Sam works on innovation policy at Progress Ireland, an independent policy think tank in Dublin, and runs the publication Fitzwilliam. Most relevant to us, he writes a summary of popular links on his personal blog. Below is a condensed version of his link from November.
Blog and short links
1. Is Google Search Getting Worse? The quality of evidence on this point is significantly low.
2. About the game theory of $1 Margarita Night. It would be great if you could have a stand-up social event at the same time and place each week that would serve as “office hours” for your group of friends, as described in this post. I feel like a lot of time is spent coordinating with society because plans are so changeable.
3. I pray for Peter Temin’s soul. I enjoyed his article reviewing the economy of the early Roman Empire. Someday, I’ll probably cover the economics of pre-industrial Europe in my book club.
4. I recently wrote for Progress Ireland about why more academics don’t start their own businesses. We also updated our efforts to support Olympic level mathematics in Ireland and launched paid subscriptions to our blog.
5. Climate predictions are relatively accurate.
6. Will AI further strengthen NIMBYism? The initial expectations for initiatives like Tract were to the contrary, and we cannot predict how these things will change.
7. Reduce VAT to generate more readers. Do you see an arbitrage opportunity in the paper market?1
8. Common sense in urban economics says that minimum standards for apartments in terms of floor space and personal preference are a terrible idea. The Irish government’s announced reduction in minimum apartment standards now faces a challenge in the High Court and could be delayed or scrapped altogether. The court in question has hinted that the European Union may need to get involved. Even if you think modestly lowering minimum standards for apartments is a bad idea, it’s baffling to see a world in which democratically elected officials had the authority to put these regulations in place in the first place, but couldn’t modestly lower them. Is this anarchism about the question of political authority, but only if the change is in the direction of less regulation rather than more regulation? Is there a worldview here, or is it just bitterness and cynicism?
9. Another thought: Have I been seeing weird flag cones this whole time?
10. Stephen Webb explains why there are too many lifeguards in the UK. I am reminded of Mitchell and Webb’s sketch about how outrageous it would be if no one drowned in the UK after a year.
I am trying to draw attention to the huge waste of public money that has resulted in a situation where no one drowns in a year. It’s just amazing what that means in terms of fencing, warning signs, swimming lessons, people coming to school to remind kids, lifebelts, maintaining waterfront roads. If a city of 500,000 people like Westchester was run with proper priorities, at least two or three people would drown each year, so it’s clearly a massive overspending.
I’m familiar with the different methodologies for calculating the value of statistical lifespans, but could someone please explain to me (a) why different countries choose different methods despite being at similar levels of development, and (b) whether there is any practical significance to the large discrepancies in these numbers? (After finishing that, I feel the same confusion about the different methodologies for calculating social discount rates.)
[E]Indeed, governments have a value for human life, which determines the appropriate level of investment in things like road safety measures. Here the UK typically falls on the lower end at around £2.5 million (for example $3 million). This compares to more than $12 million in the US and about $4.5 million in the EU.
Finally, my email inbox informed me of the Scottish lifeguard mystery.
One of my little obsessions is that ‘lifeguarding’ is a very common teenage summer job in Scotland, but apparently nowhere else in the UK. I don’t know why.
Now, every time I talk to a Scottish person, I have to stop asking them, “Did you work as a lifeguard when you were a teenager?” But whenever teenage jobs are brought up, >>65% of Scottish people I meet say they were lifeguards, compared to <<5% of English, Welsh and (Northern) Irish people. Of course I'm not sampling randomly, but that's crazy.
The BBC is even producing a lesbian romance drama set in Helensburgh, the premise of which is that being a ‘lifeguard’ is the default job for people looking for work to earn money while living with their parents.
I don’t understand.
11. New Yorkers are against noncommutative algebraic structures.
music and podcasts
1. Stevie Wonder, Innervisions. I forgot how great this album is. My favorite songs are Don’t You Worry ‘Bout a Thing and Too High. With the exception of some backing vocals and minor synthesizers, Stevie Wonder plays all instruments on this album. This is completely insane. Related “You’ll Hear It” episodes include:
2. Podcast of the Month: Dean Ball talks about how he created an AI action plan. “How do you make sane, intelligent technocrats almost isolated in a regime run by madmen?” Will this be the defining political question of the late 2020s?
3. Lauryn Hill, The Miseducation of Lauryn Hill. I realized that I should blog more about genres other than jazz and classical. I was inspired to listen to this again after Apple Music declared it the best album of all time. He’s probably my second favorite rapper after Andre 3000. It’s been a while since I’ve thought about such a thing. My favorite song is To Zion, but I recommend listening to the entire album in order.
4. Hugh Mellor tells the truth about Ramsay. I never got around to reading Truth and Probability, but it’s interesting for two reasons. First, I’ve heard philosophers argue that Cheryl Misak’s biography places Ramsey squarely in the pragmatist tradition, but I haven’t waded into this argument. Second, because of the breadth and depth of his thought, running a tiny little conference on Frank Ramsey would be a dream come true.
5. Marginal Revolutionary on the Baumol Effect. This is the original book in which Baumol proposed the idea with William Bowen.
6. Ahmad Jamal Trio “Awakening”. Notice how different the energy is on “Dolphin Dance” compared to Herbie Hancock’s original (I’m also re-listening to that album, it’s one of my favorites). You’re probably familiar with the lick around 0:40, as it’s often sampled in hip-hop.
7. Dan Wang and Stephen Kotkin talk about how historians do their jobs. This works better as a video podcast. You can hear more about Kotkin’s time living in Magnitogorsk in one of my favorite podcast episodes, the Conversations with Tyler podcast.
paper
1. Matt Clancy, How to Accelerate Technological Progress. One of the things I appreciate about Matt is that you can download his web pages as properly formatted PDFs. This is an overview of some of the material on New Things Under the Sun. The most interesting new thing I learned was a review of how long it takes to go from science to technology (answer: about 20 years) and how common independent discoveries are. I assigned this paper as one of the core readings for an economics class I teach at the University of Edinburgh (long story, I’ll post it soon).
2. The impact of various advanced AI systems on democracy. I’m currently in not one, but two reading groups, of which I’m the only one who isn’t a professional political scientist. Of the long list of authors, I only know Bruce Schneier and fellow interactor Saffron Huang.
The second page reviews some of the literature on political bias in AI. LLMs say they are biased toward “progressive/liberal” views (aren’t these often opposites? What does this mean?) when asked multiple-choice questions, but are politically neutral when given an open-ended response. My realistic feeling is that political bias is difficult to measure under current usage.
I generally had a pretty low opinion of Nature Perspectives and Nature-related editorials in general. One aspect of the “false rigor” that this paper excels at is that this 10-page relatively vague and milquetoast opinion cites a whopping 141 sources (and you don’t get the impression that the author actually read them all). Can you honestly say that this is a greater contribution to human knowledge than a thoughtful Substack post by one of them?
Footnote 9 reminds me that at some point I need to blog about James Fishkin’s deliberate polling methodology in the context of whether we can get the people of Northern Ireland to agree on something so we don’t have to dive into these sources. Nan Ransohoff recently had the misfortune of listening to me talk about this literature in more detail than I wanted to explain the reasons for my answer to her question, “What can other countries learn from what Ireland has done well in politics?” It was not a “citizen’s meeting.”
Footnote 60 cites Hugo Mercier’s heretical “argument theory of reason” as “evidence” for the following claim:
Hosting large-scale in-person events is expensive and time-consuming, and in-person discussions can be susceptible to social desirability bias. In social desirability bias, interlocutors are motivated more by the desire to win an argument than by reaching a mutually acceptable outcome.
Does the claim that people sometimes want to win an argument rather than reach a mutually agreeable outcome really need to be cited? Even if the answer is somehow yes, why is that quote a very specific and very controversial hypothesis in cognitive science that has little to do with the claim stated in the text? (Sorry for being mean.)
This also reminded me that I had an absurdly bloated 10,000 word draft of a blog responding to Mercier and Sperber’s Argumentative Theory of Reason, but I couldn’t finish it in time to submit it to the Astral Codex Ten Book Review Contest. It may or may not see the light of day. If you would like to read a copy and let us know if it is repairable, you can do so here.
This was the second time in the political science group that I instinctively sympathized with the idea of a paper, but was frustrated by its sloppy execution. Some people suspect that if I keep coming across as too pedantic, they’ll kick me out.
movies and videos
1. Joshua Oppenheimer, The Act of Murder. This is a documentary about the genocide of Chinese and Chinese people suspected of being communist sympathizers after the 1965 coup that brought Suharto to power in Indonesia. Given that the post-democratic government remains sympathetic to Suharto, publicly available archives are minimal and it is unclear how many people were killed, even in millions. To make this film, Joshua Oppenheimer spent several years in Indonesia, filming the perpetrators who willingly volunteered to re-enact executions, and learning to speak fluent Indonesian, partly with the intention of befriending them.
I think The Act of Killing is an incredible achievement and deserves to be considered perhaps the best documentary of all time. As someone who has watched a fair amount of Ken Burns, this is on a whole other level.
2. Finally, from YouTube, we’d like to introduce you to Bob James and Tame Impala from NPR Tiny Desk. I’m also watching videos of China’s Cultural Revolution. Finally, a speculative story: cosmological natural selection as a conceptual infrastructure for AI coordination. This is from the ODYSSEY conference, the second installment of the ILIAD series of coordination events. Confusingly, the all-caps title does not appear to indicate an acronym, although some nicknames are suggested (“International League of Intelligent Agent Deconfusion”). We’re buying dinner to the first person to come up with a great AI safety backronym for EPIC OF GILGAMESH.
You can read Sam’s full November link here.
[1] It’s also a good time to remind people that value-added tax causes less unnecessary pain than sales tax.
(0 comments)
Source link
