If you build a robot to fish

Scott Santens has been thinking a lot about fish lately. Specifically, he’s been reflecting on the aphorism, “If you give a man a fish, he eats for a day. If you teach a man to fish, he eats for life.” What Santens wants to know is this: “If you build a robot to fish, do all men starve, or do all men eat?”

The self-made man, a self-made myth?

A fascinating long read: The Self-Made Man: The Story of America’s Most Pliable, Pernicious, Irrepressible Myth.

The essay contains eight profiles of purportedly self-made Americans, from Benjamin Franklin to the author’s own father. In summary: Success comes from hard work, and help from others, and luck.

I’m fascinated by modern versions of the American Dream, such as Shark Tank pitches, four-hour work weeks, and multi-level marketing. And by portrayals of fictional self-made anti-heroes, from AMC’s golden age of television: Mad Men, Breaking Bad, Halt & Catch Fire. If you’re curious about these topics too, then I recommend this article.

Pope: Social media can be both help & hindrance

May 17, 2015 is the 49th World Communications Day. In anticipation, Pope Francis wrote a statement. An excerpt below:

Today the modern media, which are an essential part of life for young people in particular, can be both a help and a hindrance to communication in and between families.

The media can be a hindrance if they become a way to avoid listening to others, to evade physical contact, to fill up every moment of silence and rest, so that we forget that “silence is an integral element of communication; in its absence, words rich in content cannot exist.”

The media can help communication when they enable people to share their stories, to stay in contact with distant friends, to thank others or to seek their forgiveness, and to open the door to new encounters. By growing daily in our awareness of the vital importance of encountering others, these “new possibilities”, we will employ technology wisely, rather than letting ourselves be dominated by it.

— Pope Francis, Message for WCD 2015

So this is our call to action, for people of all faiths, or none: employ technology wisely, to share our stories, stay in contact, thank others, seek forgiveness, and open the door to new encounters.

Pope Francis chooses a perspective of optimism and opportunity:

The internet, in particular, offers immense possibilities for encounter and solidarity.  This is something truly good, a gift from God.

This is not to say that certain problems do not exist. … While these drawbacks are real, they do not justify rejecting social media; rather, they remind us that communication is ultimately a human rather than technological achievement.  …

Effective Christian witness is not about bombarding people with religious messages, but about our willingness to be available to others “by patiently and respectfully engaging their questions and their doubts as they advance in their search for the truth and the meaning of human existence.”

— Pope Francis, Message for WCD 2014

Which reminds me of:

I often tell students that the history of new media has been shaped again and again by four key innovative groups — evangelists, pornographers, advertisers, and politicians, each of whom is constantly looking for new ways to interface with their public.

— Henry Jenkins, MIT’s Director of Comparative Media Studies Program, “Why media studies should pay more attention to Christian media”


Technology is neither good nor bad; nor is it neutral.

— Melvin Kranzberg, Georgia Tech professor of the history of technology, Kranzberg’s First Law

Banjo in 5/4 time

An intense song, from Sufjan Stevens’ minister friend Vito Aiuto. Sacred Harp meets Jesus Christ Superstar. With orchestra, choir, electric guitar solos. And banjo. Probably tied for my favorite banjo song in 5/4 time:

My other favorite banjo song in 5/4 time. By John Hartford, with Dave Holland on bass:

If predictions are too precise, is insurance not insurance?

Excerpts from two blog posts with philosophical questions about insurance. In a quantified, connected world, more data enables better predictions. But, there’s tension between personalized rates vs. pooled risks. Fundamentally, does the concept of insurance require a certain degree of uncertainty?

Black boxes for people are not yet standard, but wearable sensors can monitor movement, heart rate, and heart rhythm, blood pressure and blood-oxygen levels, and glucose levels and other health-related statistics. Such information can be recorded and reported through smartphone apps, watches, and other wearable devices. …

The actual problems with health insurance markets have less to do with information asymmetry and adverse selection than with too much information. That can make some people’s insurance very expensive at actuarially sound rates. For instance if you have a cancer of a given kind, this is verifiable to the outside world, and if the treatment costs are $200,000, the cost of an insurance policy will in turn be about $200,000. Buying the policy won’t be cheaper than buying the treatments, and in that sense the market for insurance is not always present. That is a very real public policy problem

The cheap sequencing of the genome may accelerate and intensify these issues. …

— Alex Tabarrok & Tyler Cowen, The End of Asymmetric Information

I think a quantified, connected society will have some interesting consequences for businesses based on managing uncertainty. As we get better at prediction, the economics of amortization get worse. 

There are entire industries predicated on amortizing risk across populations. Insurance is a good one. …

Prediction and amortization are fundamental opposites. If you know nothing about a population’s risk factors, you use a socialized medicine model where everyone is taxed equally. Risk is shared. If you know more, you have differentiated pricing based on pre-existing conditions, smoking, etc.

In any amortized model, even with some prediction and correlation, there is an element of uncertainty. Among the population of, say, fast drivers, or smokers, you don’t know which will have an accident or die of throat cancer.

So the “perfect” prediction would be an insurance policy tailored to one person. If I knew with absolute certainty the chance that you’d have an accident and the economic impact of that accident, then your monthly insurance payment would simply be monthly deposits into a savings account whose value would equal the cost of the accident on the day of the accident. Plus, of course, the insurer’s administration and profit margins.

In other words, a tax on your life.

Obviously, no prediction system is perfect. But as we use data to make more and more accurate decisions based on the latest information, things get absurd. In the split-second before a collision, when someone slams on the brakes, we have a very good idea about their chances of being in an accident. Can we revoke their coverage or up their premiums? …

— Alistair Croll, The Selfish Economics of Big Data

Marc Andreessen’s tough love career advice

From a 2012 interview. According to Andreessen:

  • The spread of computers and the Internet will put jobs in two categories. People who tell computers what to do, and people who are told by computers what to do.
  • There’s no such thing as median income; there’s a curve, and it really matters what side of the curve you’re on. There’s no such thing as the middle class. It’s absolutely vanishing.
  • We’re in a bubble for people with a non-Ivy League, non-technical education. If you have a degree in English from a tier B state school, you’re not prepared.
  • Big companies are not going to take care of you. You can’t let history happen to you.

How Computers Learn

Peter Norvig on big data:

It turns out that a lot of problems, if you feed in a little bit of data to a very simple algorithm, it performs terribly, but if you get out into the millions and billions of examples, it starts to perform well. Having a tricky new idea isn’t as good as going out and gathering more data.

This is from Norvig’s recent talk, “How Computers Learn”:

Norvig is an excellent explainer, using many analogies, examples, and simple visuals. He starts with a real world scenario, comparing two different ways to write computer programs:

  • By programmers, with logic
  • By computers, with probabilities

Let’s go back to the real world, and think of a typical day for “Anna”:

  • She speaks into her phone, says some words, and the phone recognizes her speech, and does the right thing. …
  • Let’s say she goes into a supermarket, and the stock that’s on the shelf, some computer system has learned what everybody in that neighborhood wants to buy, and they stock the right stuff.
  • She pays with a credit card, and the credit card company has figured out, is this transaction a valid one, or is it fraudulent, should I accept it or deny it?
    She posts a picture to a social networking site, and the faces of her friends are all recognized and tagged.
  • And maybe she wants to plan a trip, and she asks for the most efficient route, and a website directs her exactly there.

So she does all these things, but every single one of them has this property: that the programmer wouldn’t necessarily know how to do it. I don’t know how to recognize speech, or recognize somebody’s face – I can’t write down the steps to do that. So I’m stuck.

So what do we do when the programmer can’t come up with a solution? The answer is, the computer could come up with a solution. We feed it some examples. And then the output of that program, rather than doing something, is it produces a new program. So it learns to write that program that we as programmers aren’t smart enough to do. …

How well does it do? The answer is, it depends on how well you train it.



Google makes its living from trying to find problems that have this kind of shape, and then trying to find the billions of examples that go with it, and then doing very well.

Norvig continues by exploring several concrete examples, with intuitive explanations instead of technical jargon:

  • Making sense of word senses
  • Translating between languages
  • Recognizing things in pictures
  • Writing captions for pictures
  • Learning to play video games by self-exploration

Physicist Richard Feynman talked about “the difference between knowing the name of something and knowing something.” Norvig truly, deeply knows machine learning – and still communicates with humility, accessibility, and a sense of wonder. I’m grateful for the opportunity to learn from him.

Further reading, for well-rounded, contrasting perspectives: