1. Cloning of Macaque Monkeys by Somatic Cell Nuclear Transfer. Here’s the paper in Cell. The scientists are from Mu-Ming Poo’s Institute of Neuroscience in Shanghai. They used the same technique as Dolly the sheep, which previously hadn’t worked in primates. How? “We found that injection of H3K9me3 demethylase Kdm4d mRNA and treatment with histone deacetylase inhibitor trichostatin A at one-cell stage following SCNT greatly improved blastocyst development and pregnancy rate of transplanted SCNT embryos in surrogate monkeys.” Ummm? I’d paraphrase as they discovered a trick. Injecting a couple of chemicals after cloning helped turn the cloned cell back into a stem cell. There’s two takeaways. First, it took them 127 attempts, to get 109 embryos, to get 21 surrogate monkey mothers, to get 6 pregnancies, to get 2 live births. Which now puts primates on par with sheep in terms of success rate using this technique. That is, still very low. Second, China don’t care. They’ve moved on. I doubt anyone doing this kind of work in China worries about what backwards countries in the west think of cloning or gene editing. Best popular story I read is here in the Atlantic.
2. Amazon Go new retail store with cashierless checkout. You come in, make sure your phone IDs you, let Amazon’s cameras and sensors automatically track what you pick up and put into your bags, then just walk out. No check out, or self check out. This is awesome! Checkout is the worst. Amazon is deliberately vague about the technology, but it includes face recognition and constant video monitoring. By far the best business and tech monopoly analysis is by Ben Thompson. Recommended. Also see Om Malik, who has a good bullet point summary. Plus the obligatory China did it already story. But note that Chinese company Bingobox uses RFID and you still have to self scan, so not as sophisticated as what Amazon is doing/attempting. For now at least.
3. Date for oldest homo sapiens outside Africa pushed back to 180k years ago. Link to paper. About 15 years ago, there was a reasonably strong argument modern humans left Africa only once about 50k years ago. By 2015 it was pushed back to 100k years ago. Now it’s 180k. To be clear, ancient DNA still seems to point to the earlier human migrations getting overrun by the one from 50-60k years ago. Takeaways? The debated find in Morocco dated 300k years ago seems more likely to be Homo sapiens after all. And it’s less clear how much we should consider Africa as separate from Arabia biogeographically, at least for humans. As Razib Kahn notes in this context: “Ever so slightly our priors as to an African genesis for our modern lineage are getting weaker.” Here’s Ewen Calaway with a good summary in nature. And Jerry Coyne with a step by step walk through of the paper.
4. AI in Medicine: Outperforming Humans since the 1970s. Jason Collins has a post where he transcribes portoins of an a16z podcast on AI in medicine. Collins focuses on the part where they discuss a 1978 text based expert system which was better than human pathologists. But wasn’t adopted due to incentives. What’s important to understand is incentives are unchanged from the 1970s. Inventing AI technology for better health is easier than changing the incentives so that better technology gets used. True for AI back then, true for AI now. My confirmed priors here are Hansonian, medicine is not about health. Also, see the health chapter from Kevin Simler and Robin Hanson’s new book.
5. Branko Milanovic argues Nassim Taleb is one of the most important thinkers today. Taleb is often strident, mind numbingly repetitive, and annoying. Side note: I’ve expressed this on twitter before, and recently realized Taleb blocked me. I am shocked! Anyway. I do find value in reading (or skimming depending on the section) his books. So I really enjoyed this Branko Milanovic post where he ties Taleb’s writing together and defends the whole perhaps better than Taleb could. Milanovic ends “very few people are able to create systems of thought that go across multiple disciplines and display internal coherence. This the uniqueness and importance of Nassim Taleb.”
6. Vengeance As Justice: Passages I Highlighted in My Copy of “Eye for an Eye”. This is the title of a post from Tanner Greer on William Ian Miller’s book Eye for an Eye. The book is about governance by honor culture. Which is the default way humans governed themselves for millennia. Here’s one graf:
These type of questions naturally lead to the topic of this book: lex talionis, the law of the talion, the principle of an eye for an eye, of justice through vengeance, retaliation sanctioned by culture and law. This understanding of justice is what propels the Icelandic sagas. But it wasn’t just a Viking tick. “Eye for an eye” was standard practice just about everywhere a few thousand years ago, from the shores of Germainia and the fields of the Greek polis to the warring tribes of Canaan and the even more distant lands of the Kurus and the Zhou. We view this understanding of justice as backward and crude. We say things like “an eye for an eye makes the whole world blind.” Miller aims to convince us otherwise. We have a lot to learn from these talionic cultures, he argues, and our world could be made a more just place if we could humble ourselves enough to learn from them.
What jumped out at me while reading Greer’s post, is the striking parallel to Robert Trivers’ tit-for-tat explanation for reciprocal altruism in evolutionary biology. Trivers highlighted altruism. But tit-for-tat (literally blow-for-a-blow) means altruism and punishment are two sides of the very same coin. Such is our nature.
Praxtime by Nathan Taylor 