Blog AF

• Voice and the uncanny valley of AI
  Benedict Evans
  

  Like I said last time Evans wrote about this, “voice interfaces seem to be adding more friction than they take away.” Changing people’s habits is hard. I constantly prompt my Echo by saying “Hey Siri,” and find myself thinking of ways to phrase my questions to make them intelligible by the machines. Once or twice a week I get developer emails about “What’s New With Alexa,” but after many months my Echo only acts as a gateway to Spotify, and a party trick whenever there are guests. Voice might be the new platform, but it is nowhere near.

• What Do Economists Actually Know?
  Russ Roberts - NewCo Shift
  

  The issue with modeling of any kind is that there are no alternative realities to compare against. We can only measure what we see, and by definition there are no counterfactuals, nor what-ifs. Making policy decisions under this state of affairs is hard, and the only thing we can do about it is internalize this limitation, knowing that we could be very wrong. Statistical analysis is a tool, and like any other tool, it is succeptible to operator error.

• If There Are an Infinite Number of Parallel Universes, Some Must Be Terrible Places
  Dean Zimmerman - Nautilus
  

  I disagree with the author’s position, but thinking of when the problem of evil meets 21st century math and science is fascinating.

• Patagonia and The North Face: saving the world – one puffer jacket at a time
  Marisa Meltzer - The Guardian
  

  Good reads on the history of large well-known retailers are not usual, as most stories in the genre end up with a strong PR flavor. This article, by virtue of showing the origin stories of two seemingly rival companies at once, achieves a good balance.

• Improving U.S. Healthcare and Coverage
  Stephen Cecchetti and Kermit Schoenholtz - Money, Banking and Financial Markets
  

  Americans are exceptional in their very own ways. This whole healthcare story is a fiasco, and I am amazed that the American people have allowed it to go this long.

• White supremacism is not nationalism
  Noah Smith - Noahpinion
  

  I am reading Benedict Anderson’s Imagined Communities and Noah’s arguments in this blog post fit right into the framework that Anderson proposes at the beginning of his book. Nations are made up. Expect a blog post about this soon.

• Triple Pendulum CHAOS!
  Jake VanDerPlas - Pythonic Perambulations
  

  When I look at demos like these, I wish I had kept learning about differential equations after sophomore year of college. Complexity is awesome, and being able to model these crazy patterns with only a few lines of code would be great.

• Update: CRISPR
  Radiolab
  

  The cutting edge of biological research gets more interesting, and more scary, the more I learn about it. Researchers are uncovering really powerful building blocks, but we have very little understanding of the complex relationships in the whole system. The ethical considerations discussed in the last part of this episode are especially worth listening to.

• Hacking The iPhone For Fun, Profit, And Maybe Espionage
  Planet Money
  

  Something I will never understand is how someone can enjoy poring over low level buffer management for hours to find an overflow condition or some obscure vulnerability. Luckily some people like watching water boil with their white hats on, and do it for the greater good, too.

Well, here’s a random set of stuff. Costa Rica, San Francisco, and México. More...

• Why All Exchange Rates Are Bad
  Timothy Taylor - Conversable Economist
  

  The Trilemma makes it so that whatever policy a government decides to follow, it must be an active choice. Currency manipulation towards a stronger, weaker, more stable, or more volatile currency is a choice, and there is no default. Like everything else in economics, and the world we live in, it is a choice about tradeoffs, and understanding who gains and who loses (and by how much!) is the key to the issue.

• A Big Little Idea Called Legibility
  Venkatesh Rao - Ribbonfarm
  

  I am currently reading several books at once, and one of them is Seeing Like a State. So far, the book has presented a bunch of ideas about the inner workings of what the author calls high modern states. In this post, Rao (who is not the author of the book) summarizes one of the most interesting ideas in the book so far: legibility, or reorganizing society to make it more understandable, and thus more governable.

• President Trump Wants a Wall? Mexico Is It
  Eduardo Porter - The New York Times
  

  The average American has no clue of how immigration policy and actual immigration patterns work. Understanding how much effort other countries put into helping the US keep illegal immigrants at bay could be helpful in the current climate.

• The Uber Conflation
  Ben Thompson - Stratechery
  

  2017 gave Uber a rough start. In an unusual post, Ben argues for a change of leadership, focusing on two questions: ‘Is Uber’s approach to regulation wrong?’ and ‘Is Uber wrong with regards to the specific issue at the center of this controversy?’

• Brains, Bodies, Minds... and Techno-Religions
  a16z Podcast
  

  One of my favorite episodes of a16z ever. Touching on the subjects of nationalisim, imagined communities, religion, governments, etc, etc, etc, and how all of these are affected by the rise of technology. I had shared a related piece from Harari a few months ago, but this podcast episode is way better.

• Crafts, Garicano, and Zingales on the Economic Future of Europe
  EconTalk
  

  Having Russ Roberts and one of his guests debate on economic topics is fun. Having him and another three guests? Even better.

This post has been sitting half-baked on my draft list for far too long. In an effort to motivate myself, and to write more, I decided to post it as is and move on. My apologies to the authors of the uncommented links.

• Manifestos and Monopolies
  Ben Thompson - Stratechery
  

  It has been interesting to see Ben apply aggregation theory to politics more and more. I agree with the views presented in this article about centralization (or lack thereof), regulation (or lack thereof), and market solutions (or lack thereof).

• Surfing, metrics and creation: Facebook and Snap
  Benedict Evans
  

  Since Snap’s S1 came out a couple of weeks ago, everyone has been discussing whether moats exists or not. The fact that their whole thesis revolves around the disintegration of sustained competitive advantages is fascinating. Evans’ index fund analogy adds an interesting idea to the mix: Facebook, Instagram, and Google must reflect reality and serve billions, while Snapchat will aim to create N things, each worthwhile to M million people, such that N*M becomes significant while not overtaking the role of the index.

• Segregation Had to Be Invented
  Alana Semuels - The Atlantic
  

  Not surprisingly, the past is different than we think it was. Thinking of the rise of segregation as a relatively new phenomenon is odd.

• Hyphen-Nation
  Bayeté Ross Smith - The New York Times
  

  A great set of interviews. I constantly think about this topic of where identities overlap and how people view themselves vs. how they are sseen by others. More so these days. Belonging, otherness, and these social dynamics are very intriguing.

• How did Europe become the richest part of the world?
  Joel Mokyr - Aeon
  
• "Incentives" as bigotry
  Chris Dillow - Stumbling and Mumbling
  
• People Actually Use Food Stamps to Buy More Food
  Noah Smith - Bloomberg View
  
• Clock
  50 Things That Made the Modern Economy
  

Unintentionally, there’s a lot of money/banking/finance content in this post. At least its not all about Trump, right?

• The Future of the Euro
  Stephen Cecchetti and Kermit Schoenholtz - Money, Banking and Financial Markets
  

  “Markets price risk, not hope.” In times like these, I wish I understood international finance better.

• Money, blockchains, and social scalability
  Nick Szabo - Unenumerated
  

  I have previously discussed Szabo, and his view on human institutions as “trust-offloading mechanisms.” In a way, money is the ultimate trust-offloading abstraction. Until the last few years, money – American Dollars, the Euro, or the Costa Rican Colon – still relied on trusting several points of failure – states, the payments networks, the certificate authorities – and our human interactions simply assumed those costs. Bitcoin and the internet have started to changed that, and further developments in technology promise much more. This is a post I’ll probably re-read again soon.

• Money as a Commons: Basic Income, Demurrage and Crypto Currencies
  Albert Wenger - Continuations
  

  Something that surpised me about monetary policy rules when I studied them in college was how much their usefulness depended on people’s expectations of their usefulness. Individual agents’ beliefs on the predictability and stability of a state’s decisions about its monetary policies were the fulcrum of all the models we studied in these intermediate economics classes. Thinking about how these rules can be baked into a crypto currency, such as bitcoin’s pre-defined velocity rule, or freicoin’s holding fees, is really interesting. Like Albert, I’m excited about these experiments.

• Snap's Apple Strategy
  Ben Thompson - Stratechery
  

  I have seen the Snap <=> Apple narrative popping up over and over across the web. Except for the “obsessive-design-focused-mission-driven-CEO” story, which I can’t really ascertain (and neither can the media!) , I haven’t really seen any good arguments to back it up. Ben comes the closest, but doesn’t quite convince me either. The Facebook <=> Microsoft analogy is much more believable, given the market conditions. I am bullish on Snapchat, and I am enjoying my spectacles (a different blog post soon?) but we’ll have to wait and see how this one plays out.

• What We Don’t Do
  Fred Wilson - AVC
  

  Thinking about this on a personal level is interesting. I know what my goals for the year are, to a certain level, but what are things that I should specifically not focus on? After all, focus means “saying no to the hundred other good ideas that there are.”

• Understanding Python Class Instantiation
  Amir Rachum
  

  These one-off explanations of snippets of source code are always enlightening. Coding seems like magic, until you discover that there is not magic, just layers upon layers of well thought out abstractions, each one understandable on its own, but magical as a whole.

• Square-Mile Street Network Visualization
  Geoff Boeing
  

  Urban development is super interesting, and thinking of how cities work by looking at their layouts is a good exercise. I wish I had worked more with geographic data while I was in the Apple Maps team. This seems like a neat library, and the fact that it is based on networkx makes me even more curious. There might be a side project brewing here.

• Anthony Bourdain's Moveable Feast
  Patrick Radden Keefe - The New Yorker
  

  My friend Dana had already recommended that I read Kitchen Confidential, but after reading this, it got bumped up a few notches on my to-do list. Parts Unknown, Bourdain’s show, is one of my go-to “half an hour to kill” shows on Netflix. It is very entertaining, and it makes an effort to show more than just the food of foreign lands. As the article describes, there is a layer of political complexity to the show that is unusual for its kind. Make half an hour to read this, and then make another one to watch an episode of the show.

• Bitcoin, the Blockchain, and Freedom in Latin America
  Russ Roberts and Jim Epstein - EconTalk
  

  The confluence of artificially cheap electricity, price controls on imported goods, and hyper-inflation are making Bitcoin more and more mainstream in places like Venezuela. While the great majority of the population probably has no idea of what cryptocurrencies are, or how to use them, it is really interesting to hear about how the fringe slowly drifts.

• Billy Bookcase
  50 Things That Made the Modern Economy
  

  There is a comment I especially liked here about how “…innovation in the modern economy isn’t just about snazzy new technologies, but boringly efficient systems. The Billy bookcase is not innovative in the way the iPhone is innovative. The innovations are about working within the limits of production, and logistics, finding tiny ways to shave more off the cost…” The iPhone example is coincidental, I’m sure, but I immediately went to “this is the argument against Tim Cook.” It is hard to appreciate how much of today’s Apple is dependent on the advanced supply chain operation that the company has built. Pushing atoms is also innovation.