My second Arizona Fall League notebook went up on Monday, covering everyone of note whom I hadn’t written up in the first one. I also held a Klawchat on Thursday.
I appeared on TSN 1050 in Toronto to talk about the League Championship Series and the Blue Jays, including prospect Ricky Tiedemann and the controversial decision to replace José Berrios with Yusie Kikuchi in what turned out to be their last playoff game.
And now, the links…
- Longreads first: Robert F. Kennedy, Jr., has profited by spreading lies and half-truths about vaccines for twenty years now. A book his group, Children’s Health Defense, and his publishing partner Skyhorse put out about the COVID-19 vaccines used the photo of a 12-year-old who died when a malformed blood vessel burst in his brain on its cover, claiming he’d died from the vaccine. He was unvaccinated, and of course they used his photo without his family’s permission. This Associated Press story talks to several families hurt by Kennedy’s lies and his so-called activism.
- A Koch-funded right-wing site, Cowboy State Daily, is injecting climate denial and transphobic rhetoric into Wyoming politics, crowding out legitimate news sites that adhere to basic journalistic standards. It’s part of the national Metric Media network, which has over 1200 sites across the country and also takes funding from the Koch family’s DonorsTrust foundation.
- Arizona continues to deny native tribes water rights they were promised under a 1908 treaty with the federal government. A third of homes on the Navajo reservation do not have access to clean water.
- The Los Angeles Times’ Michael Hiltzik looks at how anti-vaxxers are exacting a personal and professional toll on scientists, and we’re losing the war against such disinformation. Dr. Peter Hotez notes within the story the death threats, doxing, and other intimidation tactics he’s seen as a vaccine developer and advocate for vaccination and science in general.
- Meanwhile, a new study in BMJ estimates that 27 would-be mothers in the UK died during the pandemic because of confusion over the safety of COVID-19 vaccines during pregnancy.
- The Minnesota Reformer has a story on how small businesses in Minneapolis are pressured to hire off-duty cops as “security” in what sure sounds like a protection racket to me.
- Nebraska Gov. Jim Pillen (R) owns a conglomerate of hog farms that are dumping huge amounts of nitrates into ground water in the state, but when a local journalist published an exposé in the Flat Water Free Press, Gov. Pillen attacked her personally by saying “the author is from Communist China. What more do you need to know?” I think I need to know when he’s going to clean up his hog farms, and I hope more journalists there put his pig’s feet to the fire.
- Barbers Hill ISD in Texas suspended an 18-year-old Black student for wearing dreadlocks and is now forcing him to attend an alternative high school, despite the fact that Texas passed its own CROWN Act, prohibiting discrimination based on hairstyle, and the district already lost a lawsuit over the same issue in 2020. The district’s superintendent and deputy superintendent are both white, in case you were wondering.
- Book-banning zealots in Iowa are having a field day by targeting any books with a ‘sex act,’ including numerous important works of literature like, ironically, the dystopian novels 1984 and Brave New World.
- The federal government has begun distributing the $100 billion in funding for climate change mitigation projects under the Inflation Reduction Act, but it’s taking much longer than expected.
- Trump, with the help of The Heritage Foundation, has a plan to gut the federal civil service and replace tens of thousands of employees with Trump sycophants in a return to the “spoils system” of the 1800s.
- My friend Tim Grierson has a long, entertaining interview with Eddie Izzard on Cracked.com.
- Scientific American looks at the “Sleeping Beauty problem” in probability, which has generated arguments like the Monty Hall problem and the question of that dress’s color.
- A professor at UC-Davis tweeted a fairly overt threat against the lives of Jewish journalists, spurring an investigation by the university.
- If ChatGPT can do much of the work in academic grant writing, does that tell us that the grant-writing process is broken?
- Police in Newton, Iowa, have twice arrested a local resident for speaking out against the police at town meetings, so he’s suing them for violating his First Amendment (and other) rights.
- The Azeri-Armenian conflict may move to Meghri, an Armenian town in a strip of land that separates Azerbaijan from its Nakhchivan exclave. Azerbaijan calls it the Zangezur corridor, and it borders Iran, which opposes any changes to the current structure, while Turkey and Russia appear to favor some alteration that would connect the two parts of Azerbaijan even at a reduction in Armenia’s sovereignty.
- The Guardian’s Jonathan Freedland argues that Benjamin Netanyahu has walked into Hamas’s trap, and Israel needs to remove him from power before the situation gets even worse.
- Twitter Blue subscribers have become “super-spreaders” of misinformation about the Israel-Palestine conflict according to a brief analysis from the independent group NewsGuard, which attempts to rate the reliability of news sites.
- The New York Times’ Katherine Miller writes about the rambling, repetitive speeches of Donald Trump as he pounds his false claims that the 2020 election was rigged.
- St. Louis County police appear to have ignored erratic, threatening behavior from one of their own in the days leading up to him firing shots in the air at a trunk-or-treat event and shouting “you’re all going to die!”
- Deepfake porn videos are spreading too fast for the available tools, and far faster than any government is moving to stop them. Rep. Yvette Clarke (D-NY) did introduce a bill to ban deepfakes and step up enforcement, revising a similar bill she’d introduced in 2019, but it’s only at the committee stage and all sponsors are Democrats.
- David Roth writes about how Michael Lewis was taken in by fraudster Sam Bankman-Fried, as Lewis’ reputation continues to take hits after his cringy response to Michael Oher’s lawsuit against the Tuohy family. (Oher appears to have no dispute with Lewis over the book The Blind Side.)
- The Chicago man who stabbed a 6-year-old Palestinian boy 26 times, killing him, and stabbed the boy’s mother as well appears to have been radicalized by right-wing media. The spread of disinformation and hate has real consequences.
- Board game news: The Kickstarter for the roll-and-write game Merchants of Magick is 3x funded and still up for six more days.
- This year’s Essen Spiel convention was the largest ever, bringing nearly 200,000 people to the German city for the biggest board game con in the world.
The Robert F. Kennedy Jr. article is sickening.
Also, Iowa is banning 1984 and Brave New World because they contain a “Sex Act”??
!?!?!??!
How many other famous novels must be on this list? Obviously Lady Chatterlye’s Lover. The Heart is a Lonely Hunter. Stranger in a Strange Land. Probably any Sci-fi novel written by Robert Silverburg.
Maybe we should ban the bible given how raunchy it is? I’m pretty sure history says if you’re banning books like 1984 and Brave New World, you’ll be remembered as the villains. I’m sure there’s a joke to be made about banning books that warn about the dangers of censorships and big brother;
Probably. I mean, has Fahrenheit 451 ever been banned? That would be even more ironic than banning 1984 or Brave New World.
What exactly is the Iowa criteria? Does there have to be some sort of concrete reference to a sex act to ban the book? Or can they just ban any book with an implied sex act, such as “The Age of Innocence”?
Also, on a lighter note – since the links are usually just miserably depression – I am fascinated by the Sleeping Beauty problem. I can’t make up my mind about 1/2 or 1/3.
Similar question: In a card game that has 3 “witness” cards and one “assassin” card, in which one of those 4 cards is randomly set aside and the other 3 cards are dealt to the three players, there is a 1/4 probability of an “all-witness” game. But, if any one player looks at his or her card and knows he or she is a witness, then what is the probability, from that player’s perspective, of an all-witness game? (Note that this is not exactly the same as the famous Monte Hall problem).
Also, the article about Michael Lewis’s book was very well written. I don’t mean the content (which was also excellent); I mean the writing style and vocabulary. I would read any book written by the author of that article.
@Frank, I’m firmly in the 1/3 camp for the assassin game because the player is being given additional information that rules out one possibility. It’s similar in my mind to “what are the chances of flipping a coin twice and landing on heads both times if you know that the first flip was heads.” That’s where I get lost for the Sleeping Beauty problem. Is she really getting additional information if she forgets each time she wakes up? The 1/2 probability makes the most sense to me from that perspective (and the Taylor Swift vs. Usain Bolt example seems like the best case for that side).
@ Salty Scientist:
I would go with 1/3 also for the assassin game – which does actually make it very similar to the Monte Hall problem.
In the Monte Hall problem, he is always revealing a non-car door when he reduces the possible options to 2 and asks the contestant to choose. In the case of this game I described, it would always be a “witness” we are revealing and then asking whether that changes the probability. So, we have new information, and we need to act on it. Clearly, if we revealed all 3 cards, that would affect the probability of the 4th facedown card (we would know it with 100% accuracy), so, it’s obviously a case of conditional probability. We could use that knowledge and apply logic, or we would probably solve it using Bayes’ Theorem.
Or we could take it to the extreme and say, “We start with 100 witnesses and one assassin…..” which is how I typically explain the Monte Hamm 3-door problem to people who do not understand. If say, “Start with 1000 doors and only one has a car, do you really think it’s still 50/50 that your original door choice was the one with the car, or do you want to change now that I’ve eliminated 998 doors that definitely do NOT have a car?”
I’ll continue to mull over the Sleeping Beauty problem. Where is Marilyn Vos Savant when we need her ???
@Frank, I think the Sleeping Beauty problem honestly isn’t much of a problem at all, because it introduces conditional probabilities to argue for 1/3 when none exist. Obviously *if* Sleeping Beauty were given information about what day it was, that would create a conditional probability scenario, but in the original question she isn’t. TailsMonday and TailsTuesday are both simply part of the Tails probability, which is 1/2.
As you guys have noted with the Assassin game, we have additional information which changes the probability. Knowing you’re a witness absolutely increases the odds of an all-witness game to 1/3. In a slightly different vein, when the WSOP is on TV and somebody is chasing a flush draw, the broadcast will adjust that player’s chances of winning the hand if they know that some of his outs are in other players’ hands or were mucked.
@Kevin. That’s where I’m at with the Sleeping Beauty problem–is it really conditional? I think not. But, if smart people think it is, it at least gives me a lot of pause to try to see their side of the argument.
@Salty, to dig further one step, the probabilities are actually H|M = .5, T|M = .25, T|T = .25. The argument for 1/3 would seem to rest on an even probability that she was being awoken for each of the three possible states, but that isn’t really the case. Maybe in the actual community the arguments are deeper than what was presented in the article, but based on what we have? Not much of a problem at all.
@Kevin, exactly. I don’t get 1/3 as even a possibility.
As I see it, there are two different interpretations. The overarching one, which is of utmost simplicity. The probability of heads is 1/2. Full stop. It was a coin flip. Done. The number of times you wake up a girl your are needlessly torturing to confuse people is completely irrelevant. The probability a single coin flip comes up heads is and always will be 1/2.
The other interpretation is from Sleeping Beauty’s perspective. But if this is what they were going for, it would be better phrased as a betting game. “Every time Sleeping Beauty is awakened she is given the opportunity to bet on whether the coin flip had been heads or tails. How should she bet?” In the way it is actually phrased, she can still answer that the probability is 1/2 philosophically and you can’t call her wrong. But in the betting game, it is instantly clear she should bet on tails. And it would be even more so if you did the same sort of Monty Hall 100 doors expansion type modification and made it so that she would be woken up a whole bunch more additional days in the event of tails.
But what is the actual probability of tails from Sleeping Beauty’s perspective? In the case of the problem as presented, I came up with 62.5%. My reasoning is as follows (if there are holes in this let me know – I don’t claim to be an expert). 50% of the time she is awoken only once, on Monday. The other 50% of the time, she is awoken twice, once on Monday and once on Tuesday. So 75% of the time that she is awoken, it is Monday. When she is awoken on Monday there is an even probability of heads or tails, so so far it is 37.5% heads and 37.5% tails. The other 25% of the time she is awoken, it is Tuesday, and in every one of these cases, it was tails. So the probability of heads is 37.5% and the probability of tails is 37.5% + 25% = 62.5%.
And then of course I kept thinking about this a little more and became convinced my math was all wrong and now do see the 1/3.
Again using the betting scenario, if Sleeping Beauty bets on tails each time she is woken up, she will lose the bet 50% of the time, but she will win the bet twice the other 50% of the time, leading to a rate of return that equals an implied probability of 2/3 tails and 1/3 heads.
I won’t defend Lewis’ vicious and disrespectful quotes about Oher. There were inappropriate, and clearly driven by Lewis’ personal friendship with the Tuohys. (Lewis knew the father from the past, hence why he knew about the story to begin with.)
As to Going Infinite, the author from the Intelligencer article was much more balanced in his opinion and review of Going Infinite than others I have seen. But I’m not understanding the reviews calling this book a hagiography of SBF or some sort of sympathetic portrayal. (I read it a few weeks ago). I do believe Michael Lewis is biased, just like anyone would be after spending so much time with SBF. But Lewis presents a dispassionate view of SBF, along with the people around him. Even with a dispassionate view, SBF comes across as an unfeeling psychopath — who himself admitted in written communications and in conversations detailed in the book that he did not have feelings and had to make facial reactions to feign emotions to get people to trust him (much like some serial killers). Lewis also discusses SBF and his cohorts obvious crimes, without calling them as such, including their $3B fictitious trade with another firm also controlled by SBF. All so Alameda Capital could mark down a fake trading loss for tax purposes, otherwise known as tax fraud. And it’s obvious that he embezzled money from FTX to fund his own trading firm at Alameda Capital — and moved money back and forth, ignoring the fact that it was the money of FTX’s customers, not his own to play with because, once again, he has no feelings and cannot feel remorse. In the more quotidian fashion, several stories are relayed where SBF promised to do one thing to multiple people, including close “friends” and business acquaintances, only to do the opposite with absolutely no remorse. I’m puzzled as to how people walk away from this thinking Lewis made SBF out to be sympathetic.
I think people are upset that Lewis just didn’t come out and call SBF the world’s worst human being. At least the author of the Intelligencer article did point out the sections of the book, in which Lewis described SBF’s worst characteristics. I just disagree with the author that Lewis somehow excused those behaviors with one-line explanations. I actually think Lewis was trying to show the reader how SBF himself internally justified his wrongdoings, not that Lewis believed those behaviors to be appropriate.
As someone who has read seven Michael Lewis books. I say this falls in at fourth or fifth with Big Short, Flash Boys, and Moneyball above it and tied with Liar’s Poker — with Premonition and Fifth Risk falling below for being relative snoozers.
In regards to reinstituting the spoils system, the skeptic in me thinks it would be part of the grand plan of the Heritage Foundation to destroy government as we know it. Civil Service and the protections afforded to government employees allow them to use their best judgment without worries of political retaliation from the other side. It also keeps the government running during transition periods since there’s always a certain group of employees who are doing their jobs. Getting rid of it would cause mass chaos and ensure that unqualified people are holding key positions of power. I know the left has a hard time doing it but they need to embrace the system and point out that the people who would be affected are your neighbor like postal workers for example.
The ChatGPT grant writing article misses the mark for me. Much of the extra paperwork for a grant is a one time thing (facilities, equipment, budget, Bio(graphical)sketch) in that you make them once and they’re largely the same for each proposal with minor tweaks. And yes, ChatGPT can write an okay summary of an abstract or revise some of those documents. But it can’t come up with ideas de novo and write them into a compelling narrative of proposed experiments, which is 90% of the actual work of grant writing.
And as to the point of those documents? The abstract is public facing, and the public has the right to know what research their tax dollars are paying for. The facilities and equipment documents are to make sure that the proposed experiments can actually be done (and AI can’t at this point walk through a lab space and make an inventory of equipment). The Biosketch is to make sure that the applicant is qualified to do the work, and I think the article writer is likely an outlier if they are googling scientists instead of reading Biosketchs. The budget justifications may not get tons of scrutiny from reviewers (though they do get scrutiny if they’re very high), but they sure as hell get scrutinized further down the decision chain (as anyone who has had cuts made to their grants very well knows). The article laments that AI can generate these documents (though it really can’t do it well for most of them), and then ends with “perhaps it is time for funding bodies to rethink their application processes.” So what’s the alternative? Don’t include those documents at all? I’m all for ideas to streamline the process, but this really smacks of complaining about a very small part of the work that goes into grantwriting while missing the point of why we include those documents.