Saturday links 16-Feb-2019: Campus disinvitations, reducing suspensions backfires, college loans, Fortnite is all about the social

1. Campus disinvitations are way down. People have been freaking out about disinviting college speakers (in particular conservatives) for the past two years. The chart below is great, because it means now we can all go fight about something else instead. link

source: Niskanen Center

2. Stopping school suspensions backfires. New results from a large scale test of “restorative practices”, which means greatly reducing school suspensions. With a goal of helping minority students. Unfortunately it backfired. The policy disproportionately hurt African American students. The common sense reason is simple. Disruptive kids make it hard to teach. And if you don’t suspend disruptive kids, their behavior hurts at-risk kids the most. Kevin Drum is a supporter. So he says we should “take the results seriously enough to try to figure out how programs like this can be improved. We should give up on them only if we do that and they continue to fail.” If you want a harsher take, go here.

3. College student loan burden. Let me quote from the piece:

For many of us, student debt means delaying — if not entirely forgoing — homeownership, marriage, and parenthood. This new form of social stratification — between those who have student debt, and those who do not — will have ramifications for generations to come.

Millennials didn’t adopt new ideas about college. We inherited and internalized the old ones. But the cost of those ideas has risen exponentially in the past few decades. In 1983, the average full-time student borrowed $746 ($1,881 in 2018 dollars) per year. The most recent statistics from the College Board indicate that in 2018, the average annual undergraduate loan is now $4,510, while the average graduate loan hit $17,990. In 2016–2017, the average borrower left college with $37,172 in loans.

The above is an excellent point. The old norm was go to the best (most expensive) college you got into, and don’t worry about choosing a major with an eye on salary. Any degree automatically pushed you into the upper middle class. Student loans are no big deal. This was true. But isn’t any more. Yet the enticing norm lives on, zombie-like. Along those lines, College May Not Be Worth It Anymore:

Since 2000, the growth in the wage gap between high school and college graduates has slowed to a halt; 25 percent of college graduates now earn no more than does the average high school graduate. Part of the reason is oversupply. Technology increased the demand for educated workers, but that demand has been consistently outpaced by the number of people — urged on by everyone from teachers to presidents — prepared to meet it.

Why do employers demand a degree for jobs that don’t require them? Because they can.

4. Trilemma of speech moderation. Tyler Cowen argues regulation of speech on the internet faces an unsatisfactory trilemma: “When it comes to private platforms and speech regulation, you can choose two of three: scalability, effectiveness and consistency. You cannot have all three. Furthermore, this trilemma suggests that we — whether as users, citizens or indeed managers of the platforms themselves — won’t ever be happy with how speech is regulated on the internet.” Kevin Drum has a useful illustration, from his experience moderating blog comments:

  • If the blog was small, I could easily moderate comments and do it consistently.
  • If I was willing to spend lots of time on moderation, I could manage a large blog with consistent comment policies.
  • If I decided not to worry about consistency, I could manage a large blog without putting a lot of time into comment moderation.

5. Planet Labs has 200 Earth observation satellites. The smallest satellites are the size of a shoebox. So it’s all about volume. They have the largest satellite constellation in history, and can image the entire Earth once a day. Nothing special here, just thought it was cool. link

6. Three notes on China. First: China faces a housing glut, leaving 65 million empty apartments. Second: Why did Americans ignore the Unkrainian famine caused by Stalin in 1932-33 and do nothing? For the same reason we’re doing nothing about Uighur concentration camps now. We gots way too many other things to worry about. Third: “Xi Study Strong Nation” is a mobile phone app that gives you points for reading/watching Communist Party content. In particular I found this striking:

When I went home over winter break, I discovered that my mom, who generally didn’t use her mobile so much, was on it every single night, and even would not sleep until late into the night (while she generally was asleep by 10PM). I found this really strange, and only after I asked about it did I learn that the education committee at the subdistrict level had ordered teachers at all schools to download an app called “Xi Study Strong Nation,” and to earn points by fulfilling various tasks every day.

7. Fortnite is all about the social. Fortnite is a massive gaming hit. But you knew that. What I didn’t know was how important social was to the game’s allure. Quote: “Fortnite intended to merge specific shooter dynamics with the sandbox nature of Minecraft so that players could define their own style of play.” And: “Fortnite has become a daily social square – a digital mall or virtual afterschool meetup that spans neighborhoods, cities, countries and continents. This role is powered by Fortnite’s free availability, robust voice chat, cross-platform functionality, and collaborative gameplay. Accordingly, examples abound of kids, adults and families simply hanging out or catching up on Fortnite while they play.” Interesting longform if you’re curious. link

8. NASA’s Mars Opportunity rover is dead. Opportunity had a planned 90 day mission, but lasted 14 years on Mars. Main discoveries were gypsum and hematite, signs of Mars’ far wetter past. My favorite image is below.

Opportunity's northward view of tracks
source: NASA

And that’s all for now. Thanks for reading.

Saturday links 9-Feb-2019: climate and epidemics, ur DNA r belong to us, Alzheimer’s gum disease, Aphantasia

Here’s a list of links/commentary on what I found interesting this week.

1. European arrival in Americas may have caused enough death to cool climate. When Europeans discovered America in 1492, they brought along smallpox and measles. The resulting mass epidemics killed roughly 90% of the population, dropping from 60 to 5 million. Previously cultivated land, roughly the size of France, became reforested. The new plant growth lowered CO2 levels, cooling the climate. In particular causing the temperature downturn in the 1590s. This idea has been around since 2003. A new paper adds to the argument. It’s a disturbing and headline worthy idea.

What’s certain is the mass American dying and resulting reforestation happened. And on the margin this lowered CO2 levels. Fine. What’s less certain is how big this effect was compared to everything else. The paper argues noticeably big. It’s a cool paper. The BBC news story is here. And if interested, read a critique here.

2. All ur DNA r belong to us. In April last year the police solved a murder cold case by searching GEDmatch, a public DNA database. They only found 3rd and 4th cousins, but that was enough to narrow the search to 1000 people. And from there they eventually solved the murder. GEDmatch has about a million people in it. It’s open source, in the sense that people who upload their DNA have agreed to allow anyone else to use it.

What’s new is Family Tree DNA, a consumer DNA testing company, with 2 million DNA profiles on file, has been allowing the FBI use their DNA database to find criminals. And they did this without informing their users until Buzzfeed broke the story. People are rightfully upset. This is a breach of privacy and terms of service. That said, as we saw in the GEDmatch case, the public seems generally supportive of using DNA to identify criminals. So I don’t think things won’t go back to how they were.

People believe their own DNA is private. But even though only 3% of your DNA is shared by 2nd cousins, that’s enough to identify your pedigree. And even 3rd or 4th cousins are enough to find you if the FBI is motivated enough. Given shared ancestry, DNA information is far more of public good than most naively assume. All ur DNA r belong to us.

Of course this means foreign governments can use DNA to track down relatives of dissidents, and tabloids can use DNA to snoop on celebrities. So which justifications for a search are legally allowed matters. But the larger point is we’re not heading toward a world where anyone can be tracked by their DNA. We live in such a world already.

3. Jay Rosen is worried journalism for the 2020 presidential election will be like 2016. Here’s Rosen: “it’s become clearer and clearer to me that, without intervention, coverage of the 2020 campaign is likely to be a disaster for everyone except Trump and his core voters, who want to watch it all burn anyway.” It’s hard to know what’ll happen in 2020 given how much is in flux now. But agree the dynamics around journalism and Trump are unchanged from 2016. Rosen says journalists should spend less time on horse race journalism, pumping out articles on who’s ahead and who’s behind, and spend more time focusing on issues voters care about. We’ll see. Twitter thread is here, and interview is here.

4. Gum Disease and Alzheimer’s. The more sensational version of the news was We may finally know what causes Alzheimer’s – and how to stop it. (Answer: bacteria behind gum disease). This was then countered by No, we don’t know that gum disease causes Alzheimer’s. With the subtitle “A new study suggests a link between oral bacteria and Alzheimer’s, but it’s far from proven.” Which is the correct take. As for me, am I flossing more often now? Yes. Yes I am.

5. Cowen on Pinker on the Enlightenment. Steven Pinker’s book Enlightenment Now is, in my view, excellent. But the book has been attacked for being naive. Pinker responded. Then Tyler Cowen got to the heart of the matter when he noted Pinker reserves the name Enlightenment only for those aspects he likes. Not enough grey. In particular I like this bit: “I am very much an admirer of Pinker and his work, and I consider myself an optimist, especially across longer time frames.  But what is sometimes called progress does also have a dark side, and we will do better fighting that dark side if we are clearer — in our own minds and with each other — on how things have run to date.”

6. Aphantasia, I have it. Aphantasia is an inability to visualize images in your mind’s eye. So imagine an apple. Close your eyes and see it. If you have aphantasia, you can’t do that. There are matters of degree. Some people can vividly imagine pictures. Others less clearly. But if you can’t to it at all you have aphantasia. It was a shock a few years back to read about it because, having aphantasia my whole life, I had always assumed everyone was like me, and “mind’s eye” was just a metaphor. If you’re curious, this account of how shocking it was to find out about aphantasia was similar to my own. Oh. I see. I see! So that’s why book authors waste so much time on elaborate visual descriptions of the countryside. Who knew? Remembering faces is a bit harder for me than most. I can’t visualize, but if I make an effort, or just see someone often that’s fine. GPS is a godsend. My kids sometimes complain I use it for places we’ve been many times, but I love it. And my case is relatively mild. If I focus hard I can sometimes sort of imagine seeing a picture. But it’s rarely worth the bother.

I suppose this is neither here nor there. But perhaps I wanted to share because it made me realize we can go through life assuming too much. Leading separate lives, we but dimly perceive our own mind’s eye. Best to reach out and share and learn from others, letting their outside views make us see more fully a complicated, if occasionally beautiful, world.

That’s all. Thanks for reading!

Apple revoking Facebook’s certificate was a strategic err which may come back to haunt the tech industry

image credit: Peter Sellers in Dr. Strangelove, Or: How I Learned To Stop Worrying And Love The Bomb

Let me explain why I think Apple’s decision to revoke Facebook’s developer certificate was a strategic err. Not just for Apple, but for tech in general.

First the background. For an app to run on an iPhone, it has to be signed by an Apple certificate. Apple provides app developers with both an enterprise certificate, with extra privileges for internal iOS app development, and a general certificate, for the app store. Facebook broke Apple’s rule in using the enterprise certificate to sign a Facebook VPN app (giving it more privileges) for an app going to the App Store for regular customers. This app could then collect information on which other apps were installed on user’s phones, and what data traffic was sent. This was a violation of the rules, a privacy breach, and was done knowingly. Make no mistake. Facebook was bad bad bad.

In response Apple revoked Facebook’s enterprise certificate, which in turn shut down all iOS development and testing inside Facebook. It’s the nuclear option. Then it turned out Google had done something not quite as bad but similar, so (to be fair), Apple shut down Google as well. And now it turns out there are other companies which violated the terms too. After a few days, Apple re-instated the certs for everyone, so now things are getting back to normal.

Here’s my point. As Ben Thompson noted: “I wouldn’t be surprised if, in the long run, the company comes to regret this move.” This is exactly correct. I believe Apple made a strategic err they will regret. Even though it’s popular right now. For example John Gruber “Give me a break. Everybody knows that Apple’s enterprise certificate program does not in any way permit distribution of this kind of software, which wouldn’t be allowed in the App Store. But that’s exactly how Facebook was using it. There are pros and cons to Apple’s iron-clad control over native apps on iOS. This incident with Facebook is one of the pros.”

Both can be true: 1) Facebook gets away with too much and absolutely deserves severe punishment, and 2) Apple going nuclear was a mistake. No matter how deliciously satisfying. No matter if the rules said they could revoke Facebook’s certificate. Rules should be enforced with an eye toward precedent, with an assumption bad actors will eventually latch on to those same precedents.

Here’s an analogy. Suppose back in the day when Netscape had just invented the web browser and was riding high, and Microsoft was the evil empire (think circa 1997), Microsoft shut down all Netscape windows development for a rules violation. It’s similar to what Apple (good company) did to punish Facebook (bad company). But in this hypothetical the villains are swapped. It’s Microsoft (bad company) punishing Netscape (good company). While this couldn’t happen back then, with cloud centralization it can now. As Apple just demonstrated.

The cold war between the Soviet Union and the US was very fraught, but the one thing both sides got right was showing restraint with nuclear weapons, either directly or in proxy wars. The cold war bomb threat was invoked all the time. Fine. But leaders showed restraint. Knowing if one side (the good side) used them, even if justified, then the other side (the bad side) would run with that precedent and also use them. Apple could have played the brinkmanship game, publicly threatening to revoke the cert in a Bay of Pigs style confrontation with Facebook, and gotten much the same result, but without actually firing nukes.

It’s unclear exactly how and when this may come back to bite. But as a hypothetical imagine a startup using modern centralized platforms such as Amazon AWS, Microsoft github, Google maps, etc etc. And Amazon, Microsoft, Google put in place complex developer rules on what’s not allowed. So it’s hard to comply. Then if that startup later competes against one of those companies, the platform owner has the temptation to stop that company dead by disallowing development on their platform on the pretext of a rules violation. Threatening may be enough. And I’m not saying Amazon, Microsoft or Google in particular will do this. Rather that in the cloud era all centralized cloud platforms face this temptation.

Apple set a precedent. Nukes are now allowed. Even if the letter of the law says you can do it, restraint is often a wiser choice. There will be increased temptation to create platform rules which are easy to violate. Laying a trap. Just in case. Then, if the mood strikes, bombs away.

image credit: Major Kong (Slim Pickens) riding a nuke in Dr. Strangelove

Wednesday links 30-Jan-2019: Cognition models, Marx and virtue signalling, slime, iterated embryo selection, screentime

1. Scott Alexander on cognition models. Reading the book On Intelligence by Jeff Hawkins when it came out in 2004 was a revelation to me. That’s because Hawkins provided an overarching theory for how cognition works, with his memory/prediction model. From page 99: “The recalled memory is compared with the sensory input stream. It both ‘fills in’ the current input and predicts what will be seen next. By comparing the actual sensory input with recalled memory, the animal not only understands where it is but can see into the future.” Reading this in 2019, it seems unsurprising. And the framework is not original to Hawkins. But it was the first time I’d encountered a well written, convincing and unified framework for human cognition. Mind is prediction. At a base level, sense data constantly flows in, and predicted future sense data constantly flows out. At a higher level, mind is prediction on how potential possible actions can affect the world. An overarching framework is critical for science, because it stops you from drowning in details. It’s like completing the outside edge of a puzzle. Once that’s done, the frame guides you on where to place all the other pieces.

So I really liked this post by Scott Alexander, where he created his “Grand Unified Chart” below. It 100% matched my priors 🙂

Scott Alexander explains:

All of these are examples of interpreting the world through a combination of pre-existing ideas what the world should be like (first column), plus actually experiencing the world (last column). 

All of these domains share an idea that the interaction between facts and theories is bidirectional. Your facts may eventually determine what theory you have. But your theory also determines what facts you see and notice. Nor do contradictory facts immediately change a theory. The process of theory change is complicated, fiercely resisted by hard-to-describe factors, and based on some sort of idea of global tension that can’t be directly reduced to any specific contradiction.

Why do all of these areas share this same structure? I think because it’s built into basic algorithms that the brain uses for almost everything (see the Psychology and Neuroscience links above). And that in turn is because it’s just factually the most effective way to do epistemology, a little like asking “why does so much cryptography use prime numbers”.

Last point. The full title of Hawkins’ book is On Intelligence: How a New Understanding of the Brain Will Lead to the Creation of Truly Intelligent Machines. That is to say, we could add a new row to Alexander’s table for artificial intelligence. It slots in perfectly. (And yes, I’m aware some would consider today’s AI a subfield of Bayesian stats). We’re still a long way from general AI. But the point here is the framework to judge AI progress already exists. If mind is prediction from sensory streams, generalized mind (AGI) requires an ability to go beyond narrowly specific domains.

2. Tanner Greer on virtue signalling. Greer first reminds us how Marx believed economic structures and institutions drove class consciousness, which in turn determined people’s ideas. Ideas are nothing more than an expression of class interest. Which means, as Greer says:

A straight line can be drawn from statements like these and the great terrors imposed upon captive populations by Marxist dictators like Stalin, Mao, and Pol Pot. Men like Stalin did not oppose freedom of expression simply because it threatened their personal power. They opposed freedom of expression because they earnestly believed it was a pointless exercise. There is no point in debating with representatives of the bourgeoisie when the ideas of the bourgeoisie were not pliable to debate. If the opposing stance was not the product of reason, it could not be undone by reason.

Then to Greer’s analogy to signalling theory:

The central proposition of signalling theory is that the vast majority of arguments made in the public sphere are not made in good faith. An outraged tweet is not written to express genuine emotion, but to signal solidarity with the ‘right’ side. A verbose blog post is not written to persuade its readers of its argument, but of the cleverness of it author. A well circulated censure of some racist act is not written to convict the racist, but to display the Wokeness of the censor. The connecting string in all of these cases is that your arguments are less about your ideas than shaping other people’s perceptions of you. Whether you believe you are writing primarily to shape other people’s perceptions of you is immaterial. As with the Marxist theorists, signalling theorists are happy to conclude that signalling does not need to be a fully conscious process. In place of a class consciousness imposed by the material circumstances of an individual’s social status, signalling theorists trace the origins of self-interested arguments to mental social-status ‘modules’ imposed by the material circumstances of an individual’s evolutionary heritage.


Civilized discourse depends very much on both parties in duel of words recognizing the self-serving intentions that lie behind the other party’s speech… and then deciding not to mention them. This is Reason’s Pact: the bare minimum required for fruitful rational discourse to take place. We underestimate how many institutions in our society depend on us maintaining this illusion. Jettison the ideal of ‘reason’ as a governing principle, and all you have left are words as war. 

One quibble. Greer notes “Robin Hanson, doyen of modern signalling studies, fights a perpetual battle to convince his readers that none of the ulterior motives he ascribes to human sociality writ large are behind his twitter polls.” Fair point on the irony of Hanson being attacked for impure motives and signalling. But I would add in Hanson’s defense, I’ve not seen Hanson use that style of argument against his accusers. See this example. In fact the central argument of Hanson’s book on signalling is social policy could be improved if we take signalling into account. That is to say, I read Hanson as a nerdish theorist who writes about signalling with the hope of finding ways to work around it, leading to better social policies. Though perhaps that doesn’t matter. It’s being weaponized regardless. Quibble aside, an outstanding post. Worth reading in full.

3. No One Is Prepared for Hagfish Slime. It’s yucky. Has a great picture of a slime-covered Prius. And is written by the excellent Ed Yong. Recommended.

4. Iterative embryo selection. CRISPR gene editing is not reliable enough to do the many 100s or 1000s of edits it would take to dramatically impact complex traits. But one way around this problem is through iterative embryo selection: “conducing multiple generations of embryo selection in a petri dish by exploiting gametogenesis or stem cells.” So you use IVF, get an embryo, do gene edits, then replicate those embryos to get a new set of cells, select those without side effects, and repeat. The selection step allows you to weed out bad edits. This is not completely workable now. And will be applied to livestock first. But the technique will accelerate when CRISPR can do large numbers of complex edits. This was a new idea to me. link

5. Screentime worries overblown. New paper with N=355,358 argues that screen time worries are overblown, if hard to pin down. The effect of wearing glasses is bigger. Paper here. News article here.

6. On desiring your enemies to be evil. I was going to link to some Covington stories, but decided not to. Here’s C.S. Lewis from his 1952 book instead:

Suppose one reads a story of filthy atrocities in the paper. Then suppose that something turns up suggesting that the story might not be quite true, or not quite so bad as it was made out. Is one’s first feeling, ‘Thank God, even they aren’t quite so bad as that,’ or is it a feeling of disappointment, and even a determination to cling to the first story for the sheer pleasure of thinking your enemies are as bad as possible? If it is the second then it is, I am afraid, the first step in a process which, if followed to the end, will make us into devils. You see, one is beginning to wish that black was a little blacker. If we give that wish its head, later on we shall wish to see grey as black, and then to see white itself as black. Finally we shall insist on seeing everything — God and our friends and ourselves included — as bad, and not be able to stop doing it: we shall be fixed for ever in a universe of pure hatred.

And that’s all for this post. Thanks for your time!

Saturday links 19-Jan-2019: Gravitization of quantum mechanics, Revolt of the public, income sharing for tuition, Big 5 personality quiz

1. Roger Penrose on the gravitization of quantum mechanics. Physicist Roger Penrose was on Sean Carroll’s podcast (recommended). I want to highlight a particular point Penrose made at the 1:16 mark. As you may know, the theories of gravity and quantum mechanics are not reconciled. The problem is general relativity (gravity) treats space as continuous, not quantized. While quantum theory makes all interaction quantized.

One approach is to reformulate gravity to match quantum mechanics. This approach is called quantizing gravity. String theory is an example. Testing String theory (unless an indirect method is discovered) requires getting down to the Planck scale (very tiny, meaning super energetic particles). As Penrose put it on the podcast, it would require a particle accelerator the size of the solar system. Indeed.

Penrose argues we should go the other way around. Rather than reformulate gravity to match quantum mechanics, reformulate quantum mechanics to match gravity. This approach is call the gravitization of quantum mechanics (see paper). Penrose said testing this is just about within experimental reach now, using bose einstein condensates.

So what’s the point here? Funding, funding, funding! The world’s largest particle accelerator is the Large Hadron Collider (LHC) at CERN. CERN has just announced a proposal for an even larger collider, at the cost between €9 and €21 billion. Sabine Hossenfelder argues “it would make more sense to put particle physics on a pause and reconsider it in, say, 20 years to see whether the situation has changed, either because new technologies have become available or because more concrete predictions for new physics have been made.” Exactly so. I have no idea if Penrose’s ideas on the gravitization of quantum mechanics will pan out, but at least we can test them for way less than €9 billion.

2. A good Bandersnatch move review. Bandersnatch is a Netflix movie (technically an episode of Black Mirror) that lets you interactively make choices, resulting in different endings, depending on what you pick. I enjoyed this review more than the movie itself, since it clarified why video games are better suited for interactivity than movies. Here’s one bit:

The reason that letting the audience choose its own story keeps failing when the entertainment industry tries it is that it’s a bad idea. It’s the author’s job to write the story. They can then choose a way to convey that story that gives the reader freedom in how they experience it. But if the story itself is merely a loose collection of different options, each in a different genre and with a completely different tone, then what they’ve created isn’t a coherent work, but a self-indulgent mess—like “Bandersnatch”, in fact.

3. Martin Gurri’s book The Revolt of the Public. I’m about halfway through and it’s excellent (link). Gurri explains why internet publishing has led to nihilistic worldwide populist revolts. I’ll write more once I’m done. In the meantime, here’s a paragraph that captures the core idea:

The information balance of power has changed, of course. A generation ago, the public could exist only as a passive audience Information was dispensed on the industrial model: top down and one to many. That was the great age of the daily newspaper and famous anchormen on the model of Walter Cronkite. The advent of digital platforms, in a sense, created the public. People from nowhere, free of institutional entanglements, pushed the elites out of the strategic heights of the information sphere. Almost immediately, great institutions in every domain of human activity began to bleed authority a process that, as we have seen, in now approaching the terminal stage for many of them. That is my thesis for the revolt of the public.

4. CRISPR gene editing can’t make super-smart babies. This is a pretty basic point, but worth making since gene editing has been in the news. CRISPR/Cas9 gene editing is err prone, resulting in “off target” effects. So editing a single gene works only because you can make multiple attempts, starting over until you get it right. But complex traits like height or IQ are highly polygenic, depending on 1000s of genes. With each gene of very small effect. And editing 1000s of genes isn’t practical. At least for now, it won’t work. link

5. Telecoms companies sell user data to third parties. While many are upset about Facebook privacy, telecom companies are worse. If you care about this kind of thing, this article title says it all: I Gave a Bounty Hunter $300. Then He Located Our Phone.

6. NCAA should allow college athletes to cash in on endorsements. Financial exploitation of college athletes in big name sports is a tricky topic. But this seems like a good way to finesse the problem. Let Nike pay them. link

7. Income sharing to pay college tuition. The way income sharing works is you pay no college tuition up front. Then after you graduate, if you make above a threshold, you pay a fraction of your income for several years. Quote:

Lambda School gives students three payment options. They can either pay $20,000 upfront, pay $10,000 upfront and forgo 17 percent of their salary for a year (with the maximum payment capped at $15,000) or pay zero dollars upfront and forgo 17 percent of their salary for two years (with the maximum payment capped at $30,000). Allred estimates that more than 90 percent of students have opted to pay via an income share agreement.

It’s easy to dismiss this as no big deal. Why would paying before versus after graduation matter? And the New York Times piece is full of hype. But if you believe incentives are critical to making an organization perform, you can see why colleges getting paid depending on someone’s post college job success could be transformational. Incentives are critical. For example see Austen Allred talking about how the school is helping disadvantaged kids do better at job search. On the flip side, I think there’s also an incentive not to get students enrolled in programs where there’s no chance they’ll graduate. That seems harsh, but the current incentives are to enroll and accumulate debt, which is worse. Better to align all students to the programs that will help their careers the most, rather than over promise. Aligned incentives are great!

8. Personality quiz based on big 5 traits. Personality quizzes are normally junk science. And yes, that includes the popular Myers-Briggs test. Psychology breaks personality into the “big 5” traits: extraversion, agreeableness, conscientiousness, negative emotionality and openness to experience. 538 published an online quiz. It’s quick and fun to take. I got no surprises back. I’m more disagreeable and introverted than most. But very open to new experience. A good fit for writing about cool things on the internet, if I may say so. 🙂 Here’s the 538 test if you want to give it a try: link

And that’s all for this week. Thank you for reading!

Sunday links 13-Jan-2019: Apple services, China climate change, superstar cities, falling murder rates

Here’s my weekly list of links with commentary, the first for 2019.

1. Apple’s new services strategy. My new post from earlier today. I argue Apple’s shift to put their movies/TV shows/music services on other hardware platforms was a decision to commoditize the complement. link

2. Climate Change is about China. Noah Smith: “This leads to a painful but inescapable truth — no matter how much they spend, no matter how dramatically they change their societies, the U.S. and Europe won’t be able to put much of a dent in global warming on their own. Yes, the U.S. should ban coal power, tax carbon heavily and spend lots of money on building green energy infrastructure. But without a huge change in China, none of that will matter — the battle against climate change will be lost.”

In particular I liked these two graphs:

Smith’s post is here. Plus also see his related post Saving the Planet Doesn’t Mean Killing Economic Growth.

3. Nuclear power is bad, but better than the alternatives. Many people hate nuclear power. I think the best argument for nuclear power is not that it’s safe. It’s not. But rather all kinds of power generation have risks and problems. Arguing for nuclear power is arguing we don’t live in a perfect world. Solar is great, not magic. Here’s an interview with pro-nuclear activist Michael Shellenberger. One bit:

Look at France and Germany, the two biggest economies in Europe. France has 92% of its electricity from zero emission sources – Germany has 46%. And yet France pays half as much for electricity as Germany because it depends heavily on nuclear and Germany is phasing it out. Germany saw its electricity prices rise 50% over the last 15 years. France’s prices has stayed stable. So introducing huge amounts of solar and wind onto the grid makes electricity expensive; we see it all over the world. The reason has to do with trying to manage all of that unreliability.

4. Economics of superstar cities. Tech naturally clusters, creating superstar cities. Which has been going on since the 1980s. This has coincided with tightening house building restrictions, which has led to decreased mobility into growing cities, at least compared to the past. This is a bit of a hobby horse for me. So am glad to see more people writing about it. Here’s some recent articles which I thought worth reading, if you want to click through:

5. US murder rate decline. The chart below speaks for itself.

New York Time article with chart is here. Plus a nice tweetstorm by Noah Smith on same topic.

6. Some good longreads. I’d like to link to some longer articles which I enjoyed reading over the holidays. I’ll comment on why I thought each one might be worth your time.

  • Online quiz on how you pronounce certain words. Attempts to identify where you grew up. Fun, and it worked for me. link
  • A son discovers only after his death the true extent of his dad’s friendship with NBA basketball star Charles Barkley. link
  • On the poisonous allure of social media shaming. An older topic, written in first person. Surprisingly good. link
  • A personal story on life when weighing 460 pounds. Poignant. The kind of topic that would be maudlin if done badly, but when done well, is universal. This is done well. link

And that’s all for this week. Hope you have a great 2019.

Apple’s new services strategy. Apple is commoditizing (some) of its complements: music, movies, TV.

Apple announced their music and movie services will be available on non-Apple hardware. Whoa! This is huge. Ben Thompson correctly highlighted the importance, saying this shift in strategy (link $) “is not just fascinating, it’s frankly a bit stunning.” And “I think these announcements are a much bigger deal than people realize: Apple is absolutely sacrificing its traditional hardware business model, at least when it comes to TV black boxes and smart speakers, in favor of a content-centric services strategy.”

Why such a big deal? Apple’s incredible success was built on a consistent model: use proprietary software and services to make money selling hardware. That was true for the Apple II, Mac, iPod, iPhone, iPad, Watch. Despite near bankruptcy in 1997, the Apple business model remained sacrosanct. Side note: I’d argue iTunes for Windows with iPod was an exception that proved the rule.

Apple’s vertically integrated software/hardware stack allowed them to create premium experience. And premium profits. But also meant Apple services were hindered strategically. A strategy tax if you will. Because the optimal services strategy (think Google) is horizontal, putting your services on all platforms and devices. The marginal cost of extending an existing service to another platform is low, so in some sense services “want” to be everywhere. More users -> more data -> better services.

Suddenly Apple is putting Apple music on Amazon Alexa speakers. Apple movie and TV shows on Samsung, LG, Vizio and Sony TVs. And allowing AirPlay to those TVs, the ability to send content from your phone to your TV. Apple is destroying the proprietary value of their hardware. Most significantly Apple TV and HomePod, but also to a lessor extent the iPhone itself. What gives?

What’s happening now in 2019 is something I wrote about in 2013, in a post which I think holds up well.

But longer term, say 5, 10 or even 15 years, the Apple services strategy tax creates a third possible Apple bear story. Phones are becoming more and more about services. Let’s say the phone-services split was 90-10 for the first iPhone. It was primarily about what you held in your hand, not what’s in the cloud. Let’s say it’s now at 50-50. You can see where this is going. It’s possible that in a world where the phone-services split is 10-90 the other way, a company which focuses exclusively on services could wind up with a better overall premium phone experience than a company which focuses on complete systems. The strategy tax could actually dislodge Apple from the premium end of the market. Clearly this is a long term speculative scenario, but it’s a plausible future. Especially if you segment the market, where segments focused more on services will reach their tipping point earlier.

Apple makes money selling premium hardware, but what the people are paying for is premium experience. That’s the job to be done. And right now Google maps, Google email, Google voice assistant, etc are all better than Apple. Nonetheless, since I wrote that post in 2013, Apple has continued making enough moderate (if lagging) progress on its services, plus has innovated on its hardware (Face ID, camera, displays, integration of Mac and iPhone, custom ARM processors, integration of Apple Watch and Airpods) to keep their overall experience better. But this approach is showing cracks. Hence the new services strategy.

My guess is Apple’s new services strategy is, in select areas, to commoditize the complements. From Joel Spolsky’s classic post:

A complement is a product that you usually buy together with another product. Gas and cars are complements. Computer hardware is a classic complement of computer operating systems. And babysitters are a complement of dinner at fine restaurants. In a small town, when the local five star restaurant has a two-for-one Valentine’s day special, the local babysitters double their rates. (Actually, the nine-year-olds get roped into early service.) All else being equal, demand for a product increases when the prices of its complements decrease. Let me repeat that because you might have dozed off, and it’s important. Demand for a product increases when the prices of its complements decrease. 

By putting music, movies, TV on all platforms, Apple has decided to stop making those services a premium differentiator for their hardware. They are commodities. Anyone can get them. Even customers who never buy Apple hardware. Hmmm….. what about the $3 billion Apple spent buying Beats in 2014? I’d say flushed down the toilet.

To be fair, there’s another possibility. Apple may be attempting to simultaneously create premium horizontal services with music/movies/TV, and retain their traditional premium vertical stack hardware business. Ben Thompson suggests this possibility when he says “In short, it seems that Apple can have its (services) cake and eat it too: offer exclusive content across all platforms, while preserving its integrated experience (and iMessage) for its moneymakers.”

Perhaps. But the strategy conflict remains. Making a cross platform service means avoiding proprietary customization. Exactly what made those services work better on Apple hardware. That said, I know enough about huge corporations to know how this will play inside Apple. The people in the music/movie/TV services group want it to be best in market. And that’s fine. Good for them! Perhaps they’ll succeed. But even if they become the best, they won’t differentiate Apple hardware. To the hardware teams at Apple, those services are now pure commodities, and the cheaper those commodities become, the more valuable their hardware complement.

The big open question here is which other services at Apple are candidates to go horizontal. If my commoditize the complement theory is correct, it should be all services which don’t move the bar on selling Apple hardware. Let’s try three buckets:

  1. Cross platform now: music, TV, movies, itunes
  2. Cross platform likely: books, all media content, parts of iCloud
  3. Cross platform unlikely: Siri, iMessage, maps, App store, Apple pay

At first I wanted to put Siri into bucket #2. The strategy trade off is acute for voice interface. Google and Amazon can partner with everyone, getting more users and data, making their voice interfaces better and better. Apple lags. But on the other hand, Apple is famously future product oriented. And I started thinking about which future product might benefit most from tight hardware/software integration. The (or at least one) answer is augmented reality (AR) glasses. So even though the pressure for Siri to go horizontal right now is large, I suspect it’ll remain proprietary. I moved Apply pay into bucket #3 for the same reason. Whenever Apple AR glasses arrive, buying stuff with Siri voice and Apple pay will be frictionless. And in tech, frictionless is money. Pure money.

Update January 15 (two days after original post): Apple just announced they are partnering with search provider DuckDuckGo on Apple maps. Maps is a critical strategic capability, so perhaps my guess on Apple’s new services strategy was wrong. That said, I stand by my fundamental point that to the extent a service is available everywhere, it becomes a complement to selling hardware, creating an economic incentive to keep the margin on those widely available services low. And to be clear, partnering with a particular search provider is very different from putting Apple maps on Android. Which remains extremely unlikely. In any case, a lot happening right now with Apple services strategically. Something to keep a close eye on for 2019.

Sunday Links 23-Dec-2018: US-China cold war, moving to superstar cities, internet shaming

I accidentally published this post yesterday with no content. Just a blank post. Sorry about that. Below is the real thing.

1. US-China, begun the cold war has. The Cold War between the US and Soviet Union ran from 1945-1990. With the arrest of the chief financial officer of Huawei (maker of networking and mobile equipment), exports of the opioid fentanyl, and longstanding accusations of theft of tech IP, it seems like the US is headed toward a cold war with China.

Donald Trump instigated the anti-China shift. But both for better and for worse, I suspect it was inevitable. Which means the US-China cold war will outlast Trump. Here’s Ivan Krastev:

In most European capitals, policymakers and the chattering classes want to believe that before too long, Mr. Trump will be gone and the world order — including the close alliances between Europe and the United States — will return. But here’s the dirty secret that I learned in my three months in Washington: That’s not true.

The Trump moment in the end may resemble the Truman moment, when over a short period America dramatically changed its views of the world. This may be hard for Europeans to swallow, but it’s the message I am bringing back with me from Washington. The post-Trump world will not be the pre-Trump world.

Mr. Trump’s presidency has ushered in two significant changes that are likely to have staying power. First, with his administration, Americans have lost confidence in their exceptionalism. 

Second, under the Trump presidency, rivalry with China has become the organizing principle of American foreign policy. Republicans and Democrats disagree on almost everything today, but one area where there seems to be effective bipartisanship is that America must change its policy toward China.

Here’s Tyler Cowen, on similar lines:

Like it or not, the United States is the global hegemon.  In my view this is an overall positive, but for our purposes today let’s just take it as given.

If you are the global hegemon, and another country, largely hostile to your political values and geopolitical desires, engages in widespread subversion of your power and influence, you must in some way hit back.  Otherwise you will not be global hegemon for much longer.  And unlike India or the EU, China desires to build an international political and economic order which would destroy liberalism as we know it.  Imagine a world where autocracy is a much more widespread norm, the Xinjiang detentions and North Korean nuclear weapons are considered entirely appropriate behavior, Taiwan is a vassal state, and few Asian countries could allow their media to print criticism of the Chinese government, for fear of retaliation.  Institutions such as the WTO would persist only insofar as they created loopholes which gave China the benefits of membership without most of the obligations.

What really solidified my views is Peter Turchin’s books, such as Ultrasociety: How 10,000 Years of War Made Humans the Greatest Cooperators on Earth. Turchin argues human society at mass scale came about through the crucible of war. War forged cooperation within large polities, as do or die against foreign threat. External enemies and internal cohesion are two sides of the same coin.

Viewed in this light, the second world war and the subsequent cold war were secret drivers of US cohesion during the latter half of the twentieth century. Which means it’s not social media, Facebook, Trump, the internet, or Russia that’s to blame for our recent outburst of tribal conflict. It’s merely a return to normalcy, after the collapse of our previous enemy the Soviet Union.

Perhaps the lesson here is not to pretend we can achieve a world with no external enemies. Else things fall apart; the centre cannot hold. Human cooperation within requires an external enemy without. If so, the goal should be to keep those inevitable rivalries in check. Avoid hot war. Avoid brinkmanship like the Cuban Missile Crisis. Instead, the best we can hope for is to be like rival sports teams who (mostly) play by the rules. After each round we shake hands and say “good game my friend, good game”.

2. Moving to superstar cities. Below is a chart of the number of people moving per year (blue bars), and the mover rate (black line). Notice the downward trend of the black line. The mover rate was 20% in 1948, declining to 10% in 2018.

source: US census, via Timothy Taylor

Christopher Mims argues that “Technology is creating an economy in which superstar employees work for superstar firms that gather them into superstar cities, leading to a stark geographic concentration of wealth unlike any seen in the past century.” The solution is to have people move to these superstar cities. The problem is housing costs have skyrocketed in these same cities, which means only superstar employees can move there. So we should do more of what Minneapolis just did, remove zoning barriers to cheaper housing in superstar cities. This idea of housing restriction as bad and divisive policy seems to be slowly gaining ground over the past few years.

Along these lines, also see these two recent articles: 1) How rising rents contribute to homelessness, 2) Benefits of upzoning near transit and jobs.

3. Internet shaming. Helen Andrews had an early (2010) experience with internet shaming when her former fiance ranted about her failings as a human being, which went viral on YouTube. She recently wrote about internet shaming more generally. It’s a rather long post, but I thought it very good. Here’s one bit:

The more online shame cycles you observe, the more obvious the pattern becomes: Everyone comes up with a principled-sounding pretext that serves as a barrier against admitting to themselves that, in fact, all they have really done is joined a mob. Once that barrier is erected, all rules of decency go out the window, but the pretext is almost always a lie. ­Matthew Yglesias once claimed that the reason he mocked David Brooks for his divorce was because Brooks had written columns about the social value of marriage, but I do not believe him. He did it because it’s fun to humiliate your political opponents. Moira Donegan claims that she created the Shitty Media Men List—a clearinghouse of anonymous accusations optimally parked for maximum dissemination in the Google Spreadsheet cloud—for altruistic reasons and with no thought of its being used to hurt anyone, but I do not believe her. If it was about protecting women in media from harassment, then why no attempt to sort the true accusations from the false? Why the coy protestations that “I thought that the document would not be made public,” when of course she knew that it would be spread far and wide, or she wouldn’t have bothered creating it?

Read the entire piece here.

4. Track by track breakdown of Fleetwod Mac’s song Go Your Own Way. If you are interested in how pop songwriters put together songs, Song Exploder is the podcast for you. How much I like it depends on the song and artist. But in this recent episode they picked an all time classic pop song. Lindsey Buckingham does the track by track (drum part, acoustic guitar, vocals, base, etc) walk through of how the song was put together. I thought it was excellent. link

And that’s all for today. Thank you for your time!

Sunday Links 16-Dec-2018: Bitcoin bubble or scam or platform, Waymo cars, population mountains

Here’s what I’ve read recently that’s worth linking and commenting on. 

1. Is bitcoin a scam, a bubble, or a (nascent) platform?  I’d answer yes, partial yes, and a maybe. Let’s go one at a time.

First up: scam. Kevin Drum argues Bitcoin Is a Long Con Aimed at Those Least Able to Afford It. Quote:  “As near as I can tell, the Bitcoin market is split between cutthroat Chinese miners running huge racks of servers, and hopeful but clueless marks who would be better off putting their money into lottery tickets. So this is the test: Are you a cutthroat Chinese miner running huge racks of servers? No? Then you’re one of the clueless marks. Sorry.”  The data here is clear enough, with a study finding 80% of ICO offerings in 2017 were scams. One of the most famous being Centra, where music producer DJ Khaled and boxer Floyd Mayweather Jr had to pay fines to the SEC. I disagree with Kevin Drum that bitcoin is only a scams. But the answer is yes, there are plenty of bitcoin/ICO scams.

Second up: bubble. Below are two charts which I created using The first gives bitcoin value with a linear scale. The second uses log scale, which makes it easier to spot bubbles.

1. Bitcoin market price in USD, linear scale

2. Bitcoin market price in USD, log scale

Looking at the data, Noah Smith argues Yep, Bitcoin Was a Bubble:

Formally, an asset bubble is just a rapid rise and abrupt crash in prices. Defenders of the efficient-market theory argue that these price movements are based on changes in investor’s beliefs about an asset’s true value. But it’s hard to identify a reason why any rational investor would have so abruptly revised her assessment of the long-term earnings power of companies in 1929, or the long-term viability of dot-com startups in 2000, or the long-term value of housing in 2007.

Scott Sumner disagrees with Smith, arguing Bitcoin is not a bubble:

His [Smith’s] main piece of evidence is that the price rose sharply and then fell sharply.  But that sort of price pattern also occurs in 100% efficient markets that are highly volatile because the fundamental value of the asset is hard to ascertain.  And if there ever was an asset with a value that is difficult to ascertain, it’s Bitcoin.  I have no clue as to what Bitcoin should be worth, and I doubt anyone else does either. 
Bubble theories are only true if they are useful, and they are not useful.  The people who said it was a bubble at $30 were implicitly giving you advice not to buy.  Ditto for those who said it was a bubble at $300.  This advice was exceedingly non-useful; in fact if you followed the advice of bubble proponents you missed out on the opportunity to earn a massive profit investing in Bitcoin.  That’s why I don’t follow the advice of bubble theorists; it’s not useful.  BTW, I don’t own Bitcoin for unrelated reason; I prefer index stock (or bond) funds.

I happen to believe Smith and Sumner are both correct, with their disagreement being (mostly) semantic. Let me explain. Sumner argues declaring bubble offers no predictive value. You can’t make money from it. Bitcoin bubble has been declared again and again. Are the bubble declarer’s now rich? Nope. He hates the term bubble. Price volatility by itself doesn’t prove bubble. Some assets are just hard to price. I believe Smith would concede the point about predictive value. 

Smith’s additional point is bitcoin prices were driven by irrational emotion and momentum. And here I think Sumner would concede (though he’d argue all prices work more or less that way). 

So in the end whether to declare bubble is semantic. It’s a no from a “true value” point of view, since true value pricing of bitcoin is impossible for the present. But a yes from what’s driving the emotion and volatility (and yes scams). I’d lean toward a yes.

Third: bitcoin as nascent platform. Chris Dixon has been one the most articulate advocates of blockchain as platform. I’d paraphrase him as saying early in a platform lifecycle, things decentralize. Think early PC era against IBM. But later things get recentralized, for example the late Microsoft PC era. Then came the early internet decentralization. Followed by the Google/Amazon/Facebook/Apple recentralization. So what’s next for decentralizing? Dixon believes it’s blockchain. See his post here, or this recent interview.

Unfortunately a historical analogy isn’t proof. In particular one problem jumps out. Is this new platform ready to go right now? Or is it too early (the bane of all tech investors)? To see what I mean, compare this list of too early versus timed correctly: Apple newton v iPhone, Segway v electric scooters, Webvan v FreshDirect, Flooz v well…bitcoin, google glass v whenever AR glasses finally take off. You get the idea. Crypto currencies may become the next decentralizing platform, but it’s far less clear that’s happening now. For example: Out of 43 Blockchain Startups, Zero Have Delivered Products. It’s possible we’ll need another tech generation before things are ready.

To sum up. There are plenty of bitcoin scams, but don’t go overboard and say it’s a scam for VCs to make crypto investments. Bitcoin is bubbly, but be careful how you argue. As for blockchain as a decentralizing platform, timing is everything. Which means….maybe.

2. Waymo’s driverless cars not so ready to go. Announced as ready to roll out in December, the roll out happened, and….pretty quiet. From Timothy B. Lee: “It now looks to me like Waymo is nowhere close to ready for fully driverless operation in its initial <100 square mile service area, to say nothing of the rest of the Phoenix metro or other cities.” Also: “This means I have no idea how long it will take for Waymo (or anyone else) to reach full autonomy. It could take six months or it could take six years. Maybe Waymo will be forced to throw out big chunks of what they’ve built so far and start over.” Waymo (sister company to Google) is perceived as farther ahead than anyone else. So if they’re not ready, unlikely anyone else is either.

3. Ben Thompson on the state of tech at end of 2018. It’s a good post. Here’s the key paragraph:

This, then, is the state of technology in 2018: the enterprise market is thriving, and the consumer market is stagnant, dominated by the “innovations” that a few large behemoths deign to develop for consumers (probably by ripping off a smaller company). Meanwhile a backlash is brewing on both sides of the political spectrum, but with no immediately viable outlet through competition or antitrust action, the politics surrounding technology simply becomes ever more rancid.

4. Lyft continues gaining on Uber. Uber had 92% US market share in Feb 2015, 75% in Oct 2017, 69% in Oct 2018. Lyft went from 7% Feb 2015, to 25% in Oct 2017, to 28% in Oct 2018. Nice chart in the article. Both companies are going public in first quarter of 2019. More here.

5. Visualizing population density as 3D. An excellent data visualization. Population density as elevation, making big cities very spikey. If you click through, don’t miss zooming out from cities (starting with Paris) by scrolling right. Enjoyed this quite a bit. Recommended.

Source: Population Mountains

And that’s all for today. Thank you for reading. 

Monday Links 26-Nov-2018: embryo selection is a bigger deal than CRISPR, China panopticon, Saudi relations, software eats photography

Here’s links and commentary on what I read this week. Chinese CRISPR babies tops the list. So let’s get started.

1. Embryo selection is a bigger deal than CRISPR. In a surprise announcement yesterday (Nov 25), Chinese researcher He Jiankui claimed to have created the first gene edited babies. He used CRISPR-Cas9 to disable a single gene CCR5 in IVF embryos, which when disabled makes someone less susceptible to HIV. Then he implanted the modified embryos, and the mother gave birth to twins. People were skeptical, but Harvard biologist George Church said the claims were “probably accurate. I’ve been in contact with the Shenzhen team and have seen the data”.

So will gene editing lead to super babies soon? No. CRISPR is an amazing tool to edit genes, but it’s reliability is moderate. This means it should be useful for parents who carry diseases caused by very few genes, such as cystic fibrosis, Huntington’s, BRCA1, sickle cell disease. Editing a single gene or two is feasible. But traits that people might panic about, such as height, athleticism or IQ, are driven by thousands and thousands of genes, each of tiny affect. And nobody can reliably edit thousands of genes. That’s at least a decade out. Perhaps longer.

Hence this was to some extent a proof of concept publicity and technology stunt. If the point was healthy babies, the gene selected for therapy should have been one causing something like cystic fibrosis. And not disabling a correctly functioning gene, even if that gene might potentially be used by HIV as a vector. Ed Yong has a good piece up quoting ethicists and geneticists who are outraged. And it appears He Jiankui’s university in Shenzhen was unaware of his research and is conducting an investigation. With that said, it’s obvious to everyone who follows this research that China is far more open to it than the US and Europe. So in that sense it’s unsurprising China is where this is happening, and where it will continue to happen in the future.

But if CRISPR is not (yet) a big deal for super babies, what’s feasible today? Answer: Embryo selection. Which was also in the news this week. Recall that thousands of genes impact things like health, disease risk, etc. You can’t edit all those genes, but you can roll them up into a single polygenic risk score. PGS scores can assess risk of diabetes, heart disease, and yes, IQ. PGS scores are not perfect predictors, since the randomness of development still drives much of the variance. But they can give an edge. There’s nothing really new to invent. Genetic testing of IVF embryos for disease is already a thing. And if you’re choosing embryos already, why not get PGS score info to help make your choice? Your intuition here should be if a parent could magically see probabilities of say, 25 potential children, and dimly forsee their height, IQ, disease risk, and then could pick the one whose odds are best. They’d still be your kids. But their chances in life would be better than random genetics. Thus the Guardian article this week: Super-smart designer babies could be on offer soon. But is that ethical? Note this article is not about CRISPR. But about PGS scoring against a set of IVF embryos, and selecting as per the parent’s preferences. To be clear, right now the focus is on preventing disease, so picking embryo’s which score well for not having health problems like diabetes or heart disease. But the approach is amenable to selecting any polygenic trait.

Bottom line: I think embryo selection will be the single most socially disruptive technology over the next decade. AI will have to wait its turn. And yes, CRISPR is looming in the wings, possibly useful now for mendelian diseases, or maybe even traits with simple genetics such as skin or eye color. But still not yet able to create super babies. This means using IVF to select for height, athleticism, IQ is what’s most impactful now. It pushes all the right buttons. Just imagine your reaction if the politician you hate the most (or one of their adult children) takes a genetic tourism trip to China to make sure their kids have that extra special IVF selection IQ edge. People will be incredibly angry. And this anger will slice deeper into the existing political realignment going on in the US right now. On both the left and right. The babies won’t disrupt directly, but the second order effects of elites using technology to cement their children’s position at the top will make people flip out. It’s coming. And in fact is here already.

2. China continues making progress on the panopticon. This is an incremental thing. Computers will watch us all the time. It will happen everywhere. But in authoritarian countries it will be the government who watches. In US and Europe it will be regulated tech companies, which hopefully will work out better. In any case, from this story:

China’s plan to judge each of its 1.3 billion people based on their social behavior is moving a step closer to reality, with Beijing set to adopt a lifelong points program by 2021 that assigns personalized ratings for each resident.

The capital city will pool data from several departments to reward and punish some 22 million citizens based on their actions and reputations by the end of 2020, according to a plan posted on the Beijing municipal government’s website on Monday. Those with better so-called social credit will get “green channel” benefits while those who violate laws will find life more difficult.

3. Tanner Greer speaks out for ending the US-Saudi relationship. His piece is titled: The Khashoggi’s Death is the Crisis We Have Been Waiting For. Here’s one bit:

The Middle East is a distraction. America will soon be a net energy exporter. The region’s oil holds no power over us any longer.


Let us have no illusions: Riyadh has more actual American blood on their hands than any other sitting regime in the world (with the exception perhaps of the military machine run out of Rawalpindi). This was all true before the Saudis more recent and more brazen attempts to break down the rules and norms of a civilized international order—kidnapping prime ministers, killing hundreds of thousands of civilians in unnecessary wars, and so forth. If the Saudis are our ‘vital’ friends, we do not need even lukewarm enemies.

Greer sees geopolitical conflict with China as where things are headed, and the sooner the US gets out of the Middle East the better. I found his piece convincing. Worth reading.

4. The future of photography is code. Excellent piece on how capturing the real time data stream coming in from the lens allows software based image creation to exceed what you can do with just better lenses. Excerpt:

Similarly the idea of combining five, 10, or 100 images into a single HDR image seems absurd, but the truth is that in photography, more information is almost always better. If the cost of these computational acrobatics is negligible and the results measurable, why shouldn’t our devices be performing these calculations? In a few years they too will seem ordinary.

If the result is a better product, the computational power and engineering ability has been deployed with success; just as Leica or Canon might spend millions to eke fractional performance improvements out of a stable optical system like a $2,000 zoom lens, Apple and others are spending money where they can create value: not in glass, but in silicon.

Also see this related piece on how Google’s Pixel 3 is able to do amazing dim light photography based on machine learning combined with software based image synthesis. If you’re interested in photography I’d recommend both links above. If you want more, I’d suggest also reading Google’s post here as well.

5. Benedict Evans – The End of the Beginning. Slide shows and video of Evans discussing what the past 20 years of internet tech has done, and what the next 20 years will likely do. I thought it was quite good.


And that’s all for this week. Thanks for reading!