI where the author sort of resigns themself to believing that AI is an impenetrable black box while also making reasonable deductions about why it is behaving in a certain way. I would love some clarity/links to papers about why the consensus is that most of what is happening is opaque and not something that can be understood through the development of the field. I would also love links to any papers or books about like meta-AI studies. I don't mean Meta the company but rather the AI equivalent to meta-physics.
Any fellow Freemasons here? Or anyone involved in civic/service/fraternal orders or clubs? Would love to connect!
Awakening the Ancestors - a sci-fan-fic pastoral pastiche in 7 movements, submitted for EA Post Prize #1: The World in 2072 (it didn't win) - how many references can you spot? https://pathfindings.substack.com/p/awakening-the-ancestors?s=w
Pregnancy advice question.
My wife was recently diagnosed with gestational diabetes and put on a low glycemic index diet. As far as I can tell the major risk of gestational diabetes is high birth weight but the literature all seems to suggest that high birth weight is a positive predictor of outcomes such as IQ all the way up to about 12lbs (our baby is only predicted to be at ~66th percentile of birth weight which is far below that). Is there a reason why you should want to treat the gestational diabetes and reduce the birth weight or is this just a way to make the obstetrician's life easier during the delivery?
Several months ago there was a bounty announcement for a compilation of Robin Hansons best blog posts: https://www.lesswrong.com/posts/QaDwBio8MLqRvTREH/usd10k-bounty-read-and-compile-robin-hanson-s-best-posts
Does anybody know if that resulted in anything?
I wonder how we changed psychologically in the last hundred years or so due to mass empathy, caused by mass media. Before radio, TV and internet, the only emotions we came into contact with were our own and those of the people around us. Of course there were books and group gatherings, but nothing like now, where you get bombarded with other people's - real and fictional - emotions. And those generate empathy; a scary movie wouldn't be scary without empathy. So all day I 'feel' other people's emotions, from my friends on Facebook to fictional drug dealers in the Ozarks. Does anybody know if there has been anything written about how that changed us?
Say that I'm a woman on the curvier side. I'm happy with my breasts and hips but would like to slim down my waist a bit once or twice a month during parties and formal events. It looks like shapewear/corsets would be the thing I need, but it's a jungle out there. There's plenty of low-cost, low-quality (I assume) trash. The most vocal corset people seem to be BDSM cultists, not average gals that want to look a bit thinner sometimes. Thus me asking here.
Does anyone have a good beginners guide to shapewear? What are some quality brands that I can trust? Should I just bite the bullet and get a >$400 corset (money is not an issue anyway) and try to avoid getting sucked in by the vortex of "tight lacing" and "waist training" that seem to make up 95% of online corset discussion?
Hi! I'm 20 years old and from Argentina.
For the last two years, I've been doubting myself and spending a ridiculous amount of time finding the most optimal career path for me. My conclusion is that all roads lead to Software Engineering.
I plan to work on AI safety in the long term, but first, I need to start my career and focus on earning enough financial freedom.
The fastest route to getting a well-paid job in the IT field is learning Full Stack Web Development, so I'm doing that through The Odin Project.
About me: 123 IQ score (Raven's 2). Asperger's. 99th percentile for Extraversion, 98th Openness, 31st Conscientiousness (understandmyself.com). English level C1.
Any advice?
Who’s tried Kalshi, the new prediction market using dollars? It looks like they devoted the time and effort to get regulatory approval. I’m enjoying it somewhat, although I find the set of actual markets a bit dull (e.g., the closing value of the S&P 500, inflation, Biden approval ratings).
https://twitter.com/a_m_mastroianni/status/1521133090638282752
Can we get a SlateStarCodex deep dive into this? Collab or otherwise? I wanna see the data for the top 10/20/40/80 instead of just top 20. Maybe genre breakdowns?
Perhaps this is not the community for it. But if you are feeling like something is missing in your life, I recently converted from atheism to Catholicism, and I am happy to answer questions if anyone is interested in learning more about it.
Not trying to be a spoilsport, but wouldn't it be more effective to make a ACX goodreads group or something instead of this new wheel?
Currently reading Ubik which I think is fun, before that Gomorrah which is one of those books I got stuck on since it's just very depressing and even though it's not hard to read or very long it took me a month, and before that The Call: Inside the Global Saudi Religious Project by Krithika Varagur which I thought was very interesting and maybe worthy of a post by Scott.
Gladio has become such a meme but I've been very very slowly working on something about Dutch Gladio over the last three years (meaning reading some books about it and trying to get like 2 official documents) and The Call was really refreshing in 'oh yeah of course this is constantly happening everywhere!'
How strong is the evidence for the efficacy of EMDR? My wife's counselor suggested it to her, and from reading the description it seems...uh...dubious.
I’m looking to generate some conceptual trolley problems for a project I’m working on. For example: A runaway trolley will obliterate either Sexism or Racism. Which track to do you put it on? Etc.
Give me ideas of things this trolley might obliterate: good, bad or mixed. No living beings as trolley targets, please. Extra points for the absurd and/or thought-provoking. Thanks!
Someone I know with a background in computational linguistics and a degree in social work needs a job quickly. She is German and wants to stay in the US. Where should I tell her to look? I feel like someone in EA would have some ideas for this person.
Options for best budget software for personal use, and/or WFH independent contracting, and why that option works/what the drawbacks are?
Any medical students/ physicians here? I was thinking of starting a Whatsapp group where we could discuss studies and/or the practicalities of the doctor life.
If so, text me at 6098240815899 without the 8s. :)
Let me be the first to rush to accept John Schilling as our new space overlord 😁
I wrote the comment in #2 about the DSL test being incorrect. I want to point out that I don't think this statistical issue is actually particularly relevant to the debate at hand. I think basically everyone agrees that there is a strong statistical effect here, and the question of "how strong exactly" is mostly moot.
Scott's point was that there is an effect, and that effect is caused by coinfection with Strongyloides worms. Ivmmeta/Alexandros's believes it has a separate and distinct effect on covid itself. Alexandros/ivmmeta seem to have done some work to try to statistically distinguish these things by looking at infection progression and other metrics, an approach that seems superficially plausible to me. That is, plausible that you could separate these things in this way, in principle.
It seems to me like that's where the locus of the debate should be. It is still important to correct the statistical issue just as a matter of good hygiene, and Scott's use of the t-test was definitely quite inappropriate there, so that's good. I just want to point out that no matter what the outcome of this issue is, it's mostly immaterial to the broader question re: Ivermectin/COVID.
Logic/Bayes question: Suppose there is an open question about an underlying fact in a certain domain, and I have a model of how things work which predicts that an observation O is moderately more likely given hypothesis H1 than H2. You have a model which predicts the opposite. We have made observation O, update our conditional probabilities in opposite directions, and then learn that H2 is in fact correct. I should now take your underlying model a bit more seriously than my own, because you said that the observation was evidence for what turned out to be the correct hypothesis while I said the opposite.
This is a sort of meta-Bayesian reasoning, which can help us favor accurate models over inaccurate ones. But obviously it's going to produce updates strictly smaller than the update within the favored model - if your model says something is barely evidence at all, then the confirmation is barely evidence of your model.
Now, some questions: first, can we quantify these meta-Bayesian updates, and if so, how messy is the formula? Second, can we construct any scenario in which we would go up another level and have such a concept as meaningful meta-meta-Bayesian updates?
Is this summary of the US vs. European collegiate systems correct? The European college academic experience is much more focused on one's specific major, and does not feature a lot of required classes outside of it. The US college academic experience involves a broader, not deeper education, with required extracurriculars in classes outside of your major. I.e. if you study Chemistry, Comp Sci or Literature in Europe you mostly just take classes in those fields, whereas in the US you're required to take a number of extracurriculars, to be a 'broader' thinker. Thus, American students are not studying their primary subject as intensely. Do I have that right?
(This isn't even getting into medical or law school, which in the US is another program after one's Bachelor, and in Europe I believe rarely or never is).
If I do have that right- are there any advantages to the US system? I've heard many times, from many different people, that European students in a given subject are basically always ahead of American ones at the same point, given the deeper/broader dichotomy. So- why does the US structure its educational system this way? America is famously the world's #1 technological, financial, military, economic, scientific and cultural power at the moment, so I find it a bit difficult to believe the US academic system is markedly worse than Europe's. Do American college students somehow gain something substantive by being forced to take classes outside their major? (Also, how does the rest of the developed world like Canada, Japan, South Korea, Taiwan, Israel etc. do it?)
Edit to include: also, in theory, a college could offer this more intensive/less broad education program here in the US, right? America has tons of colleges & federalism- kind of surprising to see that at least a few colleges don't offer the European system instead, right?
The widely used term 'culture war' is a translation of Kulturkampf, a 19th-century political dispute between German chancellor Bismarck and the Catholic Church. But now we use it without any memory of that, as far as I can tell.
It uses 'culture' more narrowly and 'war' more broadly than I'm comfortable with. The narrow use of 'culture' obscures the multitude of values and habits on which partisan rivals tend to agree (cf. Anthropology 101). The broad use of 'war' trivializes the death and destruction in actual war.
The first fairly thorough alternative I've come up with is 'disputes over identity, respect, education, bodily autonomy, and diversity'. For short, you get 'DIREBAD' or 'IREBAD disputes'. Memorable, but also somewhat childish.
Are there better alternatives?
For those of you who have an arts and humanities streak but who've gone into business/science how do you keep up with your more artistic/humanistic pursuits? I find myself regretting not keeping up with my more artistic/humanistic interests. But the opportunities seem to require either more commitment than a working person can give or to be extremely amateur to the point many people have little to no experience. Does anyone have experience with striking that balance? Anyone have advice?
PS: I guess this is technically a request. But I don't think it degrades the discourse too much for an open thread.
I saw a headline about monkey pox being seen in Pennsylvania. Pennsylvania is quite a large state, and I don't think a case of disease here is all that meaningful.
What would be a better size/system for announcing what's going on, good or bad.
Chunk the world by population? Watersheds? Transportation networks? Different systems for different topics?
Monkey pox has since been seen in Philadelphia-- not good news, but a little more meaningful.
Why the so-called hard problem of consciousness is nonsense
I'd been thinking of writing a post on the subject – and maybe I will. But now I want to post bits of a post written by Keith Frankish, https://www.keithfrankish.com/2022/05/like-a-rainbow/. His 2nd and 3rd paragraphs:
"Here’s one: rainbows. Rainbows are real, aren’t they? You can see them with your own eyes — though you have to be in the right position, with the sun behind you. You can point them out to other people — provided they take up a similar position to you. Heck, you can even photograph them."
"But what exactly is it that’s real? It seems as if there’s an actual gauzy, multi-coloured arc stretching across the sky and curving down to meet the ground at a point to which you could walk. Our ancestors may have thought rainbows were like that. We know better, of course. There’s no real coloured arc up there. Nor are there any specific physical features arranged arcwise — the rainbow’s “atmospheric correlates”, as it were. There are just water droplets evenly distributed throughout the air and reflecting sunlight in such a way that from your vantage point there appears to be a multi-coloured arc."
That's it, really. An analogy to be sure, but a good one. He continues:
"When I reflect on my own experience, it seems to me that my consciousness is an inner world, where the world around me is rendered in private mental qualities — “qualia” — for my benefit alone. But there isn’t such a world. Neuroscience finds nothing like it in the brain, nor even anything isomorphic to it. Rather, it finds complex trains of neural activity proceeding in parallel and triggering a host of reactions — physiological, psychological, and behavioural. My sense of having a rich qualia-filled inner world is an impression created by all these processes, but the processes themselves are as different from the supposed inner world as a moisture-infused mass of air is from a colourful aerial arc."
That is, it's like rainbows. Perfectly real, but not in the most obvious way.
It's like, when Chalmers posited the "hard problem" he asks us to imagine a complex circuit diagram that accounts for everything neuroscience has to say about consciousness. That's the "easy problem," solved!
"Now," he asks, "see consciousness anywhere in that diagram?"
"No", we reply.
"That's the hard problem," and adds another box to the diagram. "Find that box, and you've solved the hard problem," says he.
No. That's just make-work for philosophers, like digging ditches and then filling them back in.
re: 2
Lots of interesting points, first person I've seen accurately portray my feelings about the aliens in public health. Speaking of injecting malevolence into health decisions.
What about mandates? I was terminated from a remote position because my organization decided that they wanted to force the decision on me, due to this I am denied any unemployment benefits that I have paid into.
If your goal was to make these things appear malevolent (justifiably or not) wouldn't you use them as a cudgel to purge newfound undesirables from society?
It seems to be the case that the bar to involuntarily confine someone experiencing mental illness (whether in a hospital or jail) is extremely low -- as soon as they do something moderately disruptive in public any cop can make this happen if they feel like it.
But the bar to involuntarily treat someone with antipsychotics is way higher, as far as I can tell. If someone says "no", then you have to go get a court order, and it's often only allowed for people who are already committed in state hospitals.
Involuntary outpatient commitment theoretically exists but seems not to happen much and also seems to be institutionally complicated. Certainly more complicated than "arrest -> charges -> no bail -> indefinite jail stay while refusing treatment".
Obviously it would be better for people to enjoy freedom and have their symptoms treated enough to function, vs. having severe symptoms while locked in a jail or hospital. So I wonder why the rules on this seem to be backwards, and how it happened.
The book review thing is awesome, and something i expect to see more of: communities using computers to communicate the community's values, and individuals using computers to help them make choices more in line with their own values.
A feature I notice about this "new" rationalism, or it may just be Scott, but the openness to being corrected is refreshing.