The ​2018 Year in Review

Supported by Greenhaven Road Capital, finding value off the beaten path.

In 2018 we passed 500 posts which is important if you live in a society with base ten numbers. 🎉

But it was a milestone because the podcast I admire most is Russ Roberts’s EconTalk. I don’t listen to every episode and it’s rarely my favorite of the week but Roberts has been professional, thoughtful, and has provided good arguments for a decade! Roberts’s episodes from 2007  include Sunstein, Michael Munger, Cowen, Taleb, Michael Lewis, Dan Pink, Taleb, Kevin Kelly, and Bogle. Multiply that lineup by a decade and you’ll get a partial sum of Russ Roberts’s work. That’s something to aspire to.

ICYMI these were the most popular posts for the year. Surprising to me these were longish ones. We’ve got more of that coming.

Books & Learning. In addition to deep books, we did a book list on China, and in the second month of the year a post on how to read more books. Fittingly, in the second to last month of the year, we did a post on not reading books. There was a warm response to each of these.

Podcasting. The medium matters and what makes a good podcast is still a work in progress. That said, I agree with what people enjoyed listening to. David Ogilvy’s work ‘rules‘, An Attitude of Factfullness, Skimm’d Lessons, and Making it in America were all episodes I enjoyed making and well received. The podcast is available on iTunes, Overcast, or Soundcloud.

Reflections. One. of my favorite books this year was Finn Murphy’s The Long Haul. ‘A Trucker’s Tales of Life on the Road’, is a view of America and the American Dream. One part of being a ‘bed-bugger’, a mover, that appealed to Murphy was the work. He could be proud of the clean trailer in the morning and he could be proud of his full trailer at night. Here’s one part:

“I discovered that moving suited me perfectly becuase I could lose myself inside the work. Many young male neurotics find out early that hard labor is a salve for an overactive mind…Hard work temporarily shut down the constant movie running in my brain that looped around in an endless cacophony of other people’s expectations, obligation, guilt, anger and rebellion.”

For context, Matthew Crawford, who wrote Shop Class as Soul Craft blurbed Murphy’s book. And both get at the shortfall of blogging. In one year there are 100x more readers than all my students in eight semesters of college instruction but the impact is reversed.

College is a deluge. This online thing is a drip. So maybe like compound interest or Wansink’s 500 extra calorie idea, the small steps we take add up to quite a trip.

The physical is easiest to see; I’ve gained/lost weight this year. The financial is next easiest to see; my account has grown/shrunk this year. The intellectual is most difficult; I’ve learned much/some this year. But eventually, after more than a decade of podcasts, the change is there and it’s easy to see. Thanks for reading.

This Blog’s Biases

Supported by Greenhaven Road Capital, finding value off the beaten path.

Traffic and automotive. This blog is annoyed by the automotive. Traffic and Happy City form the outline of my ideas about cars, roads, and how people use them. In general, they don’t bring out the best in us because of a poor feedback structure. However, looking at them through the lens of In-N-Out’s Cheeseburgers shows path dependent effects and lets us study a Chesterton Fence.

Mediums. This blog probably oversamples from people on podcasts. If someone doesn’t share their work we miss them. I’d wager we also don’t get everyone’s best ideas. Some also keep their head down so not to attract competition, avoiding alpha erosion.

Good enough. This blog probably overestimates the importance of ‘good enough.’ In marketing, it goes under satisficing, where we don’t maximize like economists but satisfice like humans. In Christensen’s disruption theory consumers switch from one product to another because the previously important area is now good enough and something else is paramount. In money, it lives in the FIRE movement, where people decide a lifestyle is good enough.

Conditions matter. This blog probably overestimates how much conditions matter. Network researcher Nicholas Christakis explained one experiment: “The gist of the experiment is that I can take you people and connect you according to one set of rules and you’re mean sons-of-bitches to each other. Or I can take you and connect you by a different set of rules and you’re sweet and kind to each other. It’s the same people, but different architecture yields different emergent properties of the system.” However…

Research and the real world. This blog probably overemphasizes the written and underemphasizes the done. Ironically, being there is advice that comes up again and again. Robert Cialdini gave one example when he said “I went into it (his first field project) to get some ideas for doing research in my laboratories. Say something this way versus that way…I realized that in a laboratory with college students I was missing the power of these techniques to really make a difference outside of the laboratory in the real world.” Facing the winds of the real world is also where you find thick data.

Kindness. This blog probably gives too much space to nice people – maybe. Here I’m selective, rationalizing more like Gerd Gigerenzer, that kindness is ‘rational’ because there’s more than the idea in a message.

Arguments. This blog probably overestimates the value of good arguments. It’s been a long time since I was part of a team working on a project so maybe people are doing this or maybe it’s a marginal benefit or maybe people don’t have the resources to disagree. But maybe not. James Mattis said:

“Take the mavericks in your service, the ones that wear rumpled uniforms and look like a bag of mud but whose ideas are so offsetting that they actually upset the people in the bureaucracy. One of your primary jobs is to take the risk and protect these people, because if they are not nurtured in your service, the enemy will bring their contrary ideas to you.”

Thanks for reading.

2018 Books

Supported by Greenhaven Road Capital, finding value off the beaten path

The (hindsight) theme for this year was continuing education. In years past I might read one book and bounce to another like Tigger the tiger might move through the Hundred Acre Wood. This year I was more like Pooh, childlike and curious. 

Current economies. Though not a book, Brian Arthur’s paper was insightful and Modern Monopolies was a framework with examples. The Gorilla Game though dated is still helpful. Breaking Smart is page-for-digital-page the best book I read this year. Videocracy is the choice for wrapping your head around YouTube.

Single stories. A Truck Full of Money is the story of Kayak, but even better, it’s written by Tracy Kidder. Made in America is the story of Walmart. Springfield Confidential is the (partial) story of The Simpsons. Reacher Said Nothing is the (academic) story of Jack Reacher. At Work is a collection of Annie Leibovitz pictures and stories. To Pixar and Beyond is the (CFA) story of Pixar. Creativity Inc. is the (CEO) story of Pixar. The Long Haul is the story of America from a trucker’s perspective. Buffett by Lowenstein was better than expected, and I probably should have read sooner.

Marketing. Take your pick of a David Ogilvy book, my favorite is Ogilvy on Advertising. I also recommend Rory Sutherland on YouTube and Terry O’Reilly on the podcast Under the Influence. Win Bigly is ‘persuasion in a world where facts don’t matter.’ Hit Makers for the subjectivity of popularity. Hidden in Plain Sight by Jan Chipchase for finding thick data because “The best way to understand how a culture adopts (or doesn’t adopt) an innovation is to go there and see it for yourself.”

History. The Worldly Philosophers filled in a gap I’d had about Malthus, Mill, Veblen, and Keynes and their economic ideas. Ditto for Churchill and Orwell, a snappy set of stories about each. For only Churchill check out Hero of the Empire, about the Boer War. Chasing the Scream is a how we got to now story of addiction.

Misc. Mastermind by Maria Konnikova is what you’d expect if you enjoy her New Yorker articles and Sherlock Holmes (I do!). Best State Ever if you want to laugh at, and with, Floridians. Thinking in Bets for decision making from a poker perspetive.

Fiction. City of Thieves is written by David Benioff of Game of Thrones, advocated by Brian Koppelman and read by Ron Pearlman – I loved it. The Midnight Line – or any Jack Reacher book. Norse Mythology, Gaiman and the old stories.

The movie was better than the book. Factfullness was championed by Bill Gates but Hans Rosling’s enthusiasm sparkles in video. Gridiron Genius was good but Lombardi’s podcast is better. The Good Neighbor is the story of Mr. Rogers, and though I haven’t seen the movie I probably have in most of the alternative histories.

Thanks for reading.

Image result for winnie the pooh quotes

Andrew Ng

Supported by Greenhaven Road Capital, finding value off the beaten path.

Andrew Ng has quite the resume to talk about AI, building the AI team at Baidu, cofounding Coursera, and leading Google Brain. His November 2017 talk on the Artificial Intelligence Channel is an overview.

We’re talking about AI now because we have enough data, passable algorithms, and people programming the two.

AI is a tool, like email and spreadsheets and it’s changing how we work. Ng gave this rule of thumb, “anything a typical person can do with less than a second of thinking we can probably now assume automate.” Ajay Agrawal said much the same, “AI doesn’t do workflow, it does tasks.” At Goldman Sachs for example, they figured out that an IPO has 146 distinct steps.

Ng gave the example of a security guard; notice, identify, categorize, respond, repeat. AI can do some of those things. AI will replace security guards, maybe. Jobs have always been about outcomes. People don’t want quarter-inch drill bits, they want quarter-inch holes. Job security will be about using new technology to create those holes.

And people adopt to adapt all the time. Blackboards became smartboards, books became digital, and grade books went online but teachers still teach.

To date, machines have vanquished only one occupation; elevator operator. Hal Varian said, “Automation generally eliminates dull, tedious, and receptive tasks.” AI, like other technologies, will make people better at their jobs. Accountants used to count, Computers used to compute.


Pedro Domingos told Shane Parrish, “People sometimes think the easiest jobs to automate are the blue collar ones but our experience is the opposite. It’s often white collar jobs that are easier to automate.” Computers don’t miss the gorilla inserted into lung scans.

Robots are not coming for our jobs but they will change them. Author of the book Humans+Machines, Paul Daugherty said, “The skills we see increasing in importance in the human plus machine age are creativity, reasoning, and socio-emotional intelligence.”

In Average is Over, Tyler Cowen proposes that income divergence will continue based on this qualifying question; can you work well with machines? The have and have-not trends “stem from some fairly basic and hard-to-reverse forces: the increasing productivity of intelligent machines, economic globalization, and the split of modern economies into both very stagnant sectors and some very dynamic sectors.”

If this were a movie about the changing economy AI could be a villain from central casting.

Ng compares AI to electricity, noting it will be ubiquitous thanks to more data and better models. Data accumulation is like the advice on planting a tree; the best time was thirty years ago, the next best time is today. That’s what Google did.

GOOG 411 began in 2007 as a directory information service. Google Economist Hal Varian said the call-in service, “learned how to recognize voices, learned how to recognize accents, learned all these different things in this very limited domain of directory information. That was enough to get started and now I think we have one of the best voice recognition systems in the world.”

We’ve been saying ‘Hey Google’ for over a decade.

Today, 2018, algorithms are less important than the data and people are more important than both. Varian said, “These days the scarcest factor by far is expertise.” Rory Sutherland worried that there are more data sets than mathematicians competent enough to handle them. But this prioritization may be shifting.

In a talk about his book AI Superpowers, Kai-Fu Lee suggests a shift of AI preeminence from the United States to China. The U.S. has the leading researchers but China has the best data, thanks in part to the number of people and the environment they live in. Mobile payments are one example, Lee explained, “People’s spending patterns are so much more valuable than their clicking patterns.”

Ng noted that more than 10% of Google and Baidu searches come through voice “because voice recognition is finally accurate enough.” Product adoption or abandonment is largely drive by if something is ‘good enough’, an idea at the heart of Disruption Theory.

AI like other tools, can improve the way people work but will never be perfect, just like other tools.

Data is the new electricity. Err, I mean oil, data is the new oil! Oops! Should have said gold, yes, data is the new gold! Wait, that’s not it either? Analogies can help but data is different.

Hal Varian said, “Some people say ‘data is the new oil’. My response is, they have one thing in common. To be useful they have to be refined. A barrel of oil isn’t worth much but turn it into gasoline, kerosene, or hydrocarbons and it’s worth something. It’s the same thing with data…(but data is non-rival)…It’s a mistake to talk about data ownership because it’s too narrow a concept.”

This is a divergence in AI opinion. People like Varian think data ownership is the wrong perspective. People like Ng believe data ownership is the only way to earn a competitive advantage, “algorithms from a company point of view are for the most part not defensible.”

Ng teaches his Stanford students about a virtual loop where data leads to a product that leads to users who generate data. 🔁 And, “after a period of time, you might have enough data to yourself have a have a defensible business.”

For example, Blue River Technology uses “cameras, computers, and artificial intelligence to allow Ag machines to see every plant in a field.”

Ng concluded his presentation with one lesson from his internet days; selling things online does not make you an internet company. “What defines the internet company is whether or not you have an architect at your organization to leverage internet capabilities to do the things that the internet allows you do really well.”

Things like a/b testing, short product cycles, and bottom-up actions. Internet companies had to “push decision making down from the CEO to the engineers and product managers because the internet products and uses are so complicated that a lot of knowledge about what needs to be done lives only in the heads of the engineers and product managers.”

In other words, it has to be a decentralized command. Pedro Domingos said, “Machine learning is computers programming themselves instead of having to be programmed by us…In general terms, tell them what you want them to do and let them figure out by themselves how to do it.”

The challenge then is to figure out what it means to be an AI company.

But there’s more to jobs than work. I won’t muddy up what those things are but The Long Haul and Shop Class as Soulcraft provide clear examples.

h/t Patrick O’Shaughnessy on Twitter.

Paul Daugherty

Supported by Greenhaven Road Capital, finding value off the beaten path.

Paul Daugherty has spent thirty-plus years at Accenture and with co-author, H. James Wilson wrote Human + Machine: Reimagining Work in the Age of AI. They emphasize similar ideas that Haskel and Westlake point out in Capitalism without Capital and Hal Varian addresses in his talk Bots vs Tots. Each author wants us to rethink how work is done and how work might be done with the rise of machines.


Actually, that’s bad phrasing.

The conquering of machines!

Wrong direction.

Seth Godin lamented that we missed a great opportunity to brand global warming. ‘Global’ and ‘Warming’ are two mostly positive words. As Scott Adams suggests, wording matters. Godin thought global warming should be called ‘atmosphere cancer.’ It’s why Paul Daugherty has “started using the term ‘collaborative intelligence’ rather than ‘artificial intelligence’. The problem with AI is it scares the public and it leads to the wrong discussions.”

We’re having this discussion now even though ‘artificial intelligence’ has been around for more than sixty years because of three changes. There have been advances in computing, there is data from new sources and at scale, and algorithms have advanced, including progress with voice and speech.

Daugherty starts one talk by saying, “If you forget the rest of my talk here’s the real thing to take away; the plus sign on the cover of the book. If you remember nothing else about what I say, just remember the plus sign because it’s the fundamental basis for the research we’ve done.”

AI is already at work. Wal-Mart has used VR to simulate Black Friday Mobs. At Accenture, someone built a system to track resumes, experiences, and current jobs openings to suggest when and where someone should considering retraining and transferring before their job becomes obsolete. There are amazing possibilities but, “the problem a lot of business executives have is ‘What do you do with it all?'”

Daugherty showed this Dilbert Cartoon where The Boss asks, “But why can’t we 3D print a blockchain and HTML it into a Bitcoin.” Rather than nonsense, Daugherty suggests baby steps.

We need to change our thinking. Robots are coming for us but as assistants, not adversaries. Machines will not take our jobs. Work will remain. With a little work in the right direction will change for the better.

“AI gives people superpowers.” Both Mercedes and Tesla tried mostly automated factories and both blundered.

AI won’t be confined to the factory. It’ll be everywhere like electricity or wifi. It’ll be in brands. AI Personality Trainers are people who tune pop up chatbots in apps. “AI is your brand.” AI is also used to flag money laundering and in some cases moves from 70% false positives to 30%. Technology gives people superpowers.

“The skills we see increasing in importance in the human plus machine age are creativity, reasoning, and socio-emotional intelligence.”

The future is difficult to predict except to say that it’s going to be different. Tim Hartford talked about how we overlook the simple things and focus on the glamorous. It wasn’t just Gutenberg’s press in 1450 but paper too! Ditto for Bladerunner where the replicants look just like humans but the phones look just the phones.


But not even baby steps happen without organizational support, support that starts at the top. “One thing we see is a senior leadership driven mindset around applying AI differently to the organization. It’s not just AI at the edges and at the small things but taking on something that matters to the organization and changing some of the behavior and culture in the organization.”

Daugherty adds that it should be a C-suite job or direct report. “You want somebody in your business understanding the context and application of AI and be accountable for it.”

In surveys of 1500 companies for the book, Daugherty found that only 1 in 10 companies were applying AI in reimagined ways. He and Wilson also found that “two-thirds of executives believe their workforce isn’t ready for AI.” But don’t go out and try to hire a machine learning specialist. “That’s what people often say, ‘I need more machine learning experts,’ or whatever the technology might be. That’s important but the bigger issue is the talent that uses AI. How do you change the culture and train the people that need to use AI for these different types of jobs?”

AI experts won’t be silver bullets. It’s your culture that matters most. Adaptability is what precedes survival. “One of the greatest examples of this is Amazon. If you look at their Go stores, they started small, tested on employees first, improvised and innovated with the technology, and now they’re scaling it. That’s a great example of behavior we see consistently with leading organizations.”

After listening to Daugherty I felt optimistic because he’s optimistic. “I strongly believe, that the more human-like technology is, the more it enhances our ability to be human.”

Hal Varian

Supported by Greenhaven Road Capital, finding value off the beaten path.

Hal Varian is the author of two economics textbooks and co-author of Information Rules: A Strategic Guide to the Network Economy. Varian spoke at the ESMT business school in Berlin in 2018.

Varian’s talk is titled Artificial Intelligence Economics and Industrial Organization. AI EIO, “kind of like old MacDonald had a farm.”

Varian represents the techno-economy as a pyramid with three layers. A base layer of data, a next layer up of information and the next layer of up knowledge. It’s only at the knowledge peak where we get action.

“There’s a lot of debate about the data economy, the information economy, the knowledge economy, but they’re all part of the same process, and no part can stand on its own.”

A series of recent breakthroughs is part of the reason the AI EIO conversation has entered the public sphere. “Many techniques have emerged that just a few years ago we thought were impossible.”

Data is our sand at the beach, models our buckets, and then it’s up to us. “These days the scarcest factor by far,” said Varian, “is expertise.” Rory Sutherland agrees, “One of my great worries about big data is that there are far more big data sets than there are mathematicians competent enough to handle them.”

Think of data as a raw material, like sand or oil – with one big difference. “Some people say ‘data is the new oil’. My response is, they have one thing in common. To be useful they have to be refined. A barrel of oil isn’t worth much but turn it into gasoline, kerosene, or hydrocarbons and it’s worth something. It’s the same thing with data.”

But data is non-rival. Two organizations can, and maybe should, have the same data. “It’s a mistake to talk about data ownership because it’s too narrow a concept.” Instead, maybe data is something to license.

Good organizations have people that argue well. Good economics have organizations that compete. Cordial clashes are fractal. “Usually pluralism is important because it allows a diversity of viewpoints and in many cases allows for a significant improvement in terms of unitizing that data.”

Varian emphasized the abundance of services for data. He also said that Google Home started as Goog-411. The call in service, “learned how to recognize voices, learned how to recognize accents, learned all these different things in this very limited domain of directory information. That was enough to get started and now I think we have one of the best voice recognition systems in the world.”

Maybe a better analogy is that data is like a river, we are the settlers, and models are the mill. It took a lot of effort to discover the data/river, and we had to be willing to venture out and harness it. Now we need to figure out how to use it well. “The challenge is finding the expertise to extract the value from the data.”

The goal isn’t perfection, just improvements. Robots won’t take our jobs but they can make our jobs better. “We think it has to be people plus machines to really be effective.” How well someone works with machines is a key part of Tyler Cowen’s book Average is Over.

The good news is that everyone is kind of figuring things out together. Cowen wrote that computers play chess moves no human understands. Varian said that computers find indicators of health (and sex) in the eye that no human understood – at least for a couple of weeks. The doctors did figure it out, and that’s kind of what we’re all doing on the banks of this river.

Varian has other good talks on YouTube, if you want more check out Bots and Tots. 

Avi Goldfarb & Ajay Agrawal

Supported by Greenhaven Road Capital, finding value off the beaten path.

Avi Goldfarb and Ajay Agrawal spoke at Google about Prediction Machines. They wrote the book because they noticed more AI startups in Toronto and more AI investing in the Bay. “We got to see both the flood of companies coming into our lab in Toronto and the excitement that was starting to percolate in the Bay Area.” They had a ‘that’s interesting‘ moment.

Like other AI presentations, Goldfarb and Agrawal articulate the nuance. “The reason we’re talking about AI in 2018 and not in 2008 is not because of C3PO or Terminator technology, it’s because of advances in machine learning.” What is machine learning? Better, faster, and cheaper predictions.

Donning economists lenses – like Hal Varian did 15 years prior – reveals this framework:

1/ When things get cheaper we use more of it.
2/ When things get cheaper we use less of substitutes.
3/ When things get cheaper we use more of compliments.

If the price of coffee drops (1) the value of tea also drops (2) but the value of sugar and cream increases (3). Now, what if that happens to something like arithmetic?

“Your computer does one thing, arithmetic. But once arithmetic is cheap enough we find all sorts of opportunities for arithmetic that we might not have thought of before.”


“Photography used to be a chemistry problem but as arithmetic became cheap we transitioned to an arithmetic based solution.”


Like accountants, who used to specialize in addition but now focus more on inquisition. “There are still accountants because it turned out the people who are best positioned to do all the arithmetic were also the best positioned to understand what to do when a machine did the arithmetic for them.”

Humans are great at finding new ways to use new tools. Cheap coffee has knock-on effects. Cheap arithmetic has knock-on effects. Cheap predictions have knock-on effects. And we are making predictions all the time about loans, tumors, behaviors, anymore.

Goldfarb wants people to ask, “What are the core compliments to prediction? What are the cream and sugar that become more valuable when coffee is cheap.” He thinks it’s decision making. “Prediction is not decision making, it is a component of decision making.” Other components are data, actions, and judgment. These are the tasks humans are better at than machines whereas machines are better at the arithmetic.

Around 17:30 Agrawal takes the lectern and provides a helpful framework. Like Paul Daugherty, Hal Varian, and Nicholas Christakis pointed out, people are excited but perplexed about AI and Machine Learning. It’s Dilbert’s boss suggesting, “we 3d print a blockchain and html it into a Bitcoin.” Daugherty said, “The problem a lot of business executives have is ‘What do you do with it all?'”

Agrawal points us in the right direction. For starters, think small. “AI doesn’t do workflow, it does tasks.” At Goldman Sachs, they figured out that an IPO has 146 distinct steps. “When organizations show up and say where do I even start, this (The AI canvas) is a coarse description of how we start.”

This canvas is simple. “The key point here is that there are senior-level people who have never written a line of code but can sit down and start filling these things out.”

The big effects will come, Agrawal said, when the returns from AI create cascades. “A common vernacular for this type of phenomenon is disruption.” For example, right now Amazon’s business model is shopping then shipping but what if their AI becomes good enough where their model is shipping then shopping. (26:10)

This may not need to happen at perfect prediction, just some good enough level. “There’s some number, it doesn’t have to be a Spinal Tap level of prediction accuracy, but there is some number where when they get to that level of prediction accuracy that someone at Amazon says ‘We’re good enough at predicting what people want, why are we waiting for them to order it, let’s just ship it.'”

“Amazon becomes transformational when the recommendations get so good that they no longer have to have the same business model as the Sears catalog.”

But betting on predictions will only work – like any other strategy – in the right culture. “The allocation of scarce resources is what makes something a strategy.”

Like marketing and design, AI researchers face the trouble with selling this to the boss. Researcher Grant McCracken said, “When a senior manager says, ‘Fine, that’s what you think, where are the numbers?’ And the best we can do is say, ‘Just trust us.’ It’s like, yeah right, ‘I’m not trusting my career, my children’s opportunity to go to college on your impression. Where’s the data?’ By data, they don’t mean, I did an ethnography in someone’s kitchen. They mean, please could I have some numbers.”

Numbers let people appear rational but we don’t have a (good) rational answer for what the killer app for AI is. Agrawal said, “I think our barrier is imagining all the things that we’re going to do.” To reimagine is one of Paul Daugherty’s key points too.

“The person who asks good questions of their data has a higher return to that part of their skillset.” Seth Godin said there are two things we should teach in school, leadership and how to solve interesting questions. With some imaginative intelligence, AI can help with the latter.

Thanks for reading.