prev <-> next
[2020-04-21 14:12:32] - Pierce: I know that was (probably?) a joke, but I can't help but think that somebody with a decent amount of cash, a high tolerance for risk, and a bold streak could be making a killing somehow right now. -Paul
[2020-04-21 14:11:48] - a: Nah, I'm guessing most of those are algorithmically driven and thus not manually checked much. I've seen similar things on CNBC before. -Paul
[2020-04-21 14:10:35] - Xpovos: Besides the lack of adequate testing (and thus insufficient data) point, is the main takeaway that we're likely to see a lot more deaths attributed to COVID-19 retroactively? -Paul
[2020-04-21 14:09:13] - paul: s&p500 +900% overnight. is it nobodys job to check these numbers? it wouldn't be hard to spot this obviously-incorrect data. ~a
[2020-04-20 21:40:28] - new recovery plan: just pay everyone to babysit a few hundred barrels of oil - pierce
[2020-04-20 21:35:27] - daniel: i feel like if you had storage facilities, you could make tons of money just holding it for a month. ~a
[2020-04-20 21:35:14] - Ignore the headline, for the most part. There's some good data (or at least analysis of missing data) in there. -- Xpovos
[2020-04-20 21:34:51] - Paul: More COVID analysis. https://www.thedailybeast.com/the-covid-19-death-undercount-is-scarier-than-you-think -- Xpovos
[2020-04-20 21:34:48] - a: Yeah oil prices are bonkers right now. -Daniel
[2020-04-20 21:05:21] - paul/daniel: found this from reddit. yeah, that's right. -$38/share. whats crazy about this story isn't that negative dollars per share are totally a thing, but HOW negative. since a barrel of oil produces something like 20 gallons of gas, -38/share would translate to -$2.00/gallon, and that doesn't even include refining. yikes. ~a
[2020-04-20 21:04:36] - or maybe it's about presidential candidates, in which case trump is relevant and kavanaugh isn't. the only way for kavanaugh to be the most relevant focus would be if this was a story about media coverage of sexual assault allegations rather than of these particular allegations. but if the story isn't about these allegations, then how can we be reporting a story about how the media covered these allegations? - pierce
[2020-04-20 21:02:13] - pierce: i thought we (miguel and paul originally) were saying it was about reporting. so sexual assault in general. i think kavenaugh is relevant to the "reporting" narrative that paul and miguel were discussing. ~a
[2020-04-20 21:00:49] - a: some, sure. but you can also find differences. if this is just a story about reade and biden, the kavanaugh stuff is irrelevant. if it's a story about sexual assault in general, then kavanaugh stuff isn't especially relevant. if it's about sexual harassment/assault by your politician boss, then clinton would be relevant but not trump or kavanaugh (since they weren't politicians at the time). - pierce
[2020-04-20 20:38:20] - the NYT article*
[2020-04-20 20:38:10] - I thought the NYT was relatively impartial? Maybe it comes out later that they had other info that shows this article wasn't impartial but I'm not sure on its face it seemed like it was tipping that far one way or another. -Daniel
[2020-04-20 20:35:38] - "there's no impartial way to report this story" isn't this a giant cop out? can't you easily find analogies/similarities/differences/metaphores to the kavanaugh reporting? ~a
[2020-04-20 20:34:39] - I guess the point I'm making is that there's no impartial way to report this story. every detail you include, every detail you omit, any relevant context you highlight or ignore, whether you make it a permanent fixture on the front page or refuse to cover it at all... every decision serves a narrative much more than it will serve the truth. - pierce
[2020-04-20 20:24:49] - but it's hard to say that the alternative is better: that they should prioritize the details that support Reade's allegations, because believing women helps in the fight against institutionalized sexual assault. even if you held that belief, it's idiotic to pretend that that version of the story wouldn't just be weaponized to support the reelection of a man with more credible, numerous, and severe allegations against him. - pierce
[2020-04-20 20:05:44] - so I'm somewhat sympathetic to the times on this, it's a really difficult tightrope to walk. if they're perceived to be under-reporting this story, or if their reporting includes any concrete or circumstancial doubt about Reade's claims, then they'll be considered shills for biden, they'll anger people who think you should always believe women, and it'll draw accusations of hypocrisy when contrasted to the kavanaugh coverage. - pierce
[2020-04-20 19:50:37] - I started listening to citations needed's episode on this topic the other day, I haven't finished it yet but it's an interesting discussion. - pierce
[2020-04-20 19:48:32] - reporting "both sides" of an issue isn't impartial when one side is demonstrably wrong (and hedging it with "critics say..." doesn't negate the free credibility you gave them). and journalism doesn't exist in a vacuum, you shouldn't pretend the raw facts are all that matter and ignore the context in which the story is being told. - pierce
[2020-04-20 19:47:43] - I will say that the original article's asides discussing assault allegations against Trump, or Reade's pro-Putin posts, are hard to justify under traditional journalistic practices since neither one is directly relevant to the incident they're reporting on. but there seems to be more "meta" discussion lately about whether those practices actually work to make journalism more objective. - pierce
[2020-04-20 19:39:52] - also Andrea has a NYT sub so I didn't get the paywall. So I can answer questions about its content but I guess would still be filtered by me. -Daniel
[2020-04-20 19:39:03] - "Examining Tara Reade’s Sexual Assault Allegation Against Joe Biden" is dismissy? I guess it seems non dissmissive to me since it then proceeds to have a whole article look into it and talk about it? -Daniel
[2020-04-20 19:13:25] - the nytimes podcast dug into some of the "behind the scenes" considerations that went into how they reported the reade allegations, fwiw. transcript link, I don't think it hits the paywall. - pierce
[2020-04-20 18:03:08] - daniel: well, i can't read miguel's nyt link because of the paywall, but even the title seems dismiss-y. ~a
[2020-04-20 17:33:09] - Daniel: Yeah, I figured it wasn't, but it felt like the kind of question you would ask which is why it took me aback. -Paul
[2020-04-20 17:26:07] - I mean there are real Mr Clark's out in the world. -Daniel
[2020-04-20 17:25:42] - paul: I don't think I sent any questions to a TMF thing. I hope it was a good question though! -Daniel
[2020-04-20 17:18:05] - a: https://peterattiamd.com/covid-19-whats-wrong-with-the-models/ This has been an interesting read so far, and it just taught me that CFR and IFR are different things and I should've made my bet based on IFR.
-Paul[2020-04-20 16:29:56] - Daniel: I was just listening to a TMF video about FI where they answered a question from a "Mr. Clark" about retirement calculators. That wasn't you, was it?
-Paul[2020-04-20 14:21:00] - mig: Who is dismissing them out of hand? That seems like an odd characterization to me. -Daniel
[2020-04-18 21:48:49] - biden is such a fucking slimeball. what a gross dude. bleh. well, at least we probably won't have to hear anything from him when trump wins his reelection. ~a
[2020-04-18 20:39:39] - paul: https://www.nytimes.com/2020/04/12/us/politics/joe-biden-tara-reade-sexual-assault-complaint.html yeah coverage has finally happened. Seems awfully quick to dismiss Reade’s claims out of hand. W-E-I-R-D. - mig
[2020-04-18 11:56:50] - Mig: https://www.cnn.com/2020/04/17/politics/joe-biden-allegations/index.html Looks like the media is picking this story up now. -Paul
[2020-04-18 11:55:19] - Nah, I was mostly kidding. I suspect your second question could get them to post their returns publicly, though. -Paul
[2020-04-18 05:05:24] - Was I being aggressive? I thought they were just regular consumer type questions. -Daniel
[2020-04-18 04:31:06] - sc2 monday! ~a
[2020-04-18 02:21:52] - a: https://www.cnn.com/2020/04/17/health/santa-clara-coronavirus-infections-study/index.html Btw, you never answered about SC2 on Monday... -Paul
[2020-04-18 01:52:58] - Sorry, I had to step away from the computer this afternoon. Thanks for the feedback! I was honestly just going to ask some questions around starting up a new company and if they all left their previous jobs and whatnot. I'll leave the aggressive questioning to Daniel.
-Paul[2020-04-17 20:51:45] - now I'm just curious what Paul's question will be. -Daniel
[2020-04-17 20:51:35] - and it's probably not as black-and-white as that, they may welcome some "tough" questions if they come out ahead for answering them well. or maybe they're super idealists about free discussion, even if it gives their company a black eye. all I'm saying is not to assume it's an open forum for any reasonable topic. - pierce
[2020-04-17 20:45:07] - it's not a generic topic, it's a subreddit centering on a specific company that appears to be moderated by people at that company. they may not have an explicit list of posting guidelines, but I strongly suspect "no posts or comments that are critical of 7investing.com" is an unspoken rule. - pierce
[2020-04-17 17:28:47] - under "all. questions. welcome." ~a
[2020-04-17 17:28:29] - i do see that daniel already posted some questions
~a[2020-04-17 17:27:50] - paul: first thing i did was look over on the right side of the screen (sidebar). often people will post their "rules" there. i don't see anything. you're in the clear. next thing i clicked on was "submit a new text post". sometimes people will put "rules" there. nothing there either. you're definitely in the clear. post anything you want. if they don't like it, fuck-em. ~a
[2020-04-17 15:23:37] - https://www.reddit.com/r/7investing/ It's a new one started up where they are actively soliciting questions, so I feel in the clear. -Paul
[2020-04-17 15:20:45] - Subreddits are just the various buckets / demarcations that break down topics. "Reddit" is just a platform that shows you all the buckets you've subscribed to (or the default set of buckets if you don't have an account or ever changed from the defaults). -Daniel
[2020-04-17 15:19:18] - paul: Mig is right that some are very heavily moderated but I think a lot of them you don't have to subscribe to the subreddit in order to post there. -Daniel
[2020-04-17 14:44:01] - paul: which subreddit. It really depends on where. Some subreddits are pretty heavily moderated. - mig
[2020-04-17 14:41:51] - Can anybody here explain reddit to me as if I were a 5 year old? There's a... subreddit? that I want to post to. Is it as easy as joining it and then creating a post? Is that too presumptuous? -Paul
[2020-04-17 14:20:52] - a: How about SC2 next Monday (20th)? -Paul
[2020-04-17 14:13:39] - a: What was that in reference to? -Paul
[2020-04-16 20:41:57] - that's from me if it's not obvious. ~a
[2020-04-16 20:14:20] - paul: back to something from yesterday: "whether we should just get rid of it even in non pandemic situations". no, i think we're in a unique situation.
[2020-04-16 19:44:36] - 9:30? every day. ~a
[2020-04-16 19:43:54] - a: When works next week for you? -Paul
[2020-04-16 19:03:34] - paul: no. i play with you or nobody. i'm fine skipping this week. ~a
[2020-04-16 19:00:34] - the exact wording of the bet was not "". it was "assuming we have a reliable way to calculate [cfr]". ~a
[2020-04-16 19:00:05] - a: I would love to do SC2, but I already used one weekday evening for a podcast recording last night and am using another of Gears of War tonight and I think Gurkie would kill me if I used another this week. Maybe next week? You all can also play without me. -Paul
[2020-04-16 19:00:00] - oof. fuck you chrome. ~a
[2020-04-16 18:59:53] - sure, ok. the exact wording of the bet was "". ~a
[2020-04-16 18:59:08] - a: Yeah, I'll feel bad if we push on a technicality like that, but at the same time a large part of my bet was based on the idea we would "catch up" with testing before the end of the year. -Paul
[2020-04-16 18:50:26] - https://www.cdc.gov/coronavirus/2019-ncov/cases-updates/cases-in-us.html
[2020-04-16 18:50:22] - CDC says CFR is at ~4.1. -- Xpovos
[2020-04-16 18:46:44] - paul: sc2 this week? i've lost track of all things when it comes to what day it is, but i'm pretty sure we haven't played yet this week. ~a
[2020-04-16 18:35:20] - xpovos: i think paul and i agreed if testing was still fucked at the end of 2020, we'd push. ~a
[2020-04-16 18:34:33] - I understand we're under-counting cases because of testing problems. But that's why CFR is such an iffy measure. I don't think we're suddenly going to start testing extra. -- Xpovos
[2020-04-16 18:29:11] - piss adrian off? i said i was going to probably lose. and i've said at least twice that i'd be happy to lose this bet. ~a
[2020-04-16 18:28:26] - And just to piss Adrian off, that last link says some German virologists calculated a "COVID-19 infection fatality rate of about 0.37 percent". -Paul
[2020-04-16 18:27:24] - Xpovos: https://reason.com/2020/04/09/preliminary-german-study-shows-a-covid-19-infection-fatality-rate-of-about-0-4-percent/ Here's another. -paul
[2020-04-16 18:25:09] - Xpovos: https://www.motherjones.com/kevin-drum/2020/04/the-united-states-has-a-very-low-covid-19-case-fatality-rate/ This is the one I looked at last. -Paul
[2020-04-16 18:24:14] - a: I'm hoping that any number of things go my way. Medication is one, but I also think the infection rate is a lot higher (bring that denominator up). -Paul
[2020-04-16 18:23:59] - xpovos: i should also mention the reason article was suggesting we should be looking at 0.2%. ~a
[2020-04-16 18:22:51] - xpovos: the bet is for 1.25%
(united states for the whole year) i think paul is hoping medication will change the results. ~a[2020-04-16 18:21:36] - Paul: What's your source? One of mine has CFR in the U.S. as 5.12 right now. I don't think it's going anywhere near 2.5% -- Xpovos
[2020-04-16 18:12:48] - a: US is still "comfortably" hovering around 3% as far as I can tell. I still think that will drop a bunch if we get widespread testing out, but I'm getting nervous about it. -Paul
[2020-04-16 18:09:59] - jeeze i suck at betting. ~a
[2020-04-16 18:09:52] - uhhh, i thought i was going to lose both of those actually. ~a
[2020-04-16 18:07:51] - a: I think we're going to break even on the S&P / CFR pair. -Paul
[2020-04-16 17:57:55] - paul: and nobody likes their shit getting ignored. virginia will totally rescind their order, and i'll be $5 richer. . . . but in the end i'll just have to give it back when trump wins re-election. ~a
[2020-04-16 17:56:07] - a: Right, like, I think we're going to start seeing the populace take this less seriously even if these stay at home orders remain in place. It's going to be really hard to convince people to stay inside for months on end while the infection rate and death rate are (hopefully) going down. -Paul
[2020-04-16 17:51:40] - pierce: are you talking to me. no restaurants in nova allow dine-in. ~a
[2020-04-16 17:49:45] - like, sitting down at a restaurant? the places that are still open here have switched to 100% delivery or maybe take-out. if you're sitting at a table, unless it's getting disinfected after each customer a normal mask isn't going to help you much. it makes droplets get deposited right in front of you instead of spraying them, which is arguably worse at a table. - pierce
[2020-04-16 17:45:37] - paul: big "meh"! i was just in a restaurant today (i'm actually kinda a pretty bad hypocrite really). and *everybody* was wearing masks. patrons, workers, everybody. maybe this will become the new normal in 2020. ~a
[2020-04-16 17:42:22] - paul: oh i actually didn't see the quote. yeah, i think that's some liberal use of the word "such as"? unless things make a turn for the worse (which is a big "unless"), schools won't be closed in 2022. ~a
[2020-04-16 17:42:08] - Pierce: Hopefully Trump, who has absolute authority over these things, is able to capably handle that.
-Paul[2020-04-16 17:41:04] - it also doesn't have to be the entire country on a continuous distancing order... if our testing gets better and our "buffer" of available medical capacity increases, we could institute temporary distancing for specific areas that are experiencing an active flare-up. - pierce
[2020-04-16 17:40:52] - xpovos: i added it to the spreadsheet. ~a
[2020-04-16 17:40:11] - a: Regardless, it's hard to imagine restaurants can survive in an environment where we're supposed to be keeping 6 feet away and wearing masks in public. What would capacity be like? 20% max? -Paul
[2020-04-16 17:39:56] - ok. ~a
[2020-04-16 17:39:35] - a: Hmm. OK. $5 that Governor Northam will not revoke, or rescind his executive order on or before June 9th. -- Xpovos
[2020-04-16 17:39:12] - a: I kinda agree, but the quote said, "social distancing measures -- such as stay-at-home orders" so they kinda disagree.
-Paul[2020-04-16 17:35:34] - paul: social distancing != sah. ~a
[2020-04-16 17:34:11] - a: "i seriously doubt stay-at-home orders will last until 2021" I think we're in agreement, then. I do think that lots of people are talking about that long, though. Just browse some headlines in CNN right now. https://www.cnn.com/world/live-news/coronavirus-pandemic-intl-04-16-20/h_eda7c30ec7790ea0aec36bdede0d4320 "The US may have to endure social distancing measures -- such as stay-at-home orders and school closures -- until 2022" -Paul
[2020-04-16 17:32:55] - $5? ~a
[2020-04-16 17:32:16] - a: OK. I'd take the other side of that bet. I think it's possible, but unlikely. -- Xpovos
[2020-04-16 17:25:33] - i guess rescinded, but at a time earlier than june 10th? ~a
[2020-04-16 17:25:02] - xpovos: moved left . . . as like in a calendar. date is changed to be an earlier time. so, neither. ~a
[2020-04-16 17:24:35] - a: "moved left" means rescinded? Or extended? -- Xpovos
[2020-04-16 17:23:03] - i also believe virginia's stay in place order will be changed (moved left). that's a guess, but we could bet on it if you want. ~a
[2020-04-16 17:21:56] - paul: maybe 10 was too small. sure, make it 15 or 20. i seriously doubt stay-at-home orders will last until 2021. ~a
[2020-04-16 17:19:49] - a: Except we're not talking about 10 days, right? Virginia has stay in place orders out to June 10th or something. There are articles throwing out dates in 2021 for various things. I feel like we're talking about months here. -Paul
[2020-04-16 17:19:05] - xpovos: agreed. what's more, new york is about 8x that. my sister lives right next to manhattan . . . ~a
[2020-04-16 17:13:53] - U.S. has climbed over the 100 deaths / 1M population count. That's a major milestone. -- Xpovos
[2020-04-16 16:54:02] - ignore the "medical perspective" all you want. but which is better? scenario A or scenario C? in both scenarios the economy was totally fucked. but come on, seriously, you think that scenario A is best for *the* *economy*? ~a
[2020-04-16 16:53:02] - here's how i think about it. there are three scenarios. lets say after the fact we could somehow figure out which was the "correct" day to end the lockdown. scenario A: we ended the lockdown 10 days too early. scenario B: we ended the lockdown on the exactly correct day! scenario C: we ended the lockdown 10 days too late. now, obviously, scenario B is best for Joe Sixpack. ~a
[2020-04-16 16:29:06] - paul: yes, you could get sick too. but that's not even the worst case scenario. ~a
[2020-04-16 16:25:50] - a: I... don't think so? Not to downplay what could happen, but how could things get worse economically than being unemployed and without any prospects? I guess you mean they could get sick too? -Paul
[2020-04-16 16:24:02] - paul: if we end the lock-down too early won't they most likely get more economically massacred? ~a
[2020-04-16 16:14:39] - a: That's part of the reason why I just can't imagine this lockdown / social distancing lasting much longer... even if it makes all the sense in the world from a medical perspective. Certain segments of the population are getting economically massacred. -Paul
[2020-04-16 16:09:51] - i think at like 25% unemployment is when things start to come unglued. ~a
[2020-04-16 16:08:10] - i'm constantly surprised by this graph. it's been updated two times (1 update per week) since it made the rounds on social media (6m, 6m, 5m). btw, i think each entry is "initial claims" so each data point is "new" people? i.e. it's like a "delta" graph? i know the unemployment rate is like an addition of each of these. i hear we're at ~20% unempoyment which is *crazy* high. ~a
[2020-04-16 15:30:04] - a: Sounds good. I'll circle back in 2027.
-Paul[2020-04-16 15:16:56] - paul: 10? ~a
[2020-04-16 15:15:02] - a: Sure, there's some luck involved, but there was no virus for the other years prior either. How many wins do I have to string up in a row against Daniel in the stock market challenge before it looks like skill might play a factor? -Paul
[2020-04-16 14:33:29] - i could have never predicted that bitcoin would go from $1 to $10,000 each. (almost up 1m%). i got lucky. . . . right place at the right time. sure, it wasn't all luck, but it was like mostly luck? ~a
[2020-04-16 14:24:06] - yeah i don't think you predicted the virus. ~a
[2020-04-16 14:06:46] - a: Want more cherry picked Freedom Portfolio data? It's up around 11% YTD. This has been a bizarrely good year so far. -Paul
[2020-04-16 14:06:07] - a: Uh... that's how I sometimes felt about Heroes of the Storm (have you ever played? might be worth playing some of that too). Starcraft 2 I feel like I know in advance how it's going to go. FFA or playing against randos? Probably going to suck.
-Paul[2020-04-15 20:45:28] - paul: i don't really agree, but i feel like this might be how you feel about sc2 multiplayer? ~a
[2020-04-15 19:41:47] - Pierce: Inflexible, you might say. -Paul
[2020-04-15 19:31:49] - pierce: edibles. ~a
[2020-04-15 19:27:58] - you definitely seem like a blunt guy, based on that. - pierce
[2020-04-15 19:21:09] - Pierce: Nope! Overrated. -Paul
[2020-04-15 19:19:52] - what about joints? have anything nice to say about those? - pierce
[2020-04-15 19:14:21] - Pierce: You're not going to trick me into saying something positive about ethanol.
-Paul[2020-04-15 18:54:41] - paul: also, ethanol is an essential medicine, but you wouldn't use it to treat your liver cancer. hydroxychloroquine is in the complementary list for treating diseases of joints. - pierce
[2020-04-15 18:46:47] - a: tries, sure.
- pierce[2020-04-15 18:44:27] - pierce: that's totally fair. and i completely agree with this. on the other hand, i think that's the point of peer review. peer review tries to suss out confounding factors and the like. ~a
[2020-04-15 18:43:13] - paul: it's actually on the who's list of essential medicines "complementary list": "specialized diagnostic or monitoring or specialist training are needed. an item may also be listed as complementary on the basis of higher costs or a less attractive cost-benefit ratio." that kinda got lost somewhere. ~a
[2020-04-15 18:41:00] - thanks, anonymous
- pierce[2020-04-15 18:40:46] - a: try, sure. but isn't that already one of the Hard Problems of science in general, especially so in medicine, and with new and fun complications because this is a fairly unique event in modern history? - pierce
[2020-04-15 18:38:50] - hbd pierce
[2020-04-15 18:35:24] - a: https://en.wikipedia.org/wiki/Hydroxychloroquine In terms of safety, I was a little surprised to see: "It is on the World Health Organization's List of Essential Medicines, the safest and most effective medicines needed in a health system." -Paul
[2020-04-15 18:34:59] - pierce: don't most clinical trials try to account for those kinds of confounding factors? ~a
[2020-04-15 18:34:20] - a: yes, and I sent you my link. It sounds like we don't have a consensus yet. -Paul
[2020-04-15 18:33:50] - a: And it sounds like the things you are hearing are the opposite, so it makes sense we have different opinions on how this should be treated. I guess we'll find out (eventually) which side ends up being right. For the record, you're still looking better for the CFR. -Paul
[2020-04-15 18:33:18] - for a hypothetical example of why the experiment might be tricky, consider that most infected people are in urban areas. urban areas also normally have worse air pollution, which correlates to respiratory illness severity. but stay-at-home orders mean that air pollution has dropped significantly. so what are your baseline assumptions? most of the test candidates are from regions where there are confounding factors. - pierce
[2020-04-15 18:33:05] - "we should try to run those trials now". we are. didn't you see my "timely" link? ~a
[2020-04-15 18:32:30] - comparison ~a
[2020-04-15 18:32:13] - a: "if we've shown its worse than a placebo" I'm of the assumption that we haven't, which is maybe where we differ. The rumblings I've heard (which are NOT from Trump) is that this is a promising treatment that has shown some efficacy in totally non-statistically significant and non-controlled trials, so we should try to run those trials now. -Paul
[2020-04-15 18:30:04] - i'm a little sad that nobody is talking about the side effects ~a
[2020-04-15 18:28:18] - pierce: the side effects are huge (heart arrhythmia is just one of them), and the positive effects are small. ~a
[2020-04-15 18:27:26] - pierce: mostly when i want us to focus on a placebo, it's because we're ignoring that this thing might be worse than a placebo. ~a
[2020-04-15 18:26:35] - pierce: well i'm not sure how serious i am about marketing a placebo. but i hear you. it's cool to know that its not a silver bullet?
~a[2020-04-15 18:25:38] - a: the wikipedia page (for what that's worth) explores a lot of the "asterisks" of the placebo effect specifically. for example, its scope seems like it might be largely limited to pain treatment. - pierce
[2020-04-15 18:24:15] - it seems absurd to discuss the moral or practical calculus as if it was as simple as figuring out if it's better or worse than placebo and just letting anyone do it. - pierce
[2020-04-15 18:21:57] - so the confluence of factors that would result in (1) selecting candidates for testing this treatment without institutional bias, (2) having people with no medical training understand the choice that they're making by participating, (3) accurately determining its effectiveness and risk profile relative to some baseline, including identifying a useful baseline, (4) doing so without siphoning resources from other promising treatments? - pierce
[2020-04-15 18:21:01] - paul: ok i don't think those things. if we've shown its worse than a placebo (which i think is mostly where we are now?), we should probably look at promoting a placebo. or you know . . . one of the other many possible better drugs? ~a
[2020-04-15 18:17:35] - a: I'm not sure I see the same breath where I mentioned those two, but I'll take a stab at what I meant: (1) If it is safe, then let those in bad shape try it. (2) Even if it isn't proven safe, some people still might want to take the chance, and as long as they understand the risks, I say that's fine too. -Paul
[2020-04-15 18:14:57] - and the ethics are gnarly as well... the factors that might make someone desperate enough to try an unproven treatment aren't random. there's plenty of historical bad faith, where we let vulnerable people be our lab rats while reinforcing the conditions that led them to be vulnerable in the first place (e.g. the Tuskeegee syphilis experiment) - pierce
[2020-04-15 18:09:49] - pierce: so you're saying it's complicated? do you have any specifics? ~a
[2020-04-15 18:09:01] - paul: "just clear them when they are proven to be safe" this has not happened yet, right? you've said "a lot might take their chance with heart arrhythmia", but then you say in the same breath "proven to be safe". regarding "proven safe AND effective" vs "proven to be safe": i'm no scientist, but it could clinical trials show both of these things at the same time? ~a
[2020-04-15 18:06:35] - FWIW, I don't think any of this stuff is as simple as it's being characterized. the placebo effect is a complicated thing that doesn't manifest consistently and isn't always better than no treatment. performing an experiment is a complicated thing even when you're not talking about a potentially life-threatening illness that's reshaping society. prioritizing research is a complicated thing even without obvious political influence. - pierce
[2020-04-15 18:04:12] - a: But the more specific point I am making here is that we maybe instead of not clearing meds until they are proven safe AND effective... we should just clear them when they are proven to be safe. -Paul
[2020-04-15 18:03:06] - a: I think I get why you are confused. So, at one level, I do believe that basically anybody should be free to try any treatment regardless of if it has been studied or not. So I do believe that right now, even if we have no idea how harmful it could be, somebody should be free to try Hydroxychloroquine to treat their coronavirus if they want as long as they know it's not FDA approved or whatever. -Paul
[2020-04-15 17:55:24] - paul: agreed. ~a
[2020-04-15 17:54:34] - a: Well, we don't know yet, right? Isn't that what the trials are for? To determine how harmful it is? -Paul
[2020-04-15 17:43:56] - iow, you can't have it both ways. either its better than placebo, and you'll take the good with the bad. or a placebo is better, and then you can't just fucking ignore the bad parts! ~a
[2020-04-15 17:42:37] - "if we think it is not harmful" we don't think this. ~a
[2020-04-15 17:41:42] - a: I'm saying that because I know of no "better" treatment, that it makes sense to look into this one (regardless of if Trump is touting it or not) and that if we think it is not harmful, then it makes sense to let people with no other good option take it. -Paul
[2020-04-15 17:21:37] - paul: ok, well i don't think you're stating your position clearly. would you suggest people should take hydroxychloroquine even if its shown that you're better off taking a placebo? or are you not saying that? ~a
[2020-04-15 17:19:18] - a: But for somebody with a bad prognosis and suffering from coronavirus... I think a lot might take their chance with heart arrhythmia and the chance of there being no benefit considering the likeliness of death otherwise. -Paul
[2020-04-15 17:17:40] - a: Right, which is why I prefaced with "assuming it is shown to be safe". Obviously, in reality, it will probably be shown to have some small percentage of minor (and maybe major) side effects that will have to be balanced by somebody to determine if it is worth the chance. -Paul
[2020-04-15 17:10:55] - "use it as a placebo to help people" ha. nice try, but placebos do not cause heart arrhythmia. ~a
[2020-04-15 17:07:17] - a: https://thehill.com/policy/healthcare/492889-hydroxychloroquine-alleviated-symptoms-did-not-cure-coronavirus-in-small Sure, there's (small) studies all over the place that show all sorts of different things. That's why I agree we need to keep studying it, but assuming it is shown to be safe, it feels like we can at least use it as a placebo to help people, right? -Paul
[2020-04-15 16:53:38] - paul: timely ~a
[2020-04-15 16:30:01] - ha. uhhh, i'm sure we've tested multiple placebos, but maybe not. ~a
[2020-04-15 16:29:35] - a: Maybe sugar is just a cure-all? (Aren't they normally sugar pills?) -Paul
[2020-04-15 16:29:33] - ah ok. ~a
[2020-04-15 16:29:15] - a: I thought I had read they work less well (when known), but still work. -Paul
[2020-04-15 16:27:40] - a: I do a quarterly update on PvtM, so every quarter I discuss the performance. -Paul
[2020-04-15 16:27:18] - paul: yeah i wonder though. do placebos work better when you don't know they're placebos. i have to believe yes, but results in this area are often illogical and surprising. ~a
[2020-04-15 16:27:11] - a: https://aporter.org/msg/?action=prev&prev=149250 I had a reference to it losing to the market back then. -Paul
[2020-04-15 16:26:22] - here. i searched through here. i also checked the pvtm, to confirm you discussed it then, heck it's the only place i would have found the 2018-12 info, right? ~a
[2020-04-15 16:25:49] - a: On the message board or on PvtM? -Paul
[2020-04-15 16:25:26] - a: Also, the whole idea that it works EVEN IF YOU KNOW it is a placebo makes no sense to me. I did know that, but when I found out it kinda broke everything I thought I knew about medicine and science. -Paul
[2020-04-15 16:24:31] - paul: due to the survivorship bias, we need to search back through the message board to see if you told us when the freedom portfolio was losing to the market: so 2018-12? yeah it doesn't look like you reminded us it was losing to the market back then.
~a[2020-04-15 16:24:24] - a: Right, I mean as non-trials. Why don't we hand it out for everything? -Paul
[2020-04-15 16:18:53] - paul: we do, actually. as part of trials. but maybe you mean as part of non-trials? i'm not sure. i think we could start. (i'm sure you know it's proven to work even when people know they're placebos, so we could even market them as placebos) ~a
[2020-04-15 16:16:25] - Random aside: The Freedom Portfolio is now beating the market by 23 percentage points (!). The weird thing is that the delta seems to expand MORE during down days thanks to stocks like TDOC/AMZN/NFLX which tend to go up when the rest of the market is down. -Paul
[2020-04-15 16:12:08] - a: But we don't typically just hand out placebos to people, right? Even though they're pretty much proven to actually help? -Paul
[2020-04-15 16:01:27] - paul: 6. whatever j&j is calling this thing ~a
[2020-04-15 15:53:38] - paul: "no better drug in the pipeline". better (?) drugs in the pipeline. 1. placebo. 2. baricitinib. 3. remdesivir. 4. lopinavir. 5. ritonavir. ~a
[2020-04-15 15:44:43] - it shouldn't be too hard to prove it's better than a placebo. unless it isn't. ~a
[2020-04-15 15:43:54] - if a placebo is better, we all should be promoting it. ~a
[2020-04-15 15:43:13] - paul: i'm not joking: it's taking resources away from a placebo. ~a
[2020-04-15 15:42:45] - a: If there's no better drug in the pipeline, then I don't quite get the objection to testing this one out. It's not taking resources from something better that we know about. -Paul
[2020-04-15 15:42:05] - a: Okay, then let's show lack of harm and then let people choose to use an unproven drug if they want. -Paul
[2020-04-15 15:18:18] - paul: 1. lack of harm, sure! that it doesn't cause heart arrhythmia. that people don't literally die from heart arrhythmia. 2. that it's better than a placebo. "what's the much better drug?" a placebo. maybe. but, if you want more, ask bill gates, he's thinking one might not be enough. ~a
[2020-04-15 14:52:39] - a: "when we could devote that money towards a much better drug" What's the much better drug? -Paul
[2020-04-15 14:52:20] - a: Is that what the red tape is? If that's the case, then what are the trials that they are doing now supposed to show? Lack of harm? -Paul
[2020-04-15 14:44:00] - "designed to determine whether hydroxychloroquine is any better than a placebo in preventing Covid-19" oh, yes, that horrible red tape. will this help better than doing nothing? what blows my mind is that we may put literally millions behind a drug that might not help, heck it may even hurt, when we could devote that money towards a much better drug. for heaven's sake, let the good be the enemy of the worthless, please. ~a
[2020-04-15 14:37:53] - https://www.cnbc.com/2020/04/15/coronavirus-us-clinical-trials-of-hydroxychloroquine-hit-warp-speed.html I legitimately wonder (no snark) what the red tape is that is being bypassed and whether we should just get rid of it even in non pandemic situations. -Paul
[2020-04-15 14:37:20] - a: I'm pretty sure one of my major problems right now is not using "spellcasters" (ie, infestors and vipers for zerg, templar and adepts and sentinels for protoss, etc) because I worry about being able to handle their micro. -Paul
[2020-04-15 13:47:55] - yeah i wish i was better at micro-ing. psionic storms are pretty hard to manage in my opinion. not just deploying them, but moving around the multiple parts of your force correctly. (like siege tanks micro, but worse). i have a hard enough time managing blink and void ray prismatic alignment. (which i only recently noticed slows them down a whole bunch making them vulnerable to an overwhelming force). ~a
[2020-04-15 03:15:09] - a: Just way too many psionic storms to weed out the zerglings and then the immortals made short work of the rest. I thought about going air eventually but it was too late and I can't imagine brood lords would've done much. -Paul
[2020-04-15 03:14:19] - a: Practicing against the AI, I found another Protoss force I found myself utterly unable to handle: Mass Templar/Archons/Immortals. The game had gone on forever (I had literally lost 3-4 max supply armies trying to take out the opposing force... mostly without denting it). I tried roaches and hydras and zerglings and ultras.... neither did much. -Paul
[2020-04-14 19:08:56] - a: https://www.washingtonexaminer.com/news/rand-paul-constitution-doesnt-allow-trump-to-be-ultimate-regulator-of-our-lives?fbclid=IwAR1cot59x1YGnbLNn2tG19eSC3ukVKl Surprising to see Rand Paul actually speak out against it, though. -Paul
[2020-04-14 19:07:47] - a: I have many, many thoughts. I'll give a glib one for now, though: "what would or could happen next?" Justice Roberts calls it a tax and says it is okay. -Paul
[2020-04-14 17:33:33] - any thoughts on whether the president of the united states has total authority? (this has been stated explicitly, and it's hardly been taken out of context) specifically if trump told a governor to open their state, and the governor decided to keep their stay-at-home-order in effect, what would or could happen next? ~a
[2020-04-13 18:43:02] - np, i got it. ~a
[2020-04-13 18:42:54] - hmm forgot quotes. my bad
[2020-04-13 18:42:29] - a: sounds good. those <a href=https://github.com/Poobslag/turbofat/pull/101/files#diff-45150bc9b5e2268a78c1cc8bf2398d6d>unit tests for piece kicks</a> came out cool by the way, the tests are literally just little ascii diagrams showing how the pieces should rotate in different cases. i'm so proud of it! ...now i'm doing 3D stuff. 3D is hard! - aaron
[2020-04-13 17:43:22] - aaron: haha, sure.
very few of the games have "cc" in the left column, which is fairly consistent with what i said. but yeah: you can/should have the content be cc-by-nc-sa, then you'll mostly not be competing with people using your game to make money. ~a[2020-04-13 17:40:14] - a: https://en.wikipedia.org/wiki/List_of_open-source_video_games Wikipedia has a list of licenses for other open source games. I'm too dumb to solve this sort of problem myself, so I'll just steal their solution and pretend I came up with it. ...That's how open source works, right? - aaron
[2020-04-13 17:20:07] - "As of Easter Sunday, 585 crew members had tested positive", yeah wow. ~a
[2020-04-13 17:18:45] - i'm actually a little surprised by this. i remember looking at the numbers two weeks ago and thinking to myself, "well, probably nobody will die though". ~a
[2020-04-13 16:41:06] - the stack-exchange link looks useful. he specifically talks about the QT model which sounds like you, but i'm not sure this is how these licenses actually work: "LGPL for non-commercial use, contributor agreement to maintain ownership of the whole project, separate commercial license" and the other people reply to that. ~a
[2020-04-13 16:40:13] - cc usually isn't designed for software. i actually don't have any concrete advice for you here. i'm sure such a license exists, but i've never heard of it (other than the one you noted: the cc-by-nc-sa). i did google this and got a few links. (more continued) ~a
[2020-04-13 16:25:18] - a: what kind of open source license do you recommend if i want to make a game where i don't care about people building, editing or distributing my game, but i don't want them to monetize it? cc-by-nc-sa? - aaron
[2020-04-12 03:25:01] - a: I suspect cannons vs batteries is if you want more DPS or more survivability... -Paul
[2020-04-12 03:24:30] - a: Interesting... What is SQ? -Paul
[2020-04-10 19:47:57] - here's a specific strategy thing i don't think i understand after watching some experienced players play: when do i make shield batteries and when do i make cannons? do i always create a mix of both? or are there times where i want to stop making one or the other? ~a
[2020-04-10 19:44:12] - it lets you compare your stats to grandmaster stats. which is interesting. you can see which stats you're the most sucky at. and how to improve. ~a
[2020-04-10 19:42:44] - just found this site which is interesting . . . http://sc2replaystats.com/ looks useful. ~a
[2020-04-10 19:26:02] - -Paul
[2020-04-10 19:25:58] - Daniel: Yes! I'll be online. I've been watching videos to practice.
[2020-04-10 18:58:05] - we still good for starcraft tonight? -Daniel
[2020-04-10 17:21:45] - it doesn't make me mad though. as i mentioned earlier, i'd be perfectly happy to lose this bet. ~a
[2020-04-10 17:20:56] - i was probably thinking "deaths". which is inverted. ~a
[2020-04-10 17:20:27] - yes. you can change it. ~a
[2020-04-10 17:20:06] - a: For our wager on CFR, isn't it written wrong in my spreadsheet? I took the under, right? -Paul
[2020-04-10 16:14:15] - a: https://reason.com/2020/04/09/preliminary-german-study-shows-a-covid-19-infection-fatality-rate-of-about-0-4-percent/ Another article to make you mad.
-Paul[2020-04-09 19:56:33] - sorry, i lost power
~a[2020-04-09 17:55:07] - being suspicious is good though. i'm also suspicious of numbers. ~a
[2020-04-09 17:54:02] - paul/daniel: you can account for these factors: just look at a large enough set of people without covid and see how many of them are dying of various things. you can also look at just the recent increase in deaths, and remove any other expected (cyclical or otherwise measurable) increase/decrease in deaths. at the end of the day, i think statisticians will be able to suss out these situations with enough of a degree of certainty. ~a
[2020-04-09 17:48:33] - a: And by significant number, I mean like 10-20%, not a majority. -Paul
[2020-04-09 17:47:32] - a: I think I agree, but I've also gotten a little suspicious of associating "Person A died and had Coronavirus" with "Person A died because of Coronavirus". I've read a few of those articles and they almost always had some other underlying issue. I'm wondering if a significant number of those people would've died anyway. -Paul
[2020-04-09 17:09:41] - I think thats why some things were listing cause as complications due to covid or something like that. I thought a lot of the time it wasn't techincally covid that killed you. It just fucks up your immune system and then some other thing that normally isn't a huge deal ends up killing you. -Daniel
[2020-04-09 15:56:49] - i know this isn't responding to your point, but i'm making a new thought: the cause of death is (and should be) covid-19 even if you had other medical conditions assuming you would have lived if it weren't for covid-19. regardless, determining cause of death is probably hard when you have lots of problems. ~a
[2020-04-09 15:56:37] - no i don't know. ~a
[2020-04-09 15:52:38] - a: Because with this hitting the elderly and people with pre-existing conditions a lot... I can see a lot of people who have COVID-19 dying even if they might've died from their other conditions regardless. -Paul
[2020-04-09 15:51:16] - a: For CFR, do you know if they count ANYBODY that dies who tests positive for COVID-19.... even if they had some other medical condition which may have been the cause of death? Do they try to identify what caused the death? -Paul
[2020-04-09 15:40:45] - a: Believe me, as somebody who lives at the intersection where one road makes a 90 degree turn and another road simply "disappears" (despite seeming to go on straight ahead).... I share your disdain. -Paul
[2020-04-09 15:39:51] - a: Yes, by most all metrics. -Paul
[2020-04-09 14:25:27] - paul: why do people feel the need to have road-names so useless like west-ox road and lawyers road. both of the road-names make turns at major intersections. so close to each-other too. who thought that was a good idea? ~a
[2020-04-08 20:49:07] - paul: are you overly white? ~a
[2020-04-08 19:53:45] - I think the field is a smart bet. About the only thing I feel certain about is that they will be young-ish and probably not overly white. Probably female? There's lots of people who fit that. -Paul
[2020-04-08 19:38:29] - paul: as for Reade's credibility. Who knows? Little coverage means really not that much info to go on.- mig
[2020-04-08 19:30:45] - I'm not taking the field. It's going to be a "safe" pick, as Biden himself was. - mig
[2020-04-08 18:49:37] - Daniel: I'd still take the field too, honestly. Some very strong choices mentioned, but the likelihood that he picks someone unknown is too high. -- Xpovos
[2020-04-08 18:47:31] - i pick michelle obama. ~a
[2020-04-08 18:46:59] - no. ~a
[2020-04-08 18:46:44] - Can I take the field against those three? I'm not sure who it will be but would be willing to play the field. -Daniel
[2020-04-08 18:44:08] - a: Kloubuchar. -- Xpovos
prev <-> next