COWEN: Here’s a real softball question. What’s the optimal rate of tax on capital income?

[laughter]

SUMMERS: Closer to the tax rate on other income than to zero would be my answer to that. A fair amount of capital income reflects rents of one kind or another. Capital income is substantially held by those at the high end. There’s a fair amount of what’s really capital income in the form of unrealized capital gains that never gets taxed.

So I think the right aggregate capital income tax rate is closer to what would go with a comprehensive income tax than it is to the alternative idea that capital income taxation is just a way of taxing future consumption, and therefore you should tax future consumption and present consumption at the same rate and the tax rate should be zero.

COWEN: If we think about the 1980s, there are a lot of models from that time — some coming from your research — where you have an infinite horizon model with a zero tax rate on capital income. At some point, enough capital accumulates so that even wages are higher. And there’s a steady-state long-run argument that still the number should be zero. What has changed that makes those models less applicable? Is it that we think the elasticity is different, or is it some other variable? What’s changed in our knowledge or your understanding?

SUMMERS: At the technical level, there’s been some mathematical work showing that some of the results that you’re referring to from the 1980s were mathematically wrong. That’s one part.

The second and more consequential part is that the premise of those models was essentially that the supply of capital was infinitely elastic. Whatever the tax rate, you would drive capital to the point where the after-tax rate of return was some fixed number.

That now looks like a very poor description of reality. We’ve seem real interest rates fluctuate substantially, and we don’t see that when real interest rates are higher, savings is lots higher, and when real interest rates are lower, savings is lots lower in the way that many people, including me in the early 1980s, would have expected. So in the absence of that kind of evidence, the argument is very much attenuated.

COWEN: What if someone said, “Well, for the special 20-year period we lived through Bernanke’s East Asian savings glut, so there was always enough capital, real rates were very low. Arguably, for demographic reasons, that’s starting to end, and we’ll end up back in an era where, actually, the supply of capital with respect to the rate of return will be high again.”

Is that possible, unlikely, too far away to matter?

SUMMERS: First, one word one should never use in economics is never. I don’t want to preclude any possibility completely.

Second, you uncharacteristically made an analytic conflation there. You conflated the idea that the savings rate would fall for a variety of reasons with the idea that the savings rate would become more elastic, which is a separate issue. I don’t see any reason to think the savings rate will become more elastic.

With respect to the savings rate falling, my reading of the evidence would be different. I think that the structural factors driving low interest rates, including longer life expectancy — which makes people save more — increased insecurity, more inequality, are more likely to be semipermanent than they are to prove transient.

I think a variety of the factors holding down investment — the demographic factor, the fact that you can buy an enormous amount of capital for a very low cost, think about my iPhone — all of that I think operates in the direction of meaning that we’re likely to have this phenomenon of low real interest rates and secular stagnation for quite a long time to come.

On the best philanthropic investment in a U.S. city

COWEN: Let’s say you’re advising a philanthropist in St. Louis, and that person has $100 million to help the city. For general background, there are poor public schools, a fair amount of crime, a lot of racial segregation, but some good universities, hospitals. It’s a biotech hub.

What kind of advice would you give? How should they start thinking about this problem?

SUMMERS: Let me see, you’re asking me about a philanthropist I’ve never met . . .

COWEN: With $100 million.

SUMMERS: . . . in a city in which I’ve spent a day and a half that has relatively generic urban problems. What should the philanthropist do?

COWEN: Or the poorer parts of Boston, if you prefer.

SUMMERS: I was trying to temporize a bit while I thought.

[laughter]

SUMMERS: But that’s not this question.

I would tell the philanthropist a few general things. One is, I would say your temptation is going to be to spread the butter uniformly across the bread and to try to do everything, and that your $100 million is a lot of money, but the budget of the city of St. Louis is measured in the low billions of dollars.

You can’t do everything. If you try to do a little bit of everything, you won’t see anything you did. So you should look for a targeted couple of interventions that will be most effective. That’s the first thing I’d say.

The second thing I’d say is, you need to be very careful to make sure that whatever you think you’re buying is what you’re actually buying. If you give more money to the health budget of the city and the city responds by reallocating its own money from healthcare to other things then you’ll have demonstrated fungibility; you won’t have spurred healthcare. So have a strategy for addressing fungibility.

I would probably say that if you can do something meaningful in education to create positive examples in education, which can then be emulated . . .

COWEN: And this is K–12.

SUMMERS: Yeah. That’s likely to be the most effective thing that you can do, but that you need to be very careful not to succeed by cannibalization. Many, too many, philanthropists interested in education decide they’re going to set up a charter school. Only their charter school is only going to admit highly motivated kids with highly motivated parents.

Their charter school’s going to pay 20 percent more than the regular schools and cherry-pick the best teachers out of the regular schools. Then they’re going to be really thrilled about how they have better achievement than the regular public schools when it’s clear from the nature of their model, selecting the kids and cherry-picking the teachers, that it is supremely nonreplicable.

I would say impose a replicability constraint on yourself and innovate in the area of education. My general view has been that a lot of the way successful innovation happens is alongside big systems.

I don’t know whether it’s preschool before kindergarten. I don’t know whether it’s summer programs. I’ve been active in Boston with Citizen Schools, an organization that does after-school programs.

I’d probably, as a vehicle for getting an education, look at something that wasn’t either joining the public schools or going to war with the public schools, but was acting constructively alongside the public schools in the after school, or preschool, or transition from high school, or summer school, or some such would probably be the advice I’d give.

On Herman Melville

COWEN: There may be a few people in the audience who don’t know you’ve actually delivered a lecture at University of Chicago Law School on the Herman Melville short story, “Paradise of Bachelors, Tartarus of Maids.”

What did you learn from engaging with that story with respect to social change, or segregation, or gender issues — anything from that story as you read it as an economist and social scientist?

SUMMERS: The first thing I learned was that I should probably stick to my day job of economics because, while I typically find myself reasonably cogent relative to others in discussing economic issues, attending that literary conference, I did not find myself relatively cogent relative to other literary experts.

I am not sure I would have made it through that experience without the help of my wife, who is a professor of English at Harvard. I guess what I took, what I learned from that was that literature had a way of evoking what an economist would call the nonpecuniary aspects of work.

They are lacking when you say nonpecuniary aspects of work and you write a mathematical function, utility of consumption, leisure, and job attributes. There was a texture of alienation and depersonalization that was provided by Melville that one missed if one thought about things in the economic paradigm, is what I learned.

On unions, wages, and labor power

COWEN: Speaking of work, your last op-ed was on labor unions. You suggested a hope for a future where labor unions are stronger.

Let me tell you my worry and see if you can talk me out of it. If I look at the labor economics literature, it seems to me the union wage premium is declining. It used to be about 15 percent. Now maybe it’s 7 or 8 percent. Some studies find it’s zero.

If you’d say it’s 7 or 8 percent, if you take the 7 or 8 percent, a lot of that’s a tax on capital but surely some of it turns up in the form of higher prices. So labor gives back some of it. Surely some of it turns up in the form of lower demand for labor, so some people don’t get that job.

If I think of unions being much stronger, even without worrying about allocative efficiency costs, I just see this one-time bump upwards of maybe 3, 4 percent, which would be a gain but wouldn’t change the fundamental reality on the ground. Can you talk me into more optimism on this? Or do you think 3 or 4 percent is a lot; let’s do it?

SUMMERS: Three or four percent is pretty large relative to real wage growth that’s taken place over the last generation is the first thing I’d say. Secondly, I’d say that, as I learned reading Melville, there are important nonpecuniary aspects of jobs.

[laughter]

SUMMERS: If you look at turnover, which is a measure of job satisfaction, it’s much lower in union employers. If you look at the way in which employee grievances are dealt with, it is more respectful of those potentially discriminated against in union workplaces. If you look at job safety, it appears to be better in union workplaces. So looking at wages is a good way to miss the local benefits of unions.

The third thing is that — again, this would depend on one’s broad policy views — but that more successful unions push for policies like the preservation of Social Security, the expansion of the healthcare safety net that are for the benefit of everyone. The political impact of a society in which labor is better organized on the broad contours of policy is another way in which stronger unions can make a difference.

But part of the argument of that piece was, I don’t know what exactly the right level of unionization is. I don’t know exactly what would be optimal.

If we got a system where, if somebody tries to organize the workers at a company and the company fires the attempted organizer, and the only remedy the organizer has is to spend five years fighting through the courts where, if they win, they’ll get back salary, less whatever they earned in whatever new job they got.

The cost-benefit for the employer has to be that it’s incredibly attractive to fire the would-be organizer. We don’t have a remotely level playing field where workers are given an opportunity to make whatever choice they prefer, and that seems wrong to me.

I look at an America where corporate profits rose last year by 16 percent and wages have been stagnant. And it seems to me that more bargaining power for workers almost certainly has to be good.

I agree with you that the technological dynamics, the globalization dynamics of a modern economy mean that it’s not clear just how large the benefits of more bargaining power will be. But it’s like the question of how much weight I should lose. I don’t know how much weight I should lose, but I know which direction I should be moving.

[laughter]

SUMMERS: That’s how I feel about more labor power.

On table tennis

COWEN: In the middle of all these conversations we have an interlude where I ask the interviewees about table tennis. I understand that this summer you played in the table tennis Jewish Olympics in Tel Aviv. Is that correct?

SUMMERS: That is correct.

COWEN: What mental qualities make for a good table tennis player?

SUMMERS: Judging by my performance, qualities that I do not possess.

[laughter]

SUMMERS: I think a deft wrist, a certain capacity for concentration, and a great deal of practice. While I practiced intensely in the run-up to the activity, there were other participants who had been practicing intensely for decades. And that gave them a substantial advantage.

I also probably was not the quickest participant in that competition, even in the 60 and over division, which was the division in which I was participating. But I found it enormously satisfying to be on an Olympic team, which is not something I ever thought about.

[laughter]

SUMMERS: You heard about that, I suspect, because some magazine wrote a story about it. The reporter who wrote the story said to me, “Mr. Summers, you used to be Treasury Secretary. You did whatever you did at Harvard. How did it feel putting a track suit on and running out with 800 other Americans into a stadium?”

I said, “Well, 10-year-olds dream a lot more about international athletic competition than they do about fiscal or monetary policy.”

[laughter]

COWEN: What is the rate of productivity improvement in table tennis amongst the very best players? Someone like Jan-Ove Waldner or Ding Ning? Are they a lot better than the best players of 20, 30 years ago or just a little better?

[laughter]

SUMMERS: I don’t know. Look, obviously I don’t know. I am struck in general by the fact that a good college track athlete can now run a four-minute mile, and nobody in the world had ever run a four-minute mile before 1954. That the world record in swimming events in 1964 wouldn’t qualify you for the NCAA championships in 2017.

In the sports where we can measure things, there’s vast and rapid improvement, huge improvement. I’m not sure why people debate whether Shaq is better or worse than Wilt Chamberlain, or whether people debate whether today’s best baseball players compare with Willie Mays, or whether they debate whether Rafael Nadal could beat Bill Tilden.

My assumption is that, of course, Rafael Nadal could beat Bill Tilden, even with the same equipment. I would assume that the same thing is true in table tennis. Better learning about technique, better training approaches, greater intensity, and better equipment, I assume mean that the quality of play in all these things is much higher than it used to be.

I think it’s a mistake that people generally make, to acknowledge progress in the places where it can be readily and concretely measured, and then be in doubt about whether there is progress in the spheres where it can’t be measured. The right first approximation is to think that there’s progress in all spheres.

To take it to some place that’s outside of athletics, where I’ve been very struck, I think the evidence is overwhelming, or at least very strong, in favor of the so-called Flynn Effect, which suggests that all over the world, if you administer any kind of constant IQ test, averages in populations have been getting smarter for a long time. I think that’s probably right, as well.

On the Fed and Dreamers

COWEN: We’re here in Washington, DC. I’m going to ask you some short bullet questions about super current policy issues. Feel free to pass. You can give just a sound bite answer, but if you want to go on, please do.

Right now on the Fed, there are only three seats being held. It seems, if things go according to plan, by the beginning of 2019 everyone on that board will have been appointed by President Trump. You’ve been a lifelong Democrat, so probably your dream candidates are not the ones who are going to be appointed. But if you could give advice to the Republicans that they possibly might listen to, what would that advice be for filling these seats?

SUMMERS: Serious, thoughtful people with real expertise who will set monetary policy based on a pragmatic reading of data rather than a strong ideology or a political orientation.

COWEN: The DREAM Act is in the news right now. What’s the best way to think about where the limits to immigration should be? You’ve spoken out in favor of renewing the DREAM Act or possibly doing more. What’s the margin at which we say, “No more”?

SUMMERS: I think, on the DREAM Act, because the people are here, they’ve invested their lives and we, as a country, made a commitment to them, I think it’s a no-brainer to find ways to enable them to stay. The right broad deal on immigration is yes, there should be immigration but at least my view is the idea of the melting pot, which has become unfashionable in many circles, is actually a good idea.

The understanding should be that if you immigrate to the United States you’re immigrating to the United States to become an American. That reflects acculturation, one crucial part of which is speaking English and understanding that you’re going to be learning English and that you’re going to be carrying on your life in English. If we had more acceptance of the idea that immigration was about becoming American, we would have more acceptance of higher levels of immigration than generate comfort right now.

But one does need to understand that any country should make policy in the interests of its current citizens. It would be in the interests of America’s current citizens to have more immigrants come for all sorts of economic reasons and many ways in which it would support the economy. But when the argument is framed in terms of broad obligation to humanity and so forth, it’s understandable that there’s some reluctance to accept that argument.

On the debt ceiling, ethereum, Bitcoin, and ICOs

COWEN: Today it was announced a three-month extension of the debt ceiling. Virtually all economists think there should be no debt ceiling. That said, what’s striking is Trump seems to be doing this with the Democrats. Do you see a shift in regime where Trump will try to govern as an independent and possibly get nothing done? Or do you think there will now be agreements with Democrats in Congress?

SUMMERS: I’d be surprised if there was much where there could be agreements with Democrats. I think the debt ceiling is, in some ways, more sui generis than it is a template for things to come. I can’t imagine, for example, on tax reform or on healthcare reform, there being some coalitional arrangement of that kind.

COWEN: You’re active on Twitter now. I saw a tweet not long ago. It read something like this. “I’m going to sell all of my Ethereum and double down on VIX . See you in hell.”

If you were tweeting back to that person, Ethereum and ICOs seem to be priced very high. VIX seems to be remarkably low for what, at least to observers, feels like a very volatile period in our history, either nationally or globally.

How do you think about those asset prices?

SUMMERS: I don’t know where Ethereum is going. It makes me nervous whenever I see an asset where I think most of the demanders are buying it in the anticipation of selling it to somebody else at a higher price rather than buying it in the anticipation of some use they’ll put it to or some set of cash flows they’ll receive.

I tend to be nervous about cyber currencies and in particular some of the more libertarian arguments that are made in their favor. “Governments are known to debauch the regular currencies,” or “this will permit avoidance of excruciating regulation” and the like, I’d be pretty skeptical of.

That said, I think blockchains have fundamental technology. There will be winners that come out of its use. Which cryptocurrency it will be I don’t know, but some of what’s said makes me uneasy. With respect to the VIX, first of all, it was up 30 percent yesterday. So whatever anomaly one saw is a lot less anomalous today.

Second, the VIX — people tend to underappreciate this. The volatility of the market moves very much with the level of the market. The reason is that if a company has $100 of debt and $100 of equity, and then the stock market goes up, it’s 50/50 levered.

If the stock market goes up by $100, then it has $100 of debt and $200 of equity and it’s only one-third levered. So when the stock market goes up, its volatility naturally goes down. And the stock market has gone way up over the last 10 months. That’s a factor operating to make its volatility go significantly down.

It’s also the case if you look at surprises. The magnitude of errors in the consensus estimates of company profits or the consensus estimates of industrial production or what have you, numbers have been coming in close to consensus to an unusual degree over the last few months.

I think all those things contribute to the relatively low level of the VIX, but those are more in the way of ex post explanations. If you had told me everything that was going on in the world and asked me to guess where the VIX would be, I would expect it to have been a little higher than it is right now.

On a rational actor model for North Korea

COWEN: Today, Dennis Rodman offered to mediate between President Donald Trump and the leader of North Korea, Kim Jong Un. You wouldn’t sell yourself as an expert on North Korea, but as an economist, people who’ve negotiated with the North Koreans swear they’re rational.

Is your instinct to apply a rational actor model to try and understand what’s going on there? Or do you think it’s more like a version of behavioral economics where other ends are being pursued?

[pause]

SUMMERS: I don’t think I can answer that. What I’d say is, my instinct is the rational actor model is probably right, but that means I’m 80 percent sure that that’s right. The consequence of getting it wrong could really be huge. So I sure want to pursue the rational actor model in a way that’s doing a lot of hedging against a set of possibilities that there are behavioral aspects.

That answer seemed a little evasive. It was.

COWEN: No, no, no. It’s perfect.

On Fed policy and secular stagnation

The Fed seems to consistently come in under 2 percent for inflation. Do you think this is just a systematic mistake? Do you think it’s a plan motivated by political economy concerns, that there’s a high political cost to being above two but a much lower cost to being below two?

Do you think the implications of the semi-liquidity trap have not been fully digested yet? Do you think it’s a lag from an earlier error still percolating through the system? How do you think about the failure to meet 2 percent in an environment where, of course, it seems that if we had inflation being a bit higher, it would not be a major cost?

SUMMERS: I think the Fed has not fully grasped the reality of secular stagnation. The reality of secular stagnation is that we’ve got a very high savings propensity and a very low investment propensity. That means that the neutral real interest rate, the real interest rate that’s consistent with full employment, is very low.

That means that interest rates, which would historically have been highly expansionary interest rates, are not highly expansionary. And the Fed has underestimated the extent to which that’s true. Therefore, they’ve been disappointed by how little inflation they’ve generated. I think that’s the primary explanation, a sort of analytical misjudgment on their part of the change in the neutral interest rate.

I think, secondarily, the factors we were referring to a few moments ago, having to do with the reductions in workers’ leverage and bargaining power, means that a degree of tightness in labor markets that would in an earlier point have set off a wage spiral, no longer sets off a wage spiral because of how nervous workers are.

So the Phillips curve relationship has either broken down or shifted, and the Fed has also underestimated that. Those are two aspects. Then separate from those two aspects, I think the Fed is confused in what it’s prepared to target. It says that it has a 2 percent inflation target. But if you have a 2 percent inflation target and it is, as the Fed claims, symmetric, that means you should be above 2 percent as often as you’ve been below 2 percent. We’ve been below 2 percent for nine years now.

We’re in the ninth year of an economic recovery. The unemployment rate is at 4.3 percent. So if there was ever a time when you were going to be above 2 percent, it would seem like now — assuming recovery continues for several more years — would be that time.

Yet not a single dot in the history of the FOMC has ever been above 2 percent, at least since the great financial crisis. So I think there’s a disconnect between what they’re prepared to forecast and what they say is the nature of the 2 percent target.

My instincts would be to be more genuinely symmetric about the 2 percent and more recognizing the current policy is not quite as expansionary as they suppose, both of which would operate in the direction of caution with respect to monetary tightening.

COWEN: If there’s an ongoing demand shortfall, as is suggested by many secular stagnation approaches, does that mean monopoly cannot be a major economic problem because that’s from the supply side, and that the supply side constraint isn’t really binding if you think of there as being multiple Lagrangians. Forgive me for getting technical for a moment. Do you see what I’m saying?

SUMMERS: That wouldn’t have been the way I’d have thought about it, Tyler, but what you’re saying might be right. I think I’d be inclined to say that, if there’s more monopoly, there’s more money going to monopoly firms where there’s a low propensity to spend it, both because the firms don’t invest and because the owners of the firms tend to be rich or endowments that have a low propensity to spend.

So the greater monopoly power, to the extent that it exists, is one factor operating to raise savings and reduce investment which contributes to demand shortfalls and secular stagnation.

I also think that there’s likely to be less entry in competition in markets that aren’t growing rapidly than there is in markets that are growing rapidly. There’s a sense in which less demand over time creates its own lack of supply.

COWEN: Let me ask you a deliberately too naive question. That is, don’t wealthy people invest almost all of their money, and thus, why is that an aggregate demand shortfall?

SUMMERS: Wealthy people don’t put it all under their mattress, they put it all in financial assets of one kind or another. But when they put it all in financial assets of one kind or another, they drive up the prices of those financial assets, which is driving down the return on those assets, which is driving down the natural or neutral level of interest rates.

COWEN: But in a Q theory framework, won’t that increase investment if the price of the stock market goes up?

SUMMERS: It operates in the direction of raising investment, but if there are limits on how far interest rates can fall — because safe interest rates can’t fall much below zero — and other interest rates have to hold spreads relative to safe interest rates, you may not be able to get the system to equilibrate.

Even if, in a particular moment, the safe interest rates are above zero, the awareness that, at some future moment when demand falls, you will not be able to reduce interest rates sufficiently, operates as a deterrent that holds asset prices and demand down.

COWEN: What’s the best framework for thinking about how the Federal Reserve’s monetary policy decisions affect emerging economies? There’s been work coming out of Princeton and Hélène Rey. How well do we understand this? And what framework do you look to first on these questions?

SUMMERS: I think we don’t fully understand why there are so many places where the long-term interest rate is more responsive to what the central bank in Washington does than what the central bank in their capital city does. That’s an important puzzle that I don’t think the economics profession fully understands at this point.

I don’t know that we’ve got a framework that is, in a profound sense, better than the Dornbusch overshooting model for thinking about a range of exchange rate fluctuations.

On NAFTA and Mexican economic performance

COWEN: If I think about the economy of Mexico, they have NAFTA, which I’ve always thought was a good idea. They’ve done significant reforms. They have a much, much higher level of professionalism in their government than they did several decades ago.

But typically they grow in the range of two to two and half percent. This, to me, is somewhat of a puzzle. What’s your take on why Mexico hasn’t done better?

SUMMERS: I agree with you that it’s a puzzle. If you had described to me in 1995, 1996, after our bailout, if you had said to me, “This is going to be the quality of Mexican economic management. This is going to be the kind of leadership that Mexico’s going to have. How fast do you think they’ll grow?” I would have expected more convergence with the United States than we’ve had.

I agree with you that it’s a puzzle, and it’s not what I would have expected. In terms of understanding it, I would highlight two things.

One is, Mexico has still some very serious rule-of-law issues, some very serious security issues associated with the risks of becoming more like Colombia than it wants to be, and I suspect those are larger issues than many outside Mexico appreciate. I think that’s one part of it.

The other part of it is a much broader phenomenon. In some ways, I think Mexico’s problem is the same problem as Michigan’s problem. We have this tremendous phenomenon of globalization, and there’s some who are able to seize the opportunity to produce and to build supply chains across many countries, to sell their products into a global economy rather than into a domestic economy.

There’s some, who by being part of the global economy, like hundreds of millions of people in China or India, are lifted to an entirely new level than they were before.

But there’s a group in between those who are levering the global economy and those who are being pulled along by the global economy, who can’t really turn the global economy into an opportunity, and don’t really want to compete with Chinese labor, who are a bit left behind and frustrated.

I think there would be many in Latin America who would be in that group, including Mexico. I think it’s a very profound social and economic problem.

On mounting Chinese debt

COWEN: Speaking of puzzles, you mentioned China. We’re still witnessing this ongoing race between Chinese debt and Chinese nominal GDP.

I would say since maybe 2006, at least I have been expecting some kind of discrete event to occur in China, where we all say, “Uh-oh, now it’s happened.” It actually hasn’t been the case. That, to me, is a puzzle, and after 11 years, I wonder what’s going on.

Can you imagine that we’re finally at a frontier, where you think Chinese nominal GDP can out-race Chinese debt, or do you think we still ought to expect some discrete event?

SUMMERS: I think it’s important to remember that a large part of Chinese debt is explicitly or implicitly government guaranteed, and that the fiscal capacity of the Chinese government is very large, especially given rapid economic growth.

I think the application of Western thinking that understates the extent to which the debts are really internal government debts probably leads to more alarm about a sudden breakdown of Chinese finances than is warranted.

COWEN: So you could imagine a smooth glide into a lower growth path? Readily imagine?

SUMMERS: Yes. I could definitely imagine that, and my best guess would be that Chinese growth will slow , but I would not be confident that China will have an event like the Asian countries had in 1997. It wouldn’t amaze me if they did, but I would not want to predict that within the next five years, they will have an event of that kind, simply because the government is standing behind so large a part of the economy.

COWEN: Do you worry about the fact that so much is guaranteed that they’re simply stacking on top, and at some point, you end up getting a lot of bad investment decisions? You only need so many roads or so many high-speed rail lines.

Until they, at some margin, pull away those guarantees, they’re just doubling down, and the debts will get bigger, provincial governments will fall, there will be defaults on corporate bonds, and then they’ll be in a very difficult situation.

Or do you think that moral hazard problem they’ll somehow keep on managing to maintain within acceptable levels?

SUMMERS: I think both are possible. That’s a very difficult question. You’re a rapidly growing Asian country, and you decide to build an airport somewhere where there aren’t very many people.

You do that because it’s going to be cheap to build it now, when there aren’t a lot of people around, and it’s important to establish the land. Then 15 years later, if you’ve grown fast, you’re a brilliant hero of masterful long-term planning, and if you’ve grown slow, you’re an idiot who’s building an airport where nobody wanted to go.

It’s hard to know which is the case. The Japanese probably look stupider than they really were because they built in anticipation of more growth than they, in fact, were able to generate.

I don’t know quite what’s going to happen with respect to China. It’s not unlike Dulles Airport. Today we think of Dulles Airport as an important strategic stroke that cemented and made possible a hugely vibrant technological economy in Northern Virginia.

I’m old enough to remember in the late ’70s or early ’80s when it was a generation after Dulles Airport was built. It was a ridiculous idea, built a million miles from anywhere, and like, “Why did we build that?”

So it’s very difficult to judge some of these infrastructure investments except in the very, very long run. Even when you judge them in the very, very long run, you can make a judgment ex post as to whether they were a good idea, but even then, it’s going to be even more difficult to make a judgment about whether they were ex ante sensible or ex ante not sensible.

On Russia

COWEN: If I think about Russia and its recent history, I’m never sure if I should feel it’s gone better than we could have expected or if it’s gone worse. If one redoes that whole history, and I’m not sure who is the relevant we here, but what could have been done differently and better with regard to Russia?

Again, with the we variable the choice variable, being a bit undefined here, what do you think of as the lost opportunity in that regard, or was there none?

SUMMERS: Factually, if you had told any of us who worked with President Clinton as he prepared for his first summit with Boris Yeltsin in 1993 where the Russian economy, Russia’s government, and Russian relations with the United States were going to be in 2017, we would have been appalled.

COWEN: Sure, but in 1988 you might have thought it was an OK deal.

SUMMERS: Right. So by the standards of what one was hoping for in the aftermath of the Berlin Wall falling and the end of the Cold War, where the models would have been West Germany 26 years after 1945, you have to say it’s been a big disappointment.

I’ve thought a lot about that. One big part of it is, 60 years of communism, not a clear defeat — it was harder to have it join the world than it was to repair Europe after the Second World War. The structural reasons why it was more difficult were things we overestimated.

Second — this is maybe the important point — in retrospect, there was not enough respect shown to what once had been a proud nation that didn’t think of itself as having been defeated. That manifested itself in some of the efforts to push NATO very far towards its borders. That manifests itself in the way in which conditionality was applied by the international financial institutions, and to some extent, by the G7 countries, including the United States.

There were probably errors in not providing enough funding quickly enough at the very beginning, at the moment of maximum malleability. Those were errors probably primarily of late Bush administration, but perhaps errors of our administration, as well, at the beginning of 1993.

In retrospect, my judgment would be that repairing Russia’s economy was a much more difficult challenge than we appreciated.

On the Larry Summers production function

COWEN: Two final questions. My first one is about what I call the Larry Summers production function.

You were successful at quite a young age, but what I find striking after reviewing a lot of what you’ve done and a lot of talks you’ve given, is I find at your current age, 62, that when you answer questions on YouTube, in general your answers are in some way better or richer than they would have been 5, 10 years ago, and they were already, obviously, quite good to have gotten you to where you are.

For people who are already famous, and to some extent well off, at the age margin of whatever to 62, for their answers still to be getting better, this is what I find striking. You don’t have to be modest here, but what is there in the Larry Summers production function that explains this?

[laughter]

SUMMERS: Part of the answer goes back to the first question that you asked. I’ve always tried to surround myself and be around extraordinarily able young people. They probably do learn some things from me, but I learn a lot from them, both from things they say and know that I don’t, and from the questions they ask, which keep me on my toes. That’s one answer.

Another answer — by the way, I’m flattered that you think it’s true. I don’t know if it’s true or not. One thing I’ve always tried to define myself by — and sometimes it’s been more successful than other times — is opposition to complacency and not being satisfied with any institution, with myself, or with anybody, and always thinking things could be better.

That’s the attitude I have to myself, as well. I’ll be on a plane back tonight, and I’ll be thinking about the various questions you asked and which questions I could have given better answers to. Then I’ll think about those answers, and the next time I’m somewhere, I’ll give better answers or be better at discussing things because I wasn’t complacent or satisfied.

Those would probably be the answers I’d give.

On final acts

COWEN: Before Q&A, my final question. This relates to a piece by your wife. She’s a professor at Harvard, in the English department, Elisa New.

In a magazine called Tablet she has a piece on what she calls final acts, namely wonderful things people do at the very end of their careers. She focuses on Philip Roth and Harold Bloom.

I would ask, can you think of a senior economist you have known or interacted with, where they truly ended their career on something wonderful as a final act, where you look at that and admire that? What would that be? You can cite relatives, by the way, if you want.

[laughter]

SUMMERS: I began by saying how much my wife admired you because of your book on food. That will be as nothing compared to the fact that you had looked up her writing . . .

[laughter]

SUMMERS: . . . in preparation for this interview.

COWEN: That’s my noncomplacency.

SUMMERS:Ken Arrow was my uncle. He was a brilliant man who could talk with extraordinary intelligence on any subject, who died this past March at the age of 94.

Six months before he died, he was the stalwart with the best attendance record at the Mathematical Economics Institute that he had led for the previous 25 years in Jerusalem, and was, I’m told, as active as any participant, and more sharply contributing than the vast majority.

To be able to do that at the age of 93 with enthusiasm and zest is extraordinary, and is a great example for me and for many others.

COWEN: Larry Summers, thank you very much.

Q&A

AUDIENCE MEMBER: First of all, thank you so much for the great exchange. The question is very simple. US economic growth is at 2 percent. Is this really the new normal, which is to say that the postwar average of 3 percent was an aberration, or are we going to find new ways? What will be the new ways — that’s the question — to go back to 3 percent?

SUMMERS: Mostly new normal because the labor force was growing much more rapidly because of population dynamics, because of immigration, and because of the tailwind from more women working through most of the postwar period than it’s likely to be going forward. So productivity may accelerate a bit, but historical average growth we’re unlikely to see.

COWEN: Next question.

AUDIENCE MEMBER: I’m mostly concerned about the digital economy. I’m very concerned that economists haven’t really explored a lot of really important issues, and as a result, policymakers are making bad choices, uninformed by economics, for example on copyright —

COWEN: Question. Question. Question.

AUDIENCE MEMBER: The question, basically: Are there two or three really important issues that economists should be looking at in the world of the digital economy — like the value of copyright — that you think would need more attention, which is a homework assignment for economists looking for hot new topics?

SUMMERS: I think there’s a lot of work to be done on intellectual property, its marketing, and the ways in which intellectual property is enforced. I think there’s a lot of work to be done on market structure in digital industries, what constitutes increased competition, and what constitutes measures that foreclose competition.

There’s a lot of work to be done on intangible infrastructures that promote exchange, blockchain being one example, but I don’t understand why there should still be three-day financial settlements on anything in the world of today’s technology.

AUDIENCE MEMBER: I’ve got a question on your paper with Natasha Sarin, and you’re slightly critical of the regulatory stress tests for being over-reliant on regulatory measures for capital. My question is, what would you redesign about the CCAR stress tests if you had a chance?

SUMMERS: I would put more emphasis on market values. It’s madness that in the spring of 2008, when the stock prices of all the banks was collapsing, everybody was saying they were totally, splendidly capitalized and all was well. I would put more reliance on market instruments.

I wouldn’t allow the market to satisfy me that everything was OK because sometimes markets are wrong and they don’t see problems. But, if markets were alarmed, I would tend to think I should be alarmed, as well, and to build more of that apparatus in.

The fact that the stress tests are saying that, even if we have an event substantially worse than the 2008 crisis, no bank will need to raise capital — that says more to me about the stress tests than it does about the banks.

Since I’m expecting more than my 1 2 regular readers to read this, let me add a preface. I’ve been managing the MathJax project for 4 years now. In the last two years, I’ve also been consulting for publishers regarding math-related workflows, in particular TeX-to-XML and XML-to-web, both on the back and front-end. Not that it says much, but I’m an invited expert on the W3C Digital Publishing Interest Group (shoutout to Jean and Tzviya!) and I also recently left the W3C Math Working Group (to which David had invited me in late 2014). MathML is a big (positive) part of my professional life.

I recently posted a terse – uhm, shall we say summary? – of my thoughts on MathML on a11ySlackers; and I promised a blog post. There’s now a 6000 word thingie sitting in my drafts which would take months to whip into shape. So I tried again and it now feels both too long and too short; oh well, maybe it leads somewhere, maybe it doesn’t.

Needless to say, opinions posted on my personal website are my personal opinions (funny how that works). In particular, they do not reflect the opinions of any of my clients, let alone the team at MathJax. I think they don’t particularly help anything or anyone specifically except, perhaps, in encouraging a more open and realistic discussion.

The gist of it

MathML is a failed web standard.^*

We can do better, we deserve better.

MathML-in-HTML5 is in the way of that.

^*Some people might prefer “browser standard”, as in “a web standard to be implemented natively in the browser” since some web standards do not rely on browser implementations. Also, “natively” as opposed to some web-components hack shipped in a browser.

MathML is a failed web standard

It doesn’t matter whether or not MathML is a good XML language. Personally, I think it’s quite alright. It’s also clearly a success in the XML publishing world, serving an important role in standards such as JATS and BITS.

The problem is: MathML has failed on the web.

Luckily, many technologies have succeeded and today MathML is neither necessary but also no longer sufficient for math on the web. Instead of one monolithic solution, we have many. We should acknowledge this and move forward towards several newer and smaller standards that actually help developers.

Here are a few reasons that make me say these things.

1. MathML has not been significantly supported by browser vendors, neither on the spec level nor on the implementation level.

You might easily think they do (Office! ChromeVox! VoiceOver!) but the browser vendors actually don’t. The partial MathML implementations in Gecko and WebKit are entirely the work of volunteers. Largely unpaid, largely unsupervised, largely unaccountable.

Not a single browser vendor has stated an intent to work on the code, not a single browser developer has been seen on the MathWG. After 18 years, not a single browser vendor is willing to dedicate even a small percentage of a developer to MathML.

This is where the story should end, really. But sadly it doesn’t. MathML’s success in the XML world has kept it alive, but not for the benefit of anyone on the web.

MathML is a poor web standard and it would be better to remove it from HTML 5.

2. Browser implementations of MathML are not used.

If you look at publicly available crawler data, you’ll notice that it’s hard to find examples of MathML that aren’t behind paywalls. If you look further, you’ll hardly find an example where people providing MathML content rely on native MathML implementations; even on Gecko and WebKit they use MathML-to-HTML5 converters. Another indicator is that, despite implementations having subtly deteriorated in the past two years, people aren’t even complaining (I mean, WebKit stopped drawing surds (try this in Safari 8) but apparently nobody cared enough to even file a bug). Actual developer problems are so extreme you can’t seriously develop anything slightly advanced with MathML (e.g., Gecko has non-existent or incomplete support for basic APIs such as style, dataset, or event handlers for MathML elements).

3. Content MathML has failed to provide usable semantics.

Ok, truth be told, I don’t know. The problem is: it’s nearly impossible to generate good Content MathML (except with massive manual labor). As far as I know there is not a single significant collection of mathematics encoded in Content MathML out there. It’s mainly ephemeral research projects and some hand-crafted projects. That’s fine, we need research after all, but that is not a standard fit for the web.

4. Presentation MathML fails front-end developers because it unnecessarily binds layout features to MathML elements instead of providing a usable set of CSS features.

Now <mstyle>, <mspace>, <mpadded>, <mphantom>, <menclose>, <mfenced>, <mtable>, <mstack> might sound funny to a web developer but it’s a serious problem. The web has found a productive separation of concern. MathML is incompatible with this approach.

5. MathML fails because it does not specify layout sufficiently.

MathML assumes an implementor would know or care about the intricacies and traditions of math layout. How do you draw a surd? Not specified. How do you draw a fraction? Not specified. How do you space things? Not specified. [But yes, dear implementor, you should support arcane mathematical layout features like movable limits, operator dictionaries, the subtle spacing and layout difference of inline- and display-style and so forth; you know why they’re important, right? RIGHT? And also make sure to implement 5 different approaches to vertical stacking, because, reasons – kthx, xxo.]

6. Presentation MathML fails because CSS is slowly implementing all layout features that mathematical layout needs, making it obsolete.

Today, lots of tools will let you render mathematics using CSS. It’s messy but it works everywhere (ok, dear IE7 user, not for you, I’m sorry). The time when MathML implementations would have significantly enhanced web layout features are past.

7. Presentation MathML fails to provide sufficient semantics.

Neil Soiffer wrote ingenious heuristics for MathPlayer which makes most people think that Presentation MathML makes mathematics accessible. That’s about as accurate as saying OCR means all images with text are actually accessible.

The reality is that even for school-level math you need both high-quality Presentation MathML (which is rare in itself) combined with powerful (but inevitably fallible) heuristics to extract meaningful semantic information; that’s acceptable in the short run but not a real solution for mathematical semantics on the web.

8. The MathML spec is not actively developed.

MathML has seen no significant activity in almost a decade. In the industrial XML world, MathML is a success and people want more features but improvements are not even brought up. It seems nobody wants to jeopardize an adoption on the web. MathML being a web standard is negatively affecting even those users who actually embrace it because MathML is stuck in maintenance mode.

Did you know the MathWG’s charter is running out this month? Would you notice if it wasn’t renewed and the WG would cease existing? Would you notice if WebKit and Gecko ripped out their MathML implementation tomorrow? I’m not sure many people would.

What to do next.

A) MathML needs to be dropped from HTML 5.

Many people I’ve met have the mistaken impression that browser manufacturers have declared an intent to implement everything in the set of standards usually called HTML 5. They have not (even if HTML 5 as a “spec” may strive for that).

I think as long as MathML is in that set of standards, the lame duck argument (“it’s a standard!”) will continue to prevent alternative developments that help the actually working solutions for mathematics on the web.

At this point, MathML is effectively preventing mathematics from aligning with today’s and tomorrow’s web. This is hurting everyone. We need to drop MathML to make room for better standards.

B) Math layout can and should be done in CSS and SVG. Let’s improve them incrementally to make it simpler.

It’s possible to generate HTML+CSS or SVG that renders any MathML content – on the server, mind you, no client-side JS required (but of course possible). The resulting markup is arguably crap – it’s span soup at its worst and some use cases are difficult to realize. But we’ve been there with HTML and CSS; people know how to solve this. It got us standards like flexbox and css-grid; it’s worth pursuing improvements to those standards that work instead of waiting for Godot.

It’s also difficult to write your own math rendering tool. But we need more ideas, not less! It shouldn’t be harder to write a simple math renderer in CSS or SVG than it is to write a RWD framework or a vector graphics library.

We don’t need Presentation MathML for this even if many projects (like MathJax) use it as an internal format. MathML’s failure as a web standard is hurting the web because it is blocking discussions about improving existing standards to help existing mathematics tools on the promise that eventually “MathML will solve everything (tm)”.

I can’t see a native MathML approach help to fill these final gaps. What existing rendering solutions need has little to do with what MathML implementations need. We don’t need underspecified layout features tied to MathML elements, we need flexible CSS features that are integrated into existing CSS. Most importantly, existing solutions can iterate on partial improvements to ensure that these help layout on the web more generally, not just the needs of one specific mathematical markup language.

We don’t need one true approach to math layout, we need flexibility for developers to be innovative and pursue new ways of solving layout problems and expressing mathematical thought on the web.

We need to get together with CSSWG/Houdini TF/etc to work out solutions that help those developers who actually solve the problem of math on the web.

To give a rough idea – From a MathJax point of view, three areas are difficult in CSS right now (and probably universally for math layout tools on the web):

• vertical stacking (although flexbox is probably already enough to fix that and CSS Ruby might also be interesting to look at for synergies)
• stretchy characters and enclosures (this is the BIG one – but they’re really just fancy borders)
• tight character bounding boxes (math layout has stronger requirements on typography than most forms of text but even existing technology is problematic)

Stretchy things are by far the biggest layout question, if only because they once led Ojan Vafai to call math layout fundamentally incompatible with CSS layout. As much as I respect his expertise, that cannot be the answer. It seems unlikely that we can’t incrementally reduce the complexity for existing rendering solutions; in any case, it has little to do with MathML.

C) We need a new approach for exposing semantics.

Since layout is practically solved (or at least achievable), we really need to solve the semantics. Presentation MathML is not sufficient, Content MathML is just not relevant.

We need to look where the web handles semantics today – that’s ARIA and HTML but also microdata, rdfa etc. Especially ARIA is an extremely urgent problem because it currently ties mathematics entirely to Presentation MathML elements (where it fails) instead of providing a way to enrich all mathematical rendering on the web.

We also need to look beyond the semantics of mathematics into the semantics of mathematics in its applications, e.g., mathematical notation out of physics (units etc), chemistry (isotopes, reactions etc) and biology (trees, graphs etc). We need to find ways to expose this information to assistive technologies, search and other tools.

Two weeks ago, a pair of researchers from Stanford University made a startling claim. Using hundreds of thousands of images taken from a dating website, they said they had trained a facial recognition system that could identify whether someone was straight or gay just by looking at them. The work was first covered by The Economist, and other publications soon followed suit, with headlines like “New AI can guess whether you're gay or straight from a photograph” and “AI Can Tell If You're Gay From a Photo, and It's Terrifying.”

As you might have guessed, it’s not as straightforward as that. (And to be clear, based on this work alone, AI can’t tell whether someone is gay or straight from a photo.) But the research captures common fears about artificial intelligence: that it will open up new avenues for surveillance and control, and could be particularly harmful for marginalized people. One of the paper’s authors, Dr Michal Kosinski, says his intent is to sound the alarm about the dangers of AI, and warns that facial recognition will soon be able to identify not only someone’s sexual orientation, but their political views, criminality, and even their IQ.

With statements like these, some worry we’re reviving an old belief with a bad history: that you can intuit character from appearance. This pseudoscience, physiognomy, was fuel for the scientific racism of the 19th and 20th centuries, and gave moral cover to some of humanity’s worst impulses: to demonize, condemn, and exterminate fellow humans. Critics of Kosinski’s work accuse him of replacing the calipers of the 19th century with the neural networks of the 21st, while the professor himself says he is horrified by his findings, and happy to be proved wrong. “It’s a controversial and upsetting subject, and it’s also upsetting to us,” he tells The Verge.

But is it possible that pseudoscience is sneaking back into the world, disguised in new garb thanks to AI? Some people say machines are simply able to read more about us than we can ourselves, but what if we’re training them to carry out our prejudices, and, in doing so, giving new life to old ideas we rightly dismissed? How are we going to know the difference?

Can AI really spot sexual orientation?

First, we need to look at the study at the heart of the recent debate, written by Kosinski and his co-author Yilun Wang. Its results have been poorly reported, with a lot of the hype coming from misrepresentations of the system’s accuracy. The paper states: “Given a single facial image, [the software] could correctly distinguish between gay and heterosexual men in 81 percent of cases, and in 71 percent of cases for women.” These rates increase when the system is given five pictures of an individual: up to 91 percent for men, and 83 percent for women.

On the face of it, this sounds like “AI can tell if a man is gay or straight 81 percent of the time by looking at his photo.” (Thus the headlines.) But that’s not what the figures mean. The AI wasn’t 81 percent correct when being shown random photos: it was tested on a pair of photos, one of a gay person and one of a straight person, and then asked which individual was more likely to be gay. It guessed right 81 percent of the time for men and 71 percent of the time for women, but the structure of the test means it started with a baseline of 50 percent — that’s what it’d get guessing at random. And although it was significantly better than that, the results aren’t the same as saying it can identify anyone’s sexual orientation 81 percent of the time.

As Philip Cohen, a sociologist at the University of Maryland who wrote a blog post critiquing the paper, told The Verge: “People are scared of a situation where you have a private life and your sexual orientation isn’t known, and you go to an airport or a sporting event and a computer scans the crowd and identifies whether you’re gay or straight. But there’s just not much evidence this technology can do that.”

Kosinski and Wang make this clear themselves toward the end of the paper when they test their system against 1,000 photographs instead of two. They ask the AI to pick out who is most likely to be gay in a dataset in which 7 percent of the photo subjects are gay, roughly reflecting the proportion of straight and gay men in the US population. When asked to select the 100 individuals most likely to be gay, the system gets only 47 out of 70 possible hits. The remaining 53 have been incorrectly identified. And when asked to identify a top 10, nine are right.

If you were a bad actor trying to use this system to identify gay people, you couldn’t know for sure you were getting correct answers. Although, if you used it against a large enough dataset, you might get mostly correct guesses. Is this dangerous? If the system is being used to target gay people, then yes, of course. But the rest of the study suggests the program has even further limitations.

What can computers really see that humans can’t?

It’s also not clear what factors the facial recognition system is using to make its judgements. Kosinski and Wang’s hypothesis is that it’s primarily identifying structural differences: feminine features in the faces of gay men and masculine features in the faces of gay women. But it’s possible that the AI is being confused by other stimuli — like facial expressions in the photos.

This is particularly relevant because the images used in the study were taken from a dating website. As Greggor Mattson, a professor of sociology at Oberlin College, pointed out in a blog post, this means that the images themselves are biased, as they were selected specifically to attract someone of a certain sexual orientation. They almost certainly play up to our cultural expectations of how gay and straight people should look, and, to further narrow their applicability, all the subjects were white, with no inclusion of bisexual or self-identified trans individuals. If a straight male chooses the most stereotypically “manly” picture of himself for a dating site, it says more about what he thinks society wants from him than a link between the shape of his jaw and his sexual orientation.

To try and ensure their system was looking at facial structure only, Kosinski and Wang used software called VGG-Face, which encodes faces as strings of numbers and has been used for tasks like spotting celebrity lookalikes in paintings. This program, they write, allows them to “minimize the role [of] transient features” like lighting, pose, and facial expression.

But researcher Tom White, who works on AI facial system, says VGG-Face is actually very good at picking up on these elements. White pointed this out on Twitter, and explained to The Verge over email how he’d tested the software and used it to successfully distinguish between faces with expressions like “neutral” and “happy,” as well as poses and background color.

Image: Kosinski and Wang

Speaking to The Verge, Kosinski says he and Wang have been explicit that things like facial hair and makeup could be a factor in the AI’s decision-making, but he maintains that facial structure is the most important. “If you look at the overall properties of VGG-Face, it tends to put very little weight on transient facial features,” Kosinski says. “We also provide evidence that non-transient facial features seem to be predictive of sexual orientation.”

The problem is, we can’t know for sure. Kosinski and Wang haven’t released the program they created or the pictures they used to train it. They do test their AI on other picture sources, to see if it’s identifying some factor common to all gay and straight, but these tests were limited and also drew from a biased dataset — Facebook profile pictures from men who liked pages such as “I love being Gay,” and “Gay and Fabulous.”

Do men in these groups serve as reasonable proxies for all gay men? Probably not, and Kosinski says it’s possible his work is wrong. “Many more studies will need to be conducted to verify [this],” he says. But it’s tricky to say how one could completely eliminate selection bias to perform a conclusive test. Kosinski tells The Verge,“You don’t need to understand how the model works to test whether it’s correct or not.” However, it’s the acceptance of the opacity of algorithms that makes this sort of research so fraught.

If AI can’t show its working, can we trust it?

AI researchers can’t fully explain why their machines do the things they do. It’s a challenge that runs through the entire field, and is sometimes referred to as the “black box” problem. Because of the methods used to train AI, these programs can’t show their work in the same way normal software does, although researchers are working to amend this.

In the meantime, it leads to all sorts of problems. A common one is that sexist and racist biases are captured from humans in the training data and reproduced by the AI. In the case of Kosinski and Wang’s work, the “black box” allows them to make a particular scientific leap of faith. Because they’re confident their system is primarily analyzing facial structures, they say their research shows that facial structures predict sexual orientation. (“Study 1a showed that facial features extracted by a [neural network] can be used to accurately identify the sexual orientation of both men and women.")

Experts say this is a misleading claim that isn’t supported by the latest science. There may be a common cause for face shape and sexual orientation — the most probable cause is the balance of hormones in the womb — but that doesn’t mean face shape reliably predicts sexual orientation, says Qazi Rahman, an academic at King’s College London who studies the biology of sexual orientation. “Biology’s a little bit more nuanced than we often give it credit for,” he tells The Verge. “The issue here is the strength of the association.”

The idea that sexual orientation comes primarily from biology is itself controversial. Rahman, who believes that sexual orientation is mostly biological, praises Kosinski and Wang’s work. “It’s not junk science,” he says. “More like science someone doesn’t like.” But when it comes to predicting sexual orientation, he says there’s a whole package of “atypical gender behavior” that needs to be considered. “The issue for me is more that [the study] misses the point, and that’s behavior.”

Reducing the question of sexual orientation to a single, measurable factor in the body has a long and often inglorious history. As Matton writes in his blog post, approaches have ranged from “19th century measurements of lesbians’ clitorises and homosexual men’s hips, to late 20th century claims to have discovered ‘gay genes,’ ‘gay brains,’ ‘gay ring fingers,’ ‘lesbian ears,’ and ‘gay scalp hair.’” The impact of this work is mixed, but at its worst it’s a tool of oppression: it gives people who want to dehumanize and persecute sexual minorities a “scientific” pretext.

Jenny Davis, a lecturer in sociology at the Australian National University, describes it as a form of biological essentialism. This is the belief that things like sexual orientation are rooted in the body. This approach, she says, is double-edged. On the one hand, it “does a useful political thing: detaching blame from same-sex desire. But on the other hand, it reinforces the devalued position of that kind of desire,” setting up hetrosexuality as the norm and framing homosexuality as “less valuable … a sort of illness.”

And it’s when we consider Kosinski and Wang’s research in this context that AI-powered facial recognition takes on an even darker aspect — namely, say some critics, as part of a trend to the return of physiognomy, powered by AI.

Your character, as plain as the nose on your face

For centuries, people have believed that the face held the key to the character. The notion has its roots in ancient Greece, but was particularly influential in the 19th century. Proponents of physiognomy suggested that by measuring things like the angle of someone’s forehead or the shape of their nose, they could determine if a person was honest or a criminal. Last year in China, AI researchers claimed they could do the same thing using facial recognition.

Their research, published as “Automated Inference on Criminality Using Face Images,” caused a minor uproar in the AI community. Scientists pointed out flaws in the study, and concluded that that work was replicating human prejudices about what constitutes a “mean” or a “nice” face. In a widely shared rebuttal titled “Physiognomy’s New Clothes,” Google researcher Blaise Agüera y Arcas and two co-authors wrote that we should expect “more research in the coming years that has similar … false claims to scientific objectivity in order to ‘launder’ human prejudice and discrimination.” (Google declined to make Agüera y Arcas available to comment on this report.)

Kosinski and Wang’s paper clearly acknowledges the dangers of physiognomy, noting that the practice “is now universally, and rightly, rejected as a mix of superstition and racism disguised as science.” But, they continue, just because a subject is “taboo,” doesn’t mean it has no basis in truth. They say that because humans are able to read characteristics like personality in other people’s faces with “low accuracy,” machines should be able to do the same but more accurately.

Kosinski says his research isn’t physiognomy because it’s using rigorous scientific methods, and his paper cites a number of studies showing that we can deduce (with varying accuracy) traits about people by looking at them. “I was educated and made to believe that it’s absolutely impossible that the face contains any information about your intimate traits, because physiognomy and phrenology were just pseudosciences,” he says. “But the fact that they were claiming things without any basis in fact, that they were making stuff up, doesn’t mean that this stuff is not real.” He agrees that physiognomy is not science, but says there may be truth in its basic concepts that computers can reveal.

For Davis, this sort of attitude comes from a widespread and mistaken belief in the neutrality and objectivity of AI. “Artificial intelligence is not in fact artificial,” she tells The Verge. “Machines learn like humans learn. We’re taught through culture and absorb the norms of social structure, and so does artificial intelligence. So it will re-create, amplify, and continue on the trajectories we’ve taught it, which are always going to reflect existing cultural norms.”

We’ve already created sexist and racist algorithms, and these sorts of cultural biases and physiognomy are really just two sides of the same coin: both rely on bad evidence to judge others. The work by the Chinese researchers is an extreme example, but it’s certainly not the only one. There’s at least one startup already active that claims it can spot terrorists and pedophiles using face recognition, and there are many others offering to analyze “emotional intelligence” and conduct AI-powered surveillance.

Facing up to what’s coming

But to return to the questions implied by those alarming headlines about Kosinski and Wang’s paper: is AI going to be used to persecute sexual minorities?

This system? No. A different one? Maybe.

Kosinski and Wang’s work is not invalid, but its results need serious qualifications and further testing. Without that, all we know about their system is that it can spot with some reliability the difference between self-identified gay and straight white people on one particular dating site. We don’t know that it’s spotted a biological difference common to all gay and straight people; we don’t know if it would work with a wider set of photos; and the work doesn’t show that sexual orientation can be deduced with nothing more than, say, a measurement of the jaw. It’s not decoded human sexuality any more than AI chatbots have decoded the art of a good conversation. (Nor do its authors make such a claim.)

Image: Faception

The research was published to warn people, say Kosinski, but he admits it’s an “unavoidable paradox” that to do so you have to explain how you did what you did. All the tools used in the paper are available for anyone to find and put together themselves. Writing at the deep learning education site Fast.ai, researcher Jeremy Howard concludes: “It is probably reasonably [sic] to assume that many organizations have already completed similar projects, but without publishing them in the academic literature.”

We’ve already mentioned startups working on this tech, and it’s not hard to find government regimes that would use it. In countries like Iran and Saudi Arabia homosexuality is still punishable by death; in many other countries, being gay means being hounded, imprisoned, and tortured by the state. Recent reports have spoken of the opening of concentration camps for gay men in the Chechen Republic, so what if someone there decides to make their own AI gaydar, and scan profile pictures from Russian social media?

Here, it becomes clear that the accuracy of systems like Kosinski and Wang’s isn’t really the point. If people believe AI can be used to determine sexual preference, they will use it. With that in mind, it’s more important than ever that we understand the limitations of artificial intelligence, to try and neutralize dangers before they start impacting people. Before we teach machines our prejudices, we need to first teach ourselves.

The thick, black-and-white rulebook packaged with every copy of the 1979 war-game The Campaign For North Africa is full of obtuse decrees, but the tabletop community always had a special appreciation for entry 52.6 - affectionately known as the “macaroni rule.” The Italian troops in World War II were outfitted with noodle rations, and in the name of historical dogma, the player responsible for the Italians is required to distribute an extra water ration to their forces, so that their pasta may be boiled. Soldiers that do not receive their “pasta point” may immediately become “disorganized,” rendering them useless in the field. It’s a fact of life really: if the Italians can’t boil their pasta, the Italians may desert.

It was a joke, by the way. Richard Berg, the legendary game designer and author of TheCampaign For North Africa, says so himself. He’ll happily admit that this was an unreasonable game for unreasonable people, but still, a pasta point? There’s attention to detail, and then there’s taking the piss. As Berg explains, the rule wasn’t even entirely factually accurate. “The reality is that the Italians cooked their pasta with the tomato sauce that came with the cans,” he says. “But I didn’t want to do a rule on that.” Yes, at the pinnacle of North Africa’s ridiculous excess, even Berg couldn’t help but poke a little fun at the obsessives in his wake.

It’ll take you about 1,500 hours (or 62 days) to complete a full play of The Campaign For North Africa. The game itself covers the famous WWII operations in Libya and Egypt between 1940 and 1943. Along with the opaque rulebook, the box includes 1,600 cardboard chits, a few dozen charts tabulating damage, morale, and mechanical failure, and a swaddling 10-foot long map that brings the Sahara to your kitchen table. You’ll need to recruit 10 total players, (five Allied, five Axis,) who will each lord over a specialized division. The Front-line and Air Commanders will issue orders to the troops in battle, the Rear and Logistics Commanders will ferry supplies to the combat areas, and lastly, a Commander-in-Chief will be responsible for all macro strategic decisions over the course of the conflict. If you and your group meets for three hours at a time, twice a month, you’d wrap up the campaign in about 20 years.

This is transparently absurd. Richard Berg knew it himself. He’s designed hundreds of war-games, focusing on everything from The Battle of Gettysburg to the Golden Age of Piracy, and The Campaign For North Africa was an outlier from the start. It was intended to be a collaborative mega-project for all of the wargaming experts employed by the storied, (and now defunct) imprint Simulations Publications Inc.

Initially, all Berg was responsible for was the map. Six months later, after the other designers had dropped out, SPI asked Berg if he was interested in finishing the game by himself. He was, and two years later he delivered history’s most infamous board game.

Berg has never completed a playthrough of The Campaign For North Africa. The game never received any of the compulsive testing required to iron-out inconsistencies and balance issues that are usually present in a freshly inked rulebook. Berg didn’t care. He never saw the point. “When I said ‘let’s publish this thing’ they said ‘but we’re still playtesting it! We don’t know if it’s balanced or not. It’s gonna take seven years to play!’ And I said ‘you know what, if someone tells you it’s unbalanced, tell them ‘we think it’s your fault, play it again.’”

The Campaign For North Africa arrived in the summer of 1979 and sold for $44 in a chunky, four-inch deep box. The game was never a massive commercial or critical success. It harbors a middling 5.8 on community tastemaker BoardGameGeek, and objectively speaking, the systems are exasperatingly finicky and require an eagle-eye for obscure rules and exceptions. In many ways, North Africa is simply a product of its time. The late ‘70s served as the commercial peak for wargaming, with dozens of new designs hitting store shelves every week. The Campaign wasn’t unique, as much as it was a standard archetype blown out to its extremes. Naturally, you do have to pay a premium price for used copies of the game on eBay, but that has more to do with the novelty of owning the “world’s longest board game” than anything else.

However, there is still a handful of players who regard Berg’s designas a triumph, rather than an extremely long-term gag. Geoff Phipps, a 54-year old software engineer living in Seattle, is one of them. Phipps never owned North Africa, but he did rent itfrom a local hobby shop after enjoying a slew of other, less-hefty Berg outings. He had no idea what he was getting into. The thing he remembers best is the way the fuel reserves worked. [Correction - 5:35pm, September 19: We initially misspelled Phipps’ last name as Phillips. We apologize for the error.]

“Every military division has a sheet of paper, and on it you’ve got a box for every battalion. It’ll tell you how many guns you have, but more interestingly, it’ll also list the fuel and water. Every game turn, three percent of the fuel evaporates, unless you’re the British before a certain date, because they used 50-gallon drums instead of jerry cans. So instead, seven percent of their fuel evaporates,” explains Phipps. “Every fucking turn you go around and make a pencil note of how much fuel you have. The pasta rule is funny, but this is what the game is about. Just doing tedious calculations all the time.”

As you may expect, Phipps did not finish The Campaign For North Africa. He and his friends played for exactly one session, resolving to get through the first day of the war for a taste of the combat systems and resource management, before quickly moved onto something that wasn’t going to demand of a decade of his time. His reasons were clear: the game is fastidious, non-intuitive, and it forces some seriously awkward fractional equations. But nearly 40 years later he still daydreams about the experience. “We did have a blast because some of the rules you’re not going to find in any other game,” says Phipps. “Just the fact that they cared about what kind of fuel tank the British had!”

As an amateur game-designer himself, Phipps plans on returning to North Africa after he’s retired to modernize some of the shortcomings in the design. The awkward flight combat module, which has caught the ire of many people in the game’s community, will be his first target, (Berg himself happily volunteers that the system “sucks.” The flight units are handled as individual planes and individual pilots, which is outstandingly fussy, even for wargame standards.) But with Phipps’ keen eye for revisions, perhaps someday he will still cross the Sahara.

Jake was enchanted in a similar way. He’s a 16-year old in Minnesota who obtained a copy of North Africa a few months ago by printing out giant PDF copies of the rulebook and map (he says it was the only way to avoid paying $400.) Like most people in the board game hobby, he learned of The Campaign For North Africa as a fable - that it was long, that it was rare, that it was occasionally silly. As he pored over the rulebook, his curiosity was piqued by the stringent regulations on the treatment of POWs, and how they could defect into their own militia and potentially plunge the campaign into an unwinnable state. Imagine that, the world’s longest board game ending with two losers.

Jake’s goal is to finish North Africa before he graduates high school. Last month he emailed the rulebooks to each of his recruited friends before their first session. Together they sat down in the family dining room to make their first moves. Jake has two years left before college, which is already cutting it close.

“For me, this is a passion. Some of my friends just like the idea of playing the world’s longest game, which is great, I don’t care,” he says. “But that’s not it for me. I love the structure, I love the complexity.”

This is the resolve of The Campaign For North Africa’s cult. They’re drawn to the game not for its cleverness or flair, but for its absurd, maximalist nature. Board games tend to prioritize a friendly communion with their players, simply because it’s difficult to sell copies of a design that nobody understands. But North Africa never got that memo. It is ornery and intentionally difficult, its commercial release feels like a grave miscalculation or an ultimate dare issued by a hysterical publisher. But its audacity touched a special few. Finally, the chance to have your courage and resilience challenged by a pile of cardboard.

Richard Berg has a pretty flat attitude towards the mystification of his most notorious work. As with every other product in his repertoire, the man built North Africa solely because someone was paying him, and he regards anyone earnestly attempting to conquer the full campaign to be either idealistic or foolish. “Has anyone completed the game? I think people have,” he says. “But the point with The Campaign For North Africa was that it was kinda fun to play for a couple weeks or a couple months. After that? Get a life.”

Berg sold his last copy of North Africa a handful of years back, because a “whole bunch of dollars seemed to be [a] more worthwhile thing to have.” He’s being flippant, but that’s not because he thinks the game was poorly conceived. “It did what it set out to do,” explains Berg. “It was supposed to be an intensive eurythmic manual, and I think it functioned at that level. Is this game something you should sit down and play? No, there are plenty of good Africa games, unless you really want to get down to that level.”

We’re in the midst of a tabletop renaissance. Global board game sales have boomed over the past few years, and a renewed interest in the hobby has seeped into coffee shops, video game publishers, and publications like ours. Despite that, the classic hexagonal historical war-game—the true bones of the industry—are a dying breed. This is the Catan generation: millennials weaned on the crisp, instinctual gameplay perfected by the German masters. Phipps has fond memories of the late-’70s “the golden age” of war-gaming - where publishers routinely tried to out-convolute each other with their designs, because surely, the more complex a game is, the grander it must be. “After that golden age the designs got better,” he says. “But at the time there’s this sense of excitement, everything is new and possible.”

Perhaps someday war-gaming will make a comeback, but in the meantime, there will always be the Campaign. The Italian water rations, the thousand-plus cardboard shards, the unrepeatable, era-specific panache to market and sell a 1,500 hour experience. It’s a blessing to be thrilled by evaporating gas, to finally find a board game that embraces your obsessiveness note for note. It’s all way too much. It is drunk and full of hubris. And yet, The Campaign For North Africa will seduce new players for the rest of time.

“You’ve got rising market power,” said Marshall Steinbaum, an economist at the Roosevelt Institute, a liberal think tank. “In general, that makes it hard for new businesses to compete with incumbents. Market power is the story that explains everything.”

That argument comes at a potent political moment. Populists on both the left and right have responded to growing public unease about the corporate giants that increasingly dominate their online and offline lives. Polling data from Gallup and other organizations shows a long-running decline in confidence in banks and other big businesses — a concern not likely to abate after high-profile data breaches at Equifax and other companies.

The start-up slump has far-reaching implications. Small businesses in general are often cited as an exemplar of economic dynamism. But it is start-ups — and particularly the small subset of companies that grow quickly — that are key drivers of job creation and innovation, and have historically been a ladder into the middle class for less-educated workers and immigrants.

Perhaps most significant, start-ups play a critical role in making the economy as a whole more productive, as they invent new products and approaches, forcing existing businesses to compete or fall by the wayside.

“Across the decades, young companies are really the heavy hitters and the consistent hitters in terms of job creation,” said Arnobio Morelix, an economist at the Kauffman Foundation, a nonprofit in Kansas City, Mo., that studies and promotes entrepreneurship.

The start-up decline might defy expectations in the age of Uber and “Shark Tank.” But however counterintuitive, the trend is backed by multiple data sources and numerous economic studies.

In 1980, according to the Census Bureau data, roughly one in eight companies had been founded in the past year; by 2015, that ratio had fallen to fewer than one in 12. The downward trend cuts across regions and industries and, at least since 2000, includes even the beating heart of American entrepreneurship, high tech.

Although the overall slump dates back more than 30 years, economists are most concerned about a more recent trend. In the 1980s and 1990s, the entrepreneurial slowdown was concentrated in sectors such as retail, where corner stores and regional brands were being subsumed by national chains. That trend, though often painful for local communities, wasn’t necessarily a drag on productivity more generally.

Since about 2000, however, the slowdown has spread to parts of the economy more often associated with high-growth entrepreneurship, including the technology sector. That decline has coincided with a period of weak productivity growth in the United States as a whole, a trend that has in turn been implicated in the patterns of fitful wage gains and sluggish economic growth since the recession. Recent research has suggested that the decline in entrepreneurship, and in other measures of business dynamism, is one cause of the prolonged stagnation in productivity.

“We’ve got lots of pieces now that say dynamism has gone down a lot since 2000,” said John Haltiwanger, a University of Maryland economist who has done much of the pioneering work in the field. “Start-ups have gone down a lot since 2000, especially in the high-tech sectors, and there are increasingly strong links to productivity.”

What is behind the decline in entrepreneurship is less clear. Economists and other experts have pointed to a range of possible explanations: The aging of the baby-boom generation has left fewer Americans in their prime business-starting years. The decline of community banks and the collapse of the market for home-equity loans may have made it harder for would-be entrepreneurs to get access to capital. Increased regulation, at both the state and federal levels, may be particularly burdensome for new businesses that lack well-staffed compliance departments. Those and other factors could well play a role, but none can fully explain the decline.

More recently, economists — especially but not exclusively on the left — have begun pointing the finger at big business, and in particular at the handful of companies that increasingly dominate many industries.

The share of employees working at large, medium and small companies in the United States.

Small (less than 100 employees)

The share of employees working at large, medium and small companies in the United States.

The evidence is largely circumstantial: The slump in entrepreneurship has coincided with a period of increasing concentration in nearly every major industry. Research from Mr. Haltiwanger and several co-authors has found that the most productive companies are growing more slowly than in the past, a hint that competitive pressures aren’t forcing companies to react as quickly to new innovations.

A recent working paper from economists at Princeton and University College London found that American companies are increasingly able to demand prices well above their costs — which according to standard economic theory would lead new companies to enter the market. Yet that isn’t happening.

“If we’re in an era of excessive profits, in competitive markets we would see record firm entry, but we see the opposite,” said Ian Hathaway, an economist who has studied the issue. That, Mr. Hathaway said, suggests that the market is not truly competitive — that existing companies have found ways to block competitors.

Experts also point to anecdotal examples that suggest that the rise of big businesses could be squelching competition. YouTube, Instagram and hundreds of lower-profile start-ups chose to sell out to industry heavyweights like Google and Facebook rather than try to take them on directly. The tech giants have likewise been accused of using the power of their platforms to favor their own offerings over those of competitors.

Most recently, Amazon openly called for a bidding war among cities for its second headquarters — hardly the kind of demand a new start-up could make. Mr. Morelix said the Amazon example was particularly striking.

“We’re saying that it’s O.K. that they shape how a city charges taxes?” Mr. Morelix said. “And what kind of regulations they have? That should be terrifying to anyone that wants a free market.”

In Washington, where for years politicians have praised small businesses while catering to big ones, issues of competition and entrepreneurship are increasingly drawing bipartisan attention. Several Republican presidential candidates referred to the start-up slump during last year’s primary campaign. Progressive Democrats such as Senators Elizabeth Warren of Massachusetts and Amy Klobuchar of Minnesota have pushed for stricter enforcement of antitrust rules. In a speech in March, Ms. Klobuchar explicitly tied the struggles of entrepreneurs to rising corporate concentration.

In July, entrepreneurs achieved a mark of political relevance: their own advocacy group. The newly formed Center for American Entrepreneurship will conduct research on the importance of new businesses to the economy and push for policies aimed at improving the start-up rate. Its founding president, John Dearie, comes from big business — he was most recently the acting head of the Financial Services Forum, which represents big financial institutions.

“Everybody loves entrepreneurship, but they’re not aware it’s in trouble,” Mr. Dearie said. “If new businesses are the engine of net new job creation, and if new businesses are the engine of innovation, and new business creation is at 30-year lows, that’s a national emergency.”

Paper wallets are a form of cold storage, meaning that the private key has never touched a computer with internet access. This is one of the most secure ways to store Bitcoins when done properly. You should never use a paper wallet you did not create yourself. For that reason, this is a tutorial to create a paper wallet in a secure fashion.

A repository of paper wallet generators can be found here. You can choose to use my repository directly if you wish, but I recommend going straight to the source. There are many options with different formats and templates, however I highly recommend BitcoinPaperWallet.com.

I am in no way affiliated with BitcoinPaperWallet.com, just a satisfied customer.

This is the most secure and well thought out design I have seen. The author has done a great job addressing the various attack vectors. His website is easy to use and provides all the relevant information needed. Furthermore, the generator is based on the popular and trusted bitaddress.org. This paper wallet is perfect for cold storage and archival, but it takes a little extra work and materials to secure it properly. Fortunately, everything required can be bought right from the website, and you can pay in Bitcoin!

Features:

• “Butterfly” design secures the private key
• Resistant to candling
• Supports BIP38 encryption
• Private key encoded in Wallet Import Format
• Designed to be printed in landscape mode, but works in portrait mode as well
• Double sided
• 2x wallets per sheet
• Notes section on the back
• Public key visible when closed

Additional Options:

*The official security stickers from bitcoinpaperwallet.com are serialized in pairs, meaning you get two of each number. This is best because each paper wallet requires two stickers, so the numbers match. The alternative link only provides one sticker for each number.

Dimensions:

• Landscape: 5.5in x 2.5in (Slightly smaller than a dollar bill)
• Portrait: 4in x 2in (Slightly bigger than a standard business card)
• QR Codes: 1in x 1in (About the size of a quarter)

For additional artworks, check out Liberty Paper Wallet (Github).

Paper wallets are extremely vulnerable to water. Consider laminating your paper wallet for extra protection. An alternative could be to vaccuum seal it using a Foodsaver Vaccuum Sealer. At the very least, you should keep the paper wallet in a ziplock bag.

Paper wallets are also extremely vulnerable to fire. However, there is not much you can do about this other than keeping multiple backups in different physical locations, such as a safety deposit box.

Remember, if you lose your paper wallet, or it is damaged, you lose all the coins stored at that address!

Paper wallets need to be created offline on a secure machine. For this tutorial, I will be using a Tails Live CD. Grab the ISO and burn it to a disk.

Save a copy of bitcoinpaperwallet onto a USB drive.

1. Shutdown your computer and boot from the CD.
2. When the Tails welcome box appears, select “More Options”.
3. Enter an administrator password of your choice and login.
4. Wait for tor to become ready.

For this tutorial, in order to get screenshots I am using a virtual machine. Do not use a virtual machine when doing this for real!

Before creating the paper wallet, you will want to ensure that your printer works correctly. You need a printer that is directly connected to your machine; don’t use a network printer (you shouldn’t be connected to any network). If possible, use a dumb printer, and try to ensure that your printer does not save a copy of printouts to internal memory. Use this time to get any drivers you need from the internet.

From the Applications menu, select System Tools -> Administration -> Printing.

Add your printer. Open up Tor Browser and ensure that you can print a webpage.

Time to disconnect from the internet. Unplug the ethernet cord, turn off any wireless cards or routers as necessary. Verify that you are not online.

1. Insert your usb drive
2. Copy the github zip file to the Tor Browser folder on the filesystem
3. Remove the usb drive
4. Unpack the zip file

Open up the html file in Tor Browser. Follow the instructions to generate your paper wallet. You have the option of using the built-in random number generator, or supplying your own random numbers using dice or cards. For maximum security, you should use dice or cards. I found that taking a deck of cards, shuffling it seven times, and picking the first 32 cards off the top worked well. Remember to shuffle the cards again after you are done! “Brain wallets” may seem convenient, however you need to have a very strong passphrase for this to be secure; it is better to use random numbers.

BIP38 Encryption?
There are pros and cons to encrypting your paper wallet. Encryption adds an extra layer of security by requiring a passphrase before being able to import the private key again, which is great if the paper wallet ever gets stolen. However if the passphrase is forgotten, the coins are lost forever. The passphrase is one more thing to remember / write down, which means it is one more thing to secure. In addition, some wallet software does not support BIP38, which may make reimporting difficult. Finally, it is important to realize that BIP38 encryption will NOT help if you chose an insecure passphrase for a brain wallet. In general, BIP38 encryption is recommended.

You will want to print two wallets by spinning the paper after each print. You will end up running the sheet of paper through the printer a total of 4 times.

Now is your chance to verify the paper wallets. Make sure you can scan the QR codes, make sure the private key and public key match, etc.

Shutdown your Tails environment. It will wipe your ram for you.

Cut out each design by following the lines on the front. Fold the wallets and apply the stickers.

If you have a laser printer, you will want to include a small 1in x 1in square of paper between the private key and the candling pattern when folding the paper. Later on, we will be laminating the paper wallet, and the heat can cause the toner on each side of the fold to fuse together.

Sign the back and write the sticker numbers. This prevents someone from simply replacing the stickers or entire wallet without your knowledge.

Put both paper wallets into a laminating pouch and run it through the laminator. Cut out each wallet, and store them in physically separate, secure locations.

You can scan the public key into Electrum or Mycelium as a watch-only wallet to keep track of your funds. Eventually you will want to spend the funds.

First, you need to delaminate the wallet. Cut a line along the edge closest to the paper where there is a tiny line of air, and peel away the laminate. A technique that worked for me was to cut along the “Private Key / Withdraw” line, then slide my knife underneath each sticker. Unfold the flap to access the private key. As you can see, it is still safe and legible, even after the lamination process.

Now comes the important part. The funds must be “swept” into an electronic wallet. You must take all the funds in one shot; do not attempt to partially spend the funds in a paper wallet. This is due to how Bitcoin Change works.

Change?
When spending Bitcoins, ALL coins from that address are moved to new addresses. The destination address will get the desired amount, and any remaining amount will be sent to a “change” address. If no “change” address is specified, the remaining amount will go to the miner that solves the block. Normally, an electronic wallet manages this for you behind the scenes, however when using a paper wallet directly, you will not have this control. This is why the entire balance of a paper wallet should be “swept” into an electronic wallet before spending.
Read more about change here

Once the paper wallet has been swept into the electronic wallet, it should be destroyed and never used again. Shred it or burn it.

“Powerlessness and the Politics of Blame”

Martha C. Nussbaum^₁

The Jefferson Lecture in the Humanities  

The John F. Kennedy Center for the Performing Arts  

Monday, May 1, 2017

_{(This lecture contains material from THE MONARCHY OF FEAR: A Philosopher Looks at Our Political Crisis, by Martha C. Nussbaum, to be published in 2018 by Simon & Schuster.)}

At the end of Aeschylus’ Oresteia, two transformations take place in the city of Athens. One is famous, the other often neglected. In the famous transformation, Athena introduces legal institutions to replace and terminate the cycle of blood vengeance. Setting up a court of law with established procedures of evidence and argument, and a jury selected by lot from the citizen body of Athens, she announces that blood guilt will now be settled by law, rather than by the Furies, ancient goddesses of revenge.  But the Furies are not simply dismissed. Instead, Athena persuades them to join the city, giving them a place of honor beneath the earth, in recognition of their importance for the health of the city. 

Typically, Athena’s move is understood to be a recognition that the legal system must incorporate and honor the retributive passions. These passions themselves remain unchanged; they simply have a new house built around them. The Furies agree to accept the constraints of law, but they retain an unchanged nature, dark and vindictive.

That reading, however, ignores the second transformation, a transformation in the character of the Furies themselves.  As the drama begins, the Furies are described as repulsive and horrifying. They are said to be black, disgusting; their eyes drip a hideous liquid. Apollo even says they vomit up clots of blood that they have ingested from their prey. They belong, he says, in some barbarian tyranny where cruelty reigns.  

Nor, when they awaken, do the Furies give the lie to these grim descriptions. As Clytemnestra’s ghost calls them, they do not speak, but simply make animal noises, moaning and whining.  When they do begin to speak, their only words are “get him get him get him get him,” as close to a predator’s hunting cry as the genre allows. As Clytemnestra says: “In your dream you pursue your prey, and you bark like a hunting dog hot on the trail of blood.” If the Furies are later given poetic speech, as the genre demands, we are never to forget this initial characterization.

What Aeschylus has done is to depict unbridled resentment. It is obsessive, destructive, existing only to inflict pain and ill. (As the distinguished 18th c. philosopher Bishop Butler observes, “No other principle, or passion, hath for its end the misery of our fellow creatures.”) Apollo’s idea is that this rabid breed belongs somewhere else, surely not in a law-abiding democracy.

Unchanged, these Furies could not be at the foundation of a legal system in a society committed to the rule of law. You don’t put wild dogs in a cage and come out with justice. But the Furies do not make the transition to democracy unchanged. Until quite late in the drama, they are still their bestial selves, threatening to disgorge their venom on the land. Then, however, Athena persuades them to alter themselves so as to join her enterprise. “Lull to repose the bitter force of your black wave of anger,” she tells them. But of course that means a virtual change of identity, so bound up are they with anger’s obsessive force. She offers them incentives to join the democracy: a place of honor, reverence from the citizens—but only if they adopt a new range of sentiments, substituting future-directed benevolence for retribution. Perhaps most fundamental of all, they must listen to the voice of persuasion. They accept her offer, and express themselves “with gentle-tempered intent.” Each, they declare, should give generously to each, in a “mindset of common love.” Not surprisingly, they are transformed physically in related ways. They apparently assume an erect posture for the procession that concludes the drama, and they receive crimson robes from a group of citizen escorts. They have become Athenians, rather than beasts. Their very name is changed: they are now the Kindly Ones (Eumenides), not the Furies.

This second transformation is just as significant as the first one, indeed crucial to the success of the first one. Aeschylus shows that a democratic legal order can’t just put a cage around retribution, it must fundamentally transform it from something hardly human, obsessive, bloodthirsty, to something human, accepting of reasons, something that protects life rather than threatening it. The Furies are still needed, because this is an imperfect world and there are always crimes to be dealt with. But they are not wanted or needed in their original form. They must become instruments of justice and human welfare. The city is liberated from the scourge of vindictive anger, which produces civil strife. In its place, the city gets forward-looking justice.  

Like modern democracies, the ancient Greek democracy had an anger problem. Read the historians, and you will see some things that are not remote: individuals litigating obsessively against people they blame for having wronged them; groups blaming other groups for their lack of power; citizens blaming prominent politicians and other elites for selling out the dearest values of the democracy; other groups blaming foreign visitors, or even women, for their own political and personal woes.  

The anger that the Greeks—and, later, the Romans—knew all too well, was an anger full of fear at one’s own human vulnerability. The Roman philosopher Lucretius even says that all political anger is an outgrowth of fear—of the terror of each human infant, who comes into the world helpless, and, unlike all other animals, can do nothing on its own to get what it needs to stay alive. Lucretius sees that as life goes on, vulnerability continues or even increases, since the awareness of death hits us hard at some point, making us realize that we are helpless with respect to the most important thing of all. This fear, he says, makes everything worse, leading to political ills to which we’ll return. For now, however, let’s focus on anger. 

The Greeks and Romans saw a lot of anger around them. But as classical scholar William Harris shows in his fine book Restraining Rage, they did not embrace or valorize anger. They did not define manliness in terms of anger, and indeed, as with those Furies, tended to impute it to women, whom they saw as lacking rationality. However much they felt and expressed anger, they waged a cultural struggle against it, seeing it as destructive of human well-being and democratic institutions. The first word of Homer’s Iliad is “anger”—the anger of Achilles that “brought thousandfold pains upon the Achaeans.” And the Iliad’s hopeful ending requires Achilles to give up his anger and to be reconciled with his enemy Priam, as both acknowledge the frailty of human life. 

I believe the Greeks and Romans are right: anger is a poison to democratic politics, and it is all the worse when fueled by a lurking fear and a sense of helplessness. As a philosopher I have been working on these ideas for some time, first in a 2016 book called Anger and Forgiveness, and now in a book in progress called The Monarchy of Fear, investigating the relationship between anger and fear. In my work, I draw not only on the Greeks and Romans, but also on some recent figures, as I shall tonight. I conclude that we should resist anger in ourselves and inhibit its role in our political culture. 

That idea, however, is radical and evokes strong opposition. For anger, with all its ugliness, is a popular emotion. Many people think that it is impossible to care for justice without anger at injustice, and that anger should be encouraged as part of a transformative process. Many also believe that it is impossible for individuals to stand up for their own self-respect without anger, that someone who reacts to wrongs and insults without anger is spineless and downtrodden. Nor are these ideas confined to the sphere of personal relations. The most popular position in the sphere of criminal justice today is retributivism, the view that the law ought to punish aggressors in a manner that embodies the spirit of justified anger. And it is also very widely believed that successful challenges against great injustice need anger to make progress.

Still, we may persist in our Aeschylean skepticism, remembering that recent years have seen three noble and successful freedom movements conducted in a spirit of non-anger: those of Mohandas Gandhi, Martin Luther King, Jr., and Nelson Mandela—surely people who stood up for their self-respect and that of others, and who did not acquiesce in injustice. 

I’ll now argue that a philosophical analysis of anger can help us support these philosophies of non-anger, showing why anger is fatally flawed from a normative viewpoint—sometimes incoherent, sometimes based on bad values, and especially poisonous when people use it to deflect attention from real problems that they feel powerless to solve.  Anger pollutes democratic politics and is of dubious value in both life and the law. I'll present my general view, and then show its relevance to thinking well about the struggle for political justice, taking our own ongoing struggle for racial justice as my example. And I’ll end by showing why these arguments make it urgent for us to learn from literature and philosophy, keeping the humanities strong in our society.

The Roots of Anger: Rage, Ideas of Unfairness  

Let’s now return briefly to that baby, following Lucretius’ brilliant analysis. Babies at birth don’t have anger as such, because anger requires causal thinking: someone did something bad to me. Fairly soon, however, that idea creeps in: those caretakers are not giving me what I desperately need. They did this to me. It’s because of them that I am cold, wet, and hungry. Experiences of being fed, held, and clothed quickly lead to expectations, expectations to demands. Instinctual self-love makes us value our own survival and comfort. But the self is threatened by others, when they don’t do what we want and expect. Psychoanalyst Melanie Klein refers to this emotional reaction in infants as “persecutory anxiety,” since it is indeed fear, but coupled with an idea of a vague threat coming from outside. I would prefer to call it fear-anger or even fear-blame. 

If we were not helpless, we would just go get what we need. But since we are initially helpless, we have to rely on others. They don’t always give us what we need, and then we lash out, blaming them. Blame gives us a strategy: now I’ll enforce my will by raging and making noise. But it also expresses an underlying picture of the world: the world ought to give us what we demand. When people don’t do that, they are bad.  

Protest and blame are positive, in a sense: they construct an orderly purposive world in which I am an agent, making demands. My life is valuable, things ought to be arranged so that I am happy and my needs are met. That hasn’t happened, so someone must be blamed. But retributive anger all too often infects the thought of blame, and often even of punishment: the people we blame ought to suffer for what they have done. Psychologist Paul Bloom has shown that retributive thinking appears very early in the lives of infants, even before they begin to use language. Infants are delighted when they see the “bad person”—a puppet who has snatched something from another puppet—beaten with a stick. Bloom calls this an early sense of justice. I prefer to call it the internal Furies that inhabit us all, and that are not securely linked to real justice. The infants’ idea looks like a version of the lex talionis: an eye for an eye, pain for pain.  It’s not hard to imagine that the crude idea of proportional payback has an early, perhaps an evolutionary, origin. It is a leap to call this an idea of justice, and I think we should not make this leap.

Defining Anger

Let’s now fast forward to human adulthood. People now experience and express not just primitive anger, but full-fledged anger. But what is anger? Philosophers are fond of definitions, which are useful to clear our heads, in this case helping us separate the potentially promising parts of anger from those that lead to nothing but trouble. And, back to the Greeks, let’s talk about Aristotle’s definition, since more or less all the definitions of anger in the Western philosophical tradition are modeled on it. (Those in Hindu traditions are very similar.)

According to Aristotle, anger is a response to a significant damage to something or someone one cares about, and a damage that the angry person believes to have been wrongfully inflicted. Aristotle adds that although anger is painful, it also contains within itself a pleasant hope for payback or retribution. So: significant damage, pertaining to one’s own values or circle of cares, and wrongfulness. Those two elements seem both true and uncontroversial, and they have been validated by modern psychological studies. Those parts of anger can go wrong in specific and local ways: we might be wrong about who did the bad thing, or how significant it was, or whether it was done wrongfully (rather than accidentally). But they are often on target.

More controversial, certainly, is the idea that the angry person wants some type of retribution, and that this is a conceptual part of what anger is. All the Western philosophers who talk about anger include this wish as a conceptual element in anger. Still, we need to pause, since it is not obvious. We should understand that the wish for retribution can be a very subtle wish: the angry person doesn’t need to wish to take revenge herself. She may simply want the law to punish the wrongdoer; or even some type of divine justice. Or she may more subtly simply want the wrongdoer’s life to go badly in future – hoping, for example, that the second marriage of her betraying spouse is a dismal failure. I think if we understand the wish in this broad way, Aristotle is right: anger does contain a sort of strike-back tendency, and that is what differentiates it from compassionate grieving.  Contemporary psychologists who study anger empirically agree with Aristotle in seeing this double movement in it, from pain to hope.   

We should understand, however, that the two parts of anger can come apart. We can feel outrage at the wrongfulness of an act or an unjust state of affairs, without wanting payback for the wrong done to us. I’ll be arguing that the outrage part is personally and socially valuable when our beliefs are correct: We need to recognize wrongful acts and protest them, expressing our concern for the violation of an important norm. And there is one species of anger, I believe, that is free of the retributive wish: its entire content is “How outrageous that is. Something must be done about that.” I call this “TransitionAnger,” because it expresses a protest, but faces forward: it gets to work finding solutions rather than dwelling on the infliction of retrospective pain.

Take parents and children. Parents often feel that children have acted wrongfully, and they are outraged. They want to protest the wrong, and somehow to hold the child accountable. But they usually avoid retributive payback. They rarely think (today at least), “now you have to suffer for what you have done,” as if that by itself was a fitting response. Instead, they ask themselves what sort of reaction will produce future improvement in the child. Usually this will not be a painful payback, and it certainly won’t obey the lex talionis, “an eye for an eye.” If their child hits a playmate, parents do not hit their child as if that were “what you deserve.” Instead, they choose strategies that are firm enough to get the child’s attention, and that express clearly that and how what the child did was wrong. And they give positive suggestions for the future, how to do things differently. So, loving parents typically have the outrage part of anger without the payback part—where their children are concerned. This will be a clue to my positive proposal for democratic society.

Retributive wishes, however, are a deep part of human nature, fostered by some parts of the major religions and by many societal cultures, although they have been denounced by religious and social radicals from Jesus and the Buddha to Mohandas Gandhi. They may have served us well in a pre-social condition, deterring aggression. But the idea that pain is made good or assuaged by pain, though extremely widespread, is a deceptive fiction. Killing the killer does not restore the dead to life, although the demand for capital punishment is endorsed by many families of victims as if it did somehow set things to rights. Pain for pain is an easy idea. But it is a false lure, creating more pain instead of rectifying the problem. As Gandhi said, “An eye for an eye makes the whole world blind.”

This wish for payback arises in all sorts of situations. Take divorce. Betrayed spouses often feel entitled to seek punitive divorce settlements and child custody arrangements, as if that somehow were their due, and as if punitive payback somehow restored the balance of power or rescued their damaged dignity. But in real life the function of payback is usually far less benign. Two people become locked in a struggle for pain, focused on the past, and often inflicting great collateral damage on children and on friends and family. In the end, the betrayer may get “his comeuppance,” but what does that achieve? Typically it does not improve the litigant’s life going forward. By focusing obsessively on the past she becomes closed to new possibilities, and she often becomes bitter and unpleasant. Retaliation is ugly, as Aeschylus shows us in his portrait of the Furies. What the payback-seeker wants is future happiness and self-respect. Payback by itself never achieves that, and it usually makes the world a lot worse for all.

But wait a minute. We all agree that wrongful acts, if they are serious enough, should be punished, and punishment is typically painful. Yes, we should agree that punishment is often useful: but why and how? We might see punishment in a retributive spirit, as payback for what has already happened. That is the attitude I have been criticizing, and it does great social harm, leading to a gruesome pile-on-the-misery strategy, as if it really compensated for the damages of crime.  But there’s a better attitude, more like that of the good parent in my example: We might try to look to the future and produce a better society, using punishment to express the value we attach to human life and safety, to deter other people from committing that crime, and, we hope, deterring that individual from committing another crime, or at least incapacitating him.   

If we think this way, however, trying to improve the future, we probably will have a lot of other thoughts before we get to punishment. Like that good parent, we will think that people don’t do wrong nearly as often, if they are basically loved and respected, if they have enough to eat, if they get a decent education, if they are healthy, and if they foresee a future of opportunity. So thinking about crime will lead us in the direction of designing a society in which people have fewer incentives to commit crime. When they do, despite our best efforts, we take that seriously, for the sake of the future.

There is one more part to Aristotle’s definition. He says that anger is always a response not to any old damage, but only to the type he calls a “down-ranking.” This does not seem to be true all the time. I can get angry at wrongs done to others, without thinking of them as a “down-ranking” of myself. Later philosophers hold on to the other parts of Aristotle’s definition but drop this restriction: anger can be a response to any wrongful act, not just a status injury. Still, let’s hold on to Aristotle’s idea, for it does cover surprisingly many cases of anger, as empirical researchers emphasize.

The status idea is important because it is the one case, I believe, where payback gives you what you want. If what you are focused on is not the murder, or theft, or rape itself, but only the way it has affected your relative status in the world, then by pushing the wrongdoer relatively lower, you really do push yourself relatively higher. And if relative status is all you care about, you needn’t be worried that the underlying harms caused by the wrongful act (murder, rape, theft) have not been made good. If you’re thinking only about relative status, then, payback sort of makes sense. Many people do think this way, and that may help explain why payback is so popular and why people do not quickly conclude that it is an empty diversion from the task of fixing the future. 

What is wrong with the status idea? Focus on relative status was common in ancient Greece: indeed it explains Achilles’ anger when Agamemnon insults him by taking “his” woman away. Focus on status was common, too, at the founding of the United States, as Lin-Manuel Miranda’s brilliant Hamilton reminds us all. Elaborate codes of honor and status led, indeed, to constant status-anxiety and to many duels responding to purported insults. What’s wrong with the obsession with status is that life is not all about reputation, it is about more substantial things: love, justice, work, family. We all know people today who are obsessed with what other people think of them, who constantly scan the Internet to see who has been insulting them. Social media may encourage this obsession, as people diss each other, count the number of “likes” some post of theirs has garnered, and so forth. As we live more and more in the eyes of others, more and more of our lives come in for rating, up or down.  But isn’t this obsession with status a sign of insecurity? And doesn’t it increase insecurity, since that person who scans the world for signs of disfavor is certain to find some? Equally important, isn’t the obsession with status a diversion from more important values? Achilles had to learn how bad it was to destroy thousands of people on account of one insult; Aaron Burr never learned much, it appears, but his example shows us what we forfeit when we become obsessed with being “in the room where it happens.”

Notice that obsession with relative status is different from a focus on human dignity or self-respect, since dignity belongs to everyone, and people are equal in dignity (at least this is how we ought to think and usually do think), so dignity does not establish a hierarchy, and nobody would be tempted to suppose that inflicting humiliation on someone else would affect my human dignity. Dignity, unlike reputation, is equal and inalienable. 

Three Errors in Anger

We’re now ready to see three ways anger can lead us astray.

1. The Obvious Errors. Anger can be misguided, and guide us badly, if it is based on wrong information about who did what to whom, about whether the bad act was really done wrongfully (with some sort of bad intent) rather than just by accident, and also if it is based on a confused sense of importance. Aristotle mentions people who get angry when someone forgets their name, and this familiar example is a case of overestimating the importance of what the person did. (Probably also a case of getting intention wrong.) Since we’re often hasty when we are angry, these errors occur often.
2. The Status Error. We also go wrong, I claim, if we think relative status is hugely important and focus on that to the exclusion of other things. This error is really a case of mistaking the importance of a particular value, but since it is so common and such a major source of anger, we have to single it out and give it a separate number.
3. The Payback Error. Finally, we very often go wrong when we permit deeply ingrained retributive thoughts to take over, making us think that pain wipes out pain, death murder, and so forth. In short, when we think that inflicting pain in the present fixes the past. We go wrong because that thought is a kind of irrational magical thinking, and because it distracts us from the future, which we can change, and often should. 

The Fourth Error in Anger: Helplessness and the “Just World”

All these errors are common, not least in the political life. We get hold of the wrong story about who did what, or we blame individuals and groups for a large systemic problem that they didn’t cause. We overestimate trivial wrongs and also, sometimes, underestimate important ones. We obsess about our own relative status (or that of our group). We think that payback will solve the problems created by the original offense, even though it does not.  

But there is more. We impute blame, often, even when there is no blame to be apportioned. The world is full of accidents. Sometimes a disaster is just a disaster. Sometimes illness and hardship are just illness and hardship. The medical profession can’t keep us completely safe from disease and death, and the wisest and most just social policies will not prevent economic woes arising from natural disasters. But in our monarchical way we expect the world to be made for our service. It gratifies our ego, and is in a deep sense comforting, to think that any bad event is someone’s fault. The act of pinning blame and pursuing the “bad guy” is deeply consoling. It makes us feel control rather than helplessness.

Psychologists have done a lot of research on people’s instinctual views of the way the world works, and they find that people have a deep-rooted need to believe that the world is just. One aspect of this “just world hypothesis” is the tendency to believe that people who are badly off cause their own misery by laziness or bad conduct. But another related aspect of this “just world” belief is the need to believe that when we encounter loss and adversity it isn’t just loss, it is someone’s wrongdoing, and that we can somehow recoup our loss by punishing the “bad guy.” 

Your parent dies in the hospital. It is very human to believe that “the doctors did it,” and to deflect one’s grief into malpractice litigation. Your marriage falls apart. Often there is fault somewhere, but sometimes it can’t be easily identified. Things do just fall apart. Still, it is human to fix blame on the “bad” spouse and try to crush that person with litigation. It makes life look more intelligible, the universe more just.

Economic woes are sometimes caused by an identifiable person or persons’ malfeasance, and sometimes by clearly stupid or unfair policies; but more often their cause is obscure and uncertain. We feel bad saying that: it makes the world look messy and ungovernable. So why not pin the blame—as the Greeks did—on groups that are easy to demonize: in place of their rhetorical category of “barbarians,” we might focus blame on immigrants, or women entering the workforce, or bankers or rich people. The Salem witch trials were once thought to be the result of group hysteria among adolescent girls. But now we know that a preponderant number of the witch-blamers were young men entering adulthood, afflicted by the usual woes of an insecure colony in a new world: economic uncertainty, a harsh climate, political instability. How easy, then, to blame the whole thing on witches, usually elderly unpopular women, who can easily be targeted and whose death brings a temporary satisfaction to the mind.

Our earliest fairy tales have this structure. Hansel and Gretel wander into the woods to search for food. The problem is hunger, compounded by the fact that their parents have to work at menial jobs and have no leisure to care for the children. But the story tells us that these very real problems are unreal, and that the real problem is a witch who lives in the woods, and turns little children into gingerbread.  Red Riding Hood goes to visit her grandmother, walking a long distance alone. The real problem in this story is aging and lack of care. The family lives far away, and grandmother is not doing well. But quickly the story deflects our attention. The problem is not this difficult human problem at all, requiring a structural solution. It is a single wolf who has broken into grandmother’s house. In both stories, when the ugly villain is killed the world is just fine. Our love of an orderly universe makes these simple fictional solutions tempting. It’s hard to wrap our minds around complicated truths, and it’s far easier to incinerate the witch than to live with hope in a world that is not made for human delectation.

Anger, Child of Fear

Anger is a distinct emotion with distinctive thoughts. It looks manly and important, not at all timorous. Nonetheless, it is the offspring of fear. How so?

First, if we were not plagued by great vulnerability, we would probably never get angry. Lucretius imagined the gods as beings who are perfect and complete, beyond our world, and he said, “They are not enslaved by gratitude, nor are they tainted by anger.” If anger is a response to a significant damage inflicted by someone else on you or someone or something you care about, then a person who is complete, who cannot be damaged, has no room for anger. (Judeo-Christian pictures of divine anger imagine God as loving humans and as deeply vulnerable to their misdeeds.)

Some moral reformers have urged us to become like Lucretius’ gods. The Greek Stoics thought that we should learn not to care at all about the “goods of fortune,” that is, anything that can be damaged by anything outside our own control. Then we would lose fear, and in the bargain we’d lose anger. Philosopher Richard Sorabji has shown that Gandhi’s views were very close to those of the Stoics.  

The problem, however, is that in losing fear we also lose love. The basis of both is a strong attachment to someone or something outside our control. There is nothing that makes us more vulnerable than loving other people, or loving a country. So much can go wrong. In one half-year the Roman philosopher and politician Cicero lost the two things he loved most in the world, when his daughter Tullia died in childbirth and the Roman Republic collapsed into tyranny. Even though his friends thought his grief excessive and urged him to be a proper Stoic, he told his best friend Atticus that he could not stop grieving, and, what’s more, he didn’t think he ought to. Taking the measure of love fully means suffering. So the solution that wipes out both fear and anger with one stroke is not one we should accept. Keeping love means keeping fear. 

And though that does not necessarily mean keeping anger, it makes it a lot harder to win the struggle with anger. Fear is not only a necessary precondition for anger, it is also a poison to anger, feeding the four errors. When we are afraid, we jump to conclusions, lashing out before we have thought carefully about the who and the how.  When problems are complex and their causes poorly understood, as economic problems tend to be, fear often leads us to pin blame on individuals or groups, conducting witch-hunts rather than pausing to figure things out. 

Fear also feeds obsession with relative status: when people feel bigger than others, they think they can’t be destroyed. But when people protect their vulnerable egos by thoughts of status, they can easily be goaded into anger, since the world is full of insults and slights. Indeed, Lucretius traces all status-competition to fear, arguing that it is a way of soothing ourselves: by putting others down we make ourselves feel powerful.   

And fear also feeds the focus on payback, since vulnerable people think that getting back at wrongdoers, even obliterating them, is a way of reestablishing lost control and dignity. Lucretius even traces all wars to fear: Feeling insecure, we rage against what threatens us, and seek to obliterate it. He omits the obvious possibility that war may be caused by a reasonable reaction to a genuine threat to our safety and our values. So we should not accept his analysis fully. I am no pacifist, nor were my primary heroes of non-anger, Martin Luther King, Jr. and Nelson Mandela. Gandhi, I think, made a large mistake by endorsing total pacifism. But even just wars, such as I believe the Second World War to have been, are often marred by zeal for the blood of the aggressor, and one could certainly argue that episodes such as the bombing of Dresden were motivated by payback rather than sound policy. Great leaders understand that we need to retain and fortify the spirit of determined protest against wrongdoing, without comforting ourselves with retributive thinking. The brilliant speech in which Winston Churchill said, “I have nothing to offer but blood, toil, tears and sweat,” refers to danger, to struggle, and to a willingness to accept great pain in order to preserve democratic values. It is conspicuous for its utter lack of retributivism. Churchill does NOT say that paying back the Nazis will remove the threat to freedom. Freedom is beautiful, and we must be prepared to suffer for it, but we must focus on defending what we love rather than “disgorging [our] venom on the land,” as Aeschylus’ Furies put it. Churchill’s speech is of a piece with the best Allied aims to rebuild Germany post-war, and we can now see the wisdom of that course, as Germany is among our most valuable allies.

Finally, helplessness, and the fear that goes with it, lead to the reflex in which we pin blame on someone in order to feel less buffeted by fortune and more in control. Even a long and difficult fight (a protracted malpractice lawsuit, divorce litigation lasting many years) is often psychologically preferable to accepting loss. 

Protest without Payback

What’s the alternative? We can keep the spirit of determined protest against injustice while letting go of the empty fantasy of payback. This forward-looking strategy includes protesting wrongdoing when it occurs, but not imputing wrongdoing where there is, instead, the murky thicket of the global economy to manage, outsourcing and automation to reconcile with our citizens’ welfare. Never seizing hold of blame as a substitute for a feeling of powerlessness, but also not yielding to despair. Even when we are confident in imputing wrongdoing to an individual or a group, we can still firmly refuse payback, but look to the future with hope, choosing strategies designed to make things better rather than to inflict the maximum pain. 

To conclude, I want to study just one example of protest without payback: the ideas of Martin Luther King, Jr., who contributed so much to our society’s ongoing struggle with racism and its search for justice. King always said that anger had a limited usefulness, in that it brought people to his protest movement, rather than sitting in despair. But once they got there, the anger had to be “purified” and “channelized.” What he means is that people must give up the payback wish and yet keep the spirit of justified protest. Instead of retribution, they need hope, and faith in the possibility of justice. In an essay of 1959, he says that the struggle for integration will continue to encounter obstacles, and that these obstacles can be met in two very different ways:

One is the development of a wholesome social organization to resist with effective, firm measures any efforts to impede progress. The other is a confused, anger-motivated drive to strike back violently, to inflict damage. Primarily, it seeks to cause injury to retaliate for wrongful suffering. . . . It is punitive—not radical or constructive. 

King, of course, was characterizing not just a deep-seated human tendency, but the actual ideas and sentiments of Malcolm X, as he understood them.  

King insisted constantly that his approach did not mean acquiescence in injustice: there is still an urgent demand, there is still a protest against unjust conditions, in which the protester takes great risks with his or her body, in what King called “direct action.” Still, the protester’s focus must turn to the future that all must work to create together, with hope and faith in the possibility of justice.

King, in short, favors and exemplifies what I’ve called Transition-Anger: the protest part of anger without the payback. To see this better, let’s study the sequence of emotions in his “I have a dream” speech. King begins, indeed, with what looks like a summons to anger: he points to the wrongful injuries of racism, which have failed to fulfill the nation’s implicit promises of equality. One hundred years after the Emancipation Proclamation, “the life of the Negro is still sadly crippled by the manacles of segregation and the chains of discrimination.” 

The next move King makes is significant: for instead of demonizing white Americans, he calmly compares them to people who have defaulted on a financial obligation: “America has given the Negro people a bad check, a check which has come back marked ‘insufficient funds.’” This begins the shift to what I’ve called Transition-Anger: for it makes us think ahead in non-retributive ways. The essential question is not how whites can be humiliated, but how can this debt be paid, and in the financial metaphor the thought of destroying the debtor is not likely to be central. 

The future now takes over, as King focuses on a time in which all may join together in pursuing justice and honoring obligations: “But we refuse to believe that the bank of justice is bankrupt. We refuse to believe that there are insufficient funds in the great vaults of opportunity of this nation.” No mention, again, of torment or payback, only of determination to ensure the protection of civil rights at last. King reminds his audience that the moment is urgent, and that there is a danger of rage spilling over: but he repudiates that behavior in advance. “In the process of gaining our rightful place, we must not be guilty of wrongful deeds. Let us not seek to satisfy our thirst for freedom by drinking from the cup of bitterness and hatred. . . . Again and again, we must rise to the majestic heights of meeting physical force with soul force.”

So, the “payback” is reconceived as the vindication of civil rights, a process that unites black and white in a quest for freedom and justice. Everyone benefits: as many white people already recognize, “their freedom is inextricably bound to our freedom.”  

King next repudiates a despair that could lead to the abandonment of effort. It is at this point that the most famous section of the speech, “I have a dream,” takes flight. And of course, this dream is one not of retributive punishment but of equality, liberty, and brotherhood. In pointed terms, King invites the African-American members of his audience to imagine brotherhood even with their former tormentors:

I have a dream that one day on the red hills of Georgia, the sons of former slaves and the sons of former slave owners will be able to sit down together at the table of brotherhood.

I have a dream that one day even the state of Mississippi, a state sweltering with the heat of injustice, sweltering with the heat of oppression, will be transformed into an oasis of freedom and justice. . . .

I have a dream that one day, down in Alabama, with its vicious racists, with its governor having his lips dripping with the words “interposition” and “nullification” – one day right there in Alabama little black boys and black girls will be able to join hands with little white boys and white girls as sisters and brothers.

There is indeed outrage in this speech, and the outrage summons up a vision of rectification, which might easily take a retributive form. But King gets busy right away reshaping retributivism into work and hope. For how, sanely and really, could injustice be made good by retributive payback? The oppressor’s pain and lowering do not make the afflicted free. Only an intelligent and imaginative effort toward justice can do that.  

It might seem strange to compare King to Aeschylus, though it’s really not strange at all, given King’s vast learning in literature and philosophy. He’s basically saying the same thing: democracy must give up the empty and destructive thought of payback and move toward a future of legal justice and human well-being. King’s opponents portrayed his stance as weak. Malcolm X said sardonically that it was like coffee that has had so much milk poured into it that it has turned white and cold, and doesn’t even taste like coffee. But that was wrong. King’s stance is strong, not weak. He resists one of the most powerful of human impulses, the retributive impulse, for the sake of the future. One of the trickiest problems in politics is to persist in a determined search for solutions, without letting fear deflect us onto the track of anger’s errors. The idea that Aeschylus and King share is that democratic citizens should face with courage the problems and, yes, the outrageous injustices that we encounter in political and social life. Lashing out in anger and fear does not solve the problem; instead, it leads, as it did in both Athens and Rome, to a spiral of retributive violence.

Lucretius tells a grim tale of human anger and fear gone wild. He imagines a world not unlike his own, in which insecurity leads to acts of aggression, which do not quiet insecurity. (At the time when he wrote, the Roman Republic was imploding, and insecurity, mounting everywhere, would shortly lead to tyranny.) In an effort to quiet fear, he imagines, people get more and more aggressive—until they think up a new way to inflict maximum damage on their enemies: putting wild beasts to work in the military.  

They even tried out bulls in the service of war.

They practiced letting wild boars loose against their enemies.

They even used fierce lions as an advance guard, equipped with a special force of armed and ferocious trainers To hold them in check and keep them in harness.

It was no use.

The lions, hot with blood, broke ranks wildly.

Trampled the troops, tossing their manes. 

In a poetic tour de force Lucretius now imagines the carnage the animals unleash. Then he pulls back. Did this really happen?, he says. Maybe it happened in some other world out in space. And what, he says, did those fictional people want to accomplish? To inflict great pain on the enemy—even if it meant that they would perish themselves! 

Lucretius’ point is that our retributive emotions are those wild beasts. People may think anger powerful, but it always gets out of hand and turns back on us. And, yet worse, half the time people don’t care. They’re so deeply sunk in payback fantasies that they’d prefer to accomplish nothing, so long as they make those people suffer.  His grim science-fiction fantasy reminds us that we’ll always defeat ourselves so long as we let ourselves be governed by fear, anger, and the politics of blame.

There is a better alternative. Aeschylus knew it, and King both knew and lived it. Making a future of justice and well-being is hard. It requires self-examination, personal risk, searching critical arguments, and uncertain initiatives to make common cause with opponents—in a spirit of hope and what we could call rational faith. It’s a difficult goal, but it is that goal that I am recommending, for both individuals and institutions.

My argument has drawn on three prominent fields in the Humanities. For evocative images that illuminate political problems and the path to political goals, I have turned to literature. In order to show how such goals have already been pursued in our own place and time, I have turned to history. But above all, since I am a philosopher, I have turned to philosophy for critical analysis and structured arguments about justice and well-being. Philosophy does not compel, or threaten, or mock. It doesn’t make bare assertions, but, instead, sets up a structure of thought in which a conclusion follows from premises the listener is free to dispute. In that way it invites dialogue, and respects the listener. Unlike the over-confident politicians that Socrates questioned (Euthyphro, Critias, Meletus), the philosophical speaker is humble and exposed: his or her position is transparent and thus vulnerable to criticism. (His or her, since Socrates said he’d like to question women, if only in the afterlife, and Plato actually taught women in his school!) 

OK: You’ve seen the steps in my argument: now, any one of you may jump in and differ with me, as Socrates recommended. That’s how democracy learns, and makes progress. I believe democracy urgently needs all three of the humanistic fields I’ve mentioned, which complement one another. All three offer practice in skills of mind and heart that are essential, especially in our current time of fear, resentment, and division, if our conversation with one another is to be respectful deliberation. But above all, to solve the political problem of anger, democracy needs to learn from, and practice, philosophical dialogue, a way of conversing—and differing—about important issues that substitutes respect for arrogance, and patient probing for overconfident boasting. If philosophical reason—and its Aeschylean partner, legal reasoning and the rule of law—prevail over anger, we may attain the result that Aeschylus imagines at the conclusion of his drama:  

That in plenty the blessings

That make life prosperous

May be made to burgeon from the earth

By the sun’s radiant beam. 

1 For comments that have helped me a lot, I’m very grateful to Douglas Baird, Ro Khanna, Brian Leiter, Saul Levmore, Richard McAdams, Charles Nussbaum, Laura Weinrib, and David Weisbach. Their generosity exemplifies the spirit of critical dialogue that I talk about at the end of this lecture.  

Disclaimer: this is a very nerdy blog entry. It is about lightweight markup languages and why I think that Org-mode is the best lightweight markup language for many use-cases. And with lightweight markup language, I do mean the syntax, the way you express headings, lists, font variations such as bold face or italic, and such things.

Please do note that this is not about Emacs. This is about Org-mode syntax and its advantages even when used outside of Emacs. You can type Org-mode in vim, notepad.exe, Atom, Notepad++, and all other text editors out there. And in my opinion it does have advantages compared to the other, common lightweight markup standards such as Markdown, AsciiDoc, Wikitext or reStructuredText.

Of course, Org-mode is my favorite syntax. Despite my personal choice you will see that I've got some pretty convincing arguments that underline my statement as well. So this is not just a matter of personal taste.

If you already have a grin on your face because you don't have any clue what this is all about: keep on reading. It makes an excellent example for making fun of nerds at your next dinner party. ;-)

Here you are. This is almost anything you need to know about Org-mode syntax:

 * This Is A Heading
 ** This Is A Sub-Heading
 *** And A Sub-Sub-Heading

 Paragraphs are separated by at least one empty line.

 *bold* /italic/ _underlined_ +strikethrough+ =monospaced=
 [[http://Karl-Voit.at][Link description]]
 http://Karl-Voit.at → link without description

 : Simple pre-formatted text such as for source code.
 : This also respects the line breaks. *bold* is not bold here.

 - list item
 - another item
   - sub-item
     1. also enumerated
     2. if you like
 - [ ] yet to be done
 - [X] item which is done	  

I've seen many coworkers who typed Org-mode markup when taking notes in their text editor. And they did not even know anything about it. So it is that intuitive I'd say.

While I was learning Org-mode, I did not even use a cheat-sheet for the syntax as I normally do. It was very natural for me to type Org-mode right from the start.

Tables are a bit more complicated like in all other markup languages I know of:

 | My Column 1 | My Column 2 | Last Column |
 |-------------+-------------+-------------|
 |          42 | foo         | bar         |
 |          23 | baz         | abcdefg     |
 |-------------+-------------+-------------|
 |          65 |             |             |	  

You most probably won't type a table like this outside of Emacs. The manual alignment without tool-support is very tedious. But even here you are able to deliver a perfectly fine Org-mode table by simply ignoring the alignment altogether:

 | My Column 1 | My Column 2 | Last Column |
 |--+-+---|
 | 42 | foo | bar |
 | 23 | baz | abcdefg |
 |--+--+-|
 | 65 | | |	  

This is an almost ridiculous argument because in my opinion a markup is of no use when it is not the same for tool A as for tool B.

However, there are markup languages that are different. For example the very widely used markup language named Markdown has many flavors to choose from:

Pandoc lists six different Markdown flavors as output formats. This is an absolutely bad situation which foils the original idea behind lightweight markup languages. When some web service tells me that I can use "Markdown" for a text field, I have to dig deeper to find out which of those many different Markdown standards the web page is talking about. After this I will have to continue and look for a cheat-sheet of this dialect because nothing is more difficult to differentiate than multiple standards that are almost the same but not really the same. A usability hell. I get furious every time I have to enter this hell.

With Org-mode, life is easy. The snippet from the previous section explains all there is. Any tool that interprets Org-mode accepts this simple and easy to remember syntax.

Many lightweight markup languages do offer multiple ways of typing headings. There are basically three ways of defining headings:

1. Prefix headings
2. Pre- and postfix headings
3. Underlined headings

Here are some examples for each category:

 Prefix headings:

 # Heading 1
 ## Heading 2
 ### Heading 3

 Pre- and postfix headings:

 = Heading 1 =
 == Heading 2 ==
 === Heading 3 ===

 Underlined headings:

 Heading 1
 =========

 Heading 2
 ~~~~~~~~~

 Heading 3
 *********	  

I prefer the prefix heading style. Org-mode use this as well with * as prefix characters. The more asterisks, the deeper the level of the heading is.

Pre- and postfix headings do offer bad usability. The user has manually synchronize the number of prefix character with the number of postfix characters. And it is totally unclear how something like = heading == with different numbers of pre/postfix characters is going to turn out when being interpreted.

And in case the user already used a markup language with simple prefix headings, it is not logical why there is the need for the postfix characters at all.

Even worse than this is the underlined heading category. The user is completely irritated for multiple reasons. Besides the tedious manual work to align the stupid heading characters with the heading title, it is not clear what characters must be used for those heading lines. If you've got a bigger document with different levels of headings you get confused which heading character stands for which heading level.

Are the tilde characters level one? Or was it the equals characters? And how about asterisks? Without a cheat-sheet, the occasional markup user is completely lost.

This gets even more worse: some markup languages let you choose your "order" of heading characters. This results in weird situations. For example one author is starting to write a reStructuredText document using her favorite heading syntax. A second author is joining in and has to analyze the document in order to know what heading syntax he must use.

In the reStructuredText mode of Emacs you can find following function:

You can visualize the hierarchy of the section adornments in the current buffer by invoking rst-display-adornments-hierarchy, bound on C-c C-a C-d. A temporary buffer will appear with fake section titles rendered in the style of the current document. This can be useful when editing other people's documents to find out which section adornments correspond to which levels.

Yes, you got it right, it is true: this function's only purpose is to generate a dummy-hierarchy of headings to visualize which markup has to be used for heading 1, which one for heading 2 and so forth just for this single document. What a bad design decision of the markup when you need such hacks just to know how a heading should look like in a markup even if you are familiar with in the first place.

Here is one more: some markup languages even allow mixed heading styles. You can use an underlined heading style for heading level 1, a prefix style for level 2, another underlining style for level 3 and so forth. Now the chaos is a perfect one.

Let's have a look at a different markup element: external links. As you already remember in Org-mode, a link looks like this:

 [[http://Karl-Voit.at][my home page]]	  

The only difficult thing here is to remember that the URL is at the beginning and the description follows after the URL. Many markup languages do add additional and unnecessary levels of difficulties.

Here are some examples from Wikipedia and comments by me where a user might be irritated.

AsciiDoc:

 http://example.com[Text]	  

The form is simple but for complex URLs, the [Text] might look like being part of the URL itself. Not beautiful but at least something I could live with.

Markdown:

 [Text](http://example.com)
 [Text](http://example.com "Title")	  

Brackets or parentheses first? Why using different kind of markup characters in the first place like only brackets? Is the Title part of the URL? Why not part of Text? Very confusing design decisions from my point of view.

reStructuredText:

 `Text <http://example.com/>`_	  

Holy moly. This is some weird stuff. First, you have to grave accents ` and not apostrophes '. Then what about the underscore character at the end? This is as complicated as you can define a simple URL. I'd even prefer the hard to type HTML version of linking. A disaster for something which has "lightweight" in its class name.

The simple syntax of Org-mode does not imply typing unnecessary characters. You don't have to manually align something like underlined headings. Anybody using a simple text editor is very fast at adding markup for headings, font variations, and so forth. The previous section proved that other markup languages clearly fail in many cases.

You don't have to use the Emacs editor to write and work with Org-mode markup text. As I mentioned above, many people already do so just because Org-mode is an intuitive and clean way of typing text characters.

When you've got text information in Org-mode markup, you can process it with many tools. Most prominent and most important examples are files pushed within a GitHub repository and the swiss army knife named Pandoc which is able to convert Org-mode to dozens of formats like HTML, odt (LibreOffice), docx (Word), LaTeX, PDF, and so forth.

As I mentioned in the beginning, this is not an article about Emacs. Nevertheless for anybody not familiar with Emacs I have to mention that with Emacs there is a tool that supports (not only) in writing Org-mode syntax in a perfect way.

You might start with mouse-only usage. There are menu items with all important functions. For the users that want to get a minimum of efficiency, the menu items show you the keyboard shortcuts you might want to use.

For Org-mode it is really easy to learn. Basically you just have to use TAB for toggle the collapsing and expanding of headings, lists, and blocks. It's Alt and the arrow keys to move around headings, list items, and even table columns/rows. Ctrl-Return creates a new heading or list item without the need of entering the markup characters and manually matching indentation levels at all.

That's it. With those three things you're good to write Org-mode syntax efficiently. The basic file open/save, finding help, exiting Emacs stuff is accessible with icons or the menu. No need to learn more keyboard shortcuts if you don't want to.

Having experienced this great tool-support, users typically are eager to learn more. You don't have to. You might be happy with Org-mode for capturing minutes of meetings and your shopping list. However, others do master a few additional things and write whole eBookswithin Org-mode.

Lightweight markup languages are designed to be used with a minimum effort compared to full-blown and therefore more complicated markup languages such as HTML or LaTeX.

Some are doing their job better than others. In my experience, many design decisions of widely adapted markups such as Markdown or reStructuredText (and others) are questionable from a usability point of view. At least I do have some issues when I have to use them in my daily life.

Unfortunately, I hardly see any people out there using Org-mode as a markup language outside of Emacs although there are very good reasons for it as an easy to learn and easy to use markup language.

With this blog article I wanted to point out the usefulness of Org-mode even when you are not using Emacs as an writing tool.

See also the discussion on this article on reddit.

But this strategy means that Equifax is entrenched in consumers’ financial lives whether they like it or not — or even know it. Equifax’s approach amplified the consequences of the breach, reported this month, that exposed the personal information for up to 143 million people.

Ordinary people are not Equifax’s customers. They are the company’s product. The “Big Three” credit bureaus, Equifax, Experian and TransUnion, collect 4.5 billion pieces of data each month to feed into their credit reports.

From birth to death, the record grows. Decades’ worth of addresses and identifying information, including drivers’ licenses and Social Security numbers. Utility accounts like telephone and cable subscriptions. Criminal records, medical debt, as well as rental and eviction histories.

Equifax’s records on any given individual, scattered throughout dozens of databases, typically stretch across hundreds or thousands of pages.

Equifax now faces a consumer backlash over its response to the hacking attack. The anger has been intensified by the actions of three senior executives who sold shares worth $1.8 million in the days after the breach was discovered. The stock, which had tripled in the last five years, is down 30 percent since the attack. Equifax said the executives were unaware of the breach when they sold their stock.

Customers have been less vocal, given their dependency on the bureaus. Financial firms readily hand over their data because they rely on the credit reports — and the scores they are used to generate — to size up potential customers. The data, over which Equifax and the other bureaus have a stranglehold, is one of the best predictors of risk.

“We don’t really have a choice to opt out of the credit report system,” said Pete Mills, senior vice president of residential policy at the Mortgage Bankers Association, which represents some of Equifax’s biggest clients, home loan providers. “We spend a lot of money trying to protect our customers, and then we give that data to others,” like the credit bureaus.

Equifax said it was supporting customers who may have been affected by the data breach. “We value our customers and have been in close communication with them,” said Wyatt Jefferies, a company spokesman.

Under Mr. Smith, Equifax has been creative in developing new markets and services. The company expanded globally, often by acquiring local competitors; it now operates in 24 countries.

New analytic products have been a priority. Equifax has a team of mathematicians who mine its data to develop algorithms predicting how consumers will behave. Those insights are sold to companies like lenders.

At a financial conference last year, Mr. Smith described a new system that searched four billion public tweets for keywords like “car” and “automotive lease.” It paired the tweets with a person’s Equifax credit file. In real time, the credit bureau could identify potential buyers and provide its customer, a company selling car leases, with everything it wanted to know about those people.

The corporate culture shifted under Mr. Smith and became more focused on increasing profit, said David Galas, who left Equifax in 2011 after 13 years.

“It was run a little more like a sports team,” said Mr. Galas, who served most recently as a vice president. “You immediately had to get out there and perform, and if you didn’t perform, you were cut.”

Equifax’s roots as a behind-the-scenes data collector stretch back to 1899, when it began as the Retail Credit Company. Grocers and other retailers kept notes on their customers to determine who could be trusted to run tabs and pay them. Two brothers in Atlanta went door to door to collect that information. They compiled it into a publication called “The Merchant’s Guide” and sold annual subscriptions for $25.

The company and its competitors swept through the country, employing thousands of investigators to investigate people’s lives. Their reports were widely available for sale to anyone except the individuals themselves.

In the 1960s, the credit bureaus’ secrecy and unchecked power prompted alarm within Congress. The hearings that followed exposed the more unsavory practices, like including unverified gossip about people’s marital indiscretions in their reports. The bureaus amassed personal dossiers so detailed that J. Edgar Hoover was covetous.

“The F.B.I. is constantly in our files,” an executive at a credit bureau testified.

Congress responded by passing the Fair Credit Reporting Act, which created some safeguards. For the first time, people were allowed to review their own files and report errors.

But the strongest agencies just kept growing, often by acquiring rivals. By the late 1990s, three big national players were left.

With little competition, the bureaus saw an opening for a new sales market: capitalizing on consumers’ curiosity and concern about their credit files.

In 2001, Equifax teamed up with Fair Isaac to let people buy their three-digit FICO credit scores. Today, Equifax charges people $40 to see all three of their reports. (Consumers are entitled to one free credit report from each of the bureaus annually.)

The company’s consumer business generates $400 million in annual sales, much of it through resellers. Using Equifax data, LifeLock sells identity theft protection, a booming business since the breach.

Such sales, while strong, are eclipsed by the money Equifax makes from human resources products. It entered the market in 2007 with the purchase of Talx, which verified employment for companies.

Mr. Smith viewed Talx as a beachhead into a lucrative new data field: payroll information. When Equifax bought the company, Talx held 142 million employment records. The unit now has 300 million.

“It’s been a nearly 10-year investment, but now it’s paying off for Equifax,” said Brett Horn, an investment analyst at Morningstar. “They have something their rivals don’t.”

A few expansion efforts fizzled, especially in tightly regulated markets. In 1995, Equifax teamed with AT&T to develop health care products, including electronic patient records. The effort quietly died a year later, right around the time that Congress passed a strict medical privacy bill.

As the industry expanded, safety became a sales pitch. “We have been blessed in our rich history to never have a major breach,” Mr. Smith said at a financial conference shortly after joining the company in 2005.

In one document, Equifax called itself the “trusted stewards of data.”

“If you’re not ahead of security risk,” the pitch read, “you’re behind it.”

After previous smaller breaches, the bureaus have been reluctant to offer consumers the strongest form of protection, credit freezes, free of charge. Freezing a file prevents new credit lines from being opened, which locks out identity thieves.

After Experian’s servers were attacked two years ago, exposing personal details on 15 million T-Mobile customers, consumer advocates urged both companies to provide free credit freezes at all three bureaus.

Doing that would set a terrible precedent and “haunt” all future breaches, Experian’s senior vice president of government affairs and public policy said in a response intended for executives at his company and T-Mobile. The reply was accidentally emailed to one of the advocates.

Giving in to the demand “will not satiate their hatred for Experian,” he added. Instead, he suggested responding with a letter explaining why fraud alerts were good enough. “We could turn our response into a good P.R. approach if done right,” he wrote.

Experian said in a statement that the opinions in the email did not reflect the company’s position. The company said it had provided affected individuals with free credit monitoring and credit freezes at Experian at no charge.

Equifax’s own response to its breach has been marred by blunders.

An Equifax website was supposed to allow customers to determine if they had been affected; it didn’t work correctly. The company’s Twitter account accidentally steered people toward a fake site. And when millions of consumers went to freeze their Equifax credit files, some had to pay for the service. After people protested, the company waived the fees.

From a business perspective, it will be paramount for Equifax to keep its customers — financial firms and other big companies — happy.

Six of America’s largest financial services companies — American Express, Bank of America, Capital One, Citibank, Discover and JPMorgan Chase — declined to comment on whether the breach would alter their relationships with Equifax. Walmart, the nation’s largest private employer, and Kroger, the second biggest, said they were comfortable continuing to send Equifax their payroll data.

Still, some — mainly smaller organizations — are beginning to rethink their relationship with the company.

Summit Credit Union in Madison, Wis., filed a lawsuit against Equifax. The firm is seeking compensation for the economic harm that it said it was likely to suffer from the breach.

“This situation has caused us all to pause,” said Sandi Papenfuhs, senior vice president of consumer lending at another firm, First Tech Federal Credit Union in Beaverton, Ore. “Anytime someone is not securing member data to the same degree that we do and we expect, we will take action on that relationship accordingly.”

But her credit union will continue to send Equifax data. Withholding information would only hurt consumers, she explained, because it would create an incomplete picture of their credit history.

“I am unaware of a way to just stop, from any individual lender perspective,” Ms. Papenfuhs said, “and not cause consumer harm.”

Web apps are impossible to secure

At the end of the 1990’s a horrible realisation was dawning on the software industry: security bugs in C/C++ programs weren’t rare one-off mistakes that could be addressed with ad-hoc processes. They were everywhere. People began to realise that if a piece of C/C++ was exposed to the internet, exploits would follow.

We can see how innocent the world was back then by reading the SANS report on Code Red from 2001:

“Representatives from Microsoft and United States security agencies held a press conference instructing users to download the patch available from Microsoft and indicated it as “a civic duty” to download this patch. CNN and other news outlets following the spread of Code Red urged users to patch their systems.”

Windows did have automatic updates, but if I recall correctly they were not switched on by default. The idea that software might change without the user’s permission was something of a taboo.

The industry began to change, but only with lots of screaming and denial. Back then it was conventional wisdom amongst Linux and Mac users that this was somehow a problem specific to Microsoft … that their systems were built by a superior breed of programmer. So whilst Microsoft accepted that it faced an existential crisis and introduced the “secure development lifecycle” (a huge retraining and process program) its competitors did very little. Redmond added a firewall to Windows XP and introduced code signing certificates. Mobile code became restricted. As it became apparent that security bugs were bottomless “Patch Tuesday” was introduced. Clever hackers kept discovering that bug types once considered benign were nonetheless exploitable, and exploit mitigations once considered strong could be worked around. The Mac and Linux communities slowly woke up to the fact that they were not magically immune to viruses and exploits.

The final turning point came in 2008 when Google launched Chrome, a project notable for the fact that it had put huge effort into a complex but completely invisible renderer sandbox. In other words, the industry's best engineers were openly admitting they could never write secure C++ no matter how hard they tried. This belief and design has become a de-facto standard.

Now it’s the web’s turn

Unfortunately, the web has not led us to the promised land of trustworthy apps. Whilst web apps are kind of sandboxed from the host OS, and that’s good, the apps themselves are hardly more robust than Windows code was circa 2001. Instead of fixing our legacy problems for good the web just replaced one kind of buffer overflow with another. Where desktop apps have exploit categories like “double free”, “stack smash”, “use after free” etc, web apps fix those but then re-introduce their own very similar mistakes: SQL injection, XSS, XSRF, header injection, MIME confusion, and so on.

This leads to a simple thesis:

I put it to you that it’s impossible to write secure web apps.

Let’s get the pedantry out of the way. I’m not talking about literally all web apps. Yes you can make a secure HTML Hello World, good for you.

I’m talking about actual web apps of decent size, written under realistic conditions, and it’s not a claim I make lightly. It’s a belief I developed during my eight years at Google, where I watched the best and brightest web developers ship exploitable software again and again.

The Google security team is one of the world’s best, perhaps the best, and they put together this helpful guide to some of the top mistakes people make as part of their internal training program. Here’s their advice on securely sending data to the browser for display:

To fix, there are several changes you can make. Any one of these changes will prevent currently possible attacks, but if you add several layers of protection (“defense in depth”) you protect against the possibility that you get one of the protections wrong and also against future browser vulnerabilities. First, use an XSRF token as discussed earlier to make sure that JSON results containing confidential data are only returned to your own pages. Second, your JSON response pages should only support POSTrequests, which prevents the script from being loaded via a script tag. Third, you should make sure that the script is not executable. The standard way of doing this is to append some non-executable prefix to it, like ])}while(1);</x>. A script running in the same domain can read the contents of the response and strip out the prefix, but scripts running in other domains can't.

NOTE: Making the script not executable is more subtle than it seems. It’s possible that what makes a script executable may change in the future if new scripting features or languages are introduced. Some people suggest that you can protect the script by making it a comment by surrounding it with /* and */, but that's not as simple as it might seem. (Hint: what if someone included */ in one of their snippets?)

Reading this ridiculous pile of witchcraft and folklore always makes me laugh. It should be a joke, but it’s actually basic stuff that every web developer at Google is expected to know, just to put some data on the screen.

Actually you can do all of that and it still doesn’t work. The HEIST attack allows data to be stolen from a web app that implements even all the above mitigations and it doesn’t require any mistakes. It exploits unfixable design flaws in the web platform itself. Game over.

Not really! It gets worse! Protecting REST/JSON endpoints is only one of many different security problems a modern web developer must understand. There are dozens more (here’s an interesting example and another fun one).

My experience has been that attempting to hire a web developer that has even heard of all these landmines always ends in failure, let alone hiring one who can reliably avoid them. Hence my conclusion: if you can’t hire web devs that understand how to write secure web apps then writing secure web apps is impossible.

The core problem

Virtually all security problems on the web come from just a few core design issues:

• Buffers that don’t specify their length
• Protocols designed for documents not apps
• The same origin policy

Losing track of the size of your buffers is a classic source of vulnerabilities in C programs and the web has exactly the same problem: XSS and SQL injection exploits are all based on creating confusion about where a code buffer starts and a data buffer ends. The web is utterly dependent on textual protocols and formats, so buffers invariably must be parsed to discover their length. This opens up a universe of escaping, substitution and other issues that didn’t need to exist.

The fix: All buffers should be length prefixed from database, to frontend server, to user interface. There should never be a need to scan something for magic characters to determine where it ends. Note that this requires binary protocols, formats and UI logic throughout the entire stack.

HTTP and HTML were designed for documents. When Egor Homakov was able to break Authy’s 2-factor authentication product by simply typing “../sms” inside the SMS code input field, he succeeded because like all web services Authy is built on a stack designed for hypertext, not software. Path traversal is helpful if what you’re accessing is an actual set of directories with HTML files in them, as Sir Tim intended. If you’re presenting a programming API as “documents” then path traversal can be fatal.

REST was bad enough when it returned XML, but nowadays XML is unfashionable and instead the web uses JSON, a format so badly designed it actually has an entire section in its wiki page just about security issues.

The fix: Let’s stop pretending REST is a good idea. REST is a bad idea that twists HTTP into something it’s not, only to work around the limits of the browser, another tool twisted into being something it was never meant to be. This can only end in tears. Taking into account the previous point, client/server communication should be using binary protocols that are designed specifically for the RPC use case.

The same origin policy is another developer experience straight out of a Stephen King novel. Quoth the wiki:

The behavior of same-origin checks and related mechanisms is not well-defined in a number of corner cases … this historically caused a fair number of security problems.

In addition, many legacy cross-domain operations predating JavaScript are not subjected to same-origin checks.

Lastly, certain types of attacks, such as DNS rebinding or server-side proxies, permit the host name check to be partly subverted.

The SOP is a result of Netscape bolting code onto a document format. It doesn’t actually make any sense and you wouldn’t design an app platform that way if you had more than 10 days to do it in. Still, we can’t really blame them as Netscape was a startup working under intense time pressure, and as we already covered above, back then nobody was thinking much about security anyway. For a 10 day coding marathon it could have been worse.

Regardless of our sympathy it’s the SOP that lies at the heart of the HEIST attack, and HEIST appears to break almost all real web apps in ways that probably can’t be fixed, at least not without breaking backwards compatibility. That’s one more reason writing secure web apps is impossible.

The fix: apps need a clear identity and shouldn’t be sharing security tokens with each other by default. If you don’t have permission to access a server you shouldn’t be able to send it messages. Every platform except the web gets this right.

There are a bunch of other design problems in the web that make it hard to secure, but the above examples are hopefully enough to convince.

A new open-ended puzzle game from the maker of Manufactoria.

• Build complex electronics from a variety of simple components, like Adders, Latches and Multiplexers.
• Travel back to the 60s to Silicon Valley's very first startup, and do your best to keep the whole thing from imploding.
• Solve more than sixty puzzles, from straightforward introductions to alarmingly elaborate head-scratchers.
• ...and if those aren't challenging enough already, try to optimize your machines and reach the best possible solutions.
• Original soundtrack by Craig Barnes, composer for Streets of Rogue and Enyo.

"The best game about CPU design that I can imagine" - Zach Barth, developer of SpaceChem and TIS-100.

Buy the game now on Steam or itch.io!

Preview on Gaming Nexus.

Development news and community:

From an office at Carnegie Mellon, my colleague John Miller and I had evolved a computer program with a taste for genocide.

This was certainly not our intent. We were not scholars of race, or war. We were interested in the emergence of primitive cooperation. So we built machines that lived in an imaginary society, and made them play a game with each other—one known to engender complex social behavior just as surely as a mushy banana makes fruit flies.

The game is called Prisoner’s Dilemma. It takes many guises, but it is at heart a story about two individuals that can choose to cooperate or to cheat. If they both cheat, they both suffer. If they both cooperate, they both prosper. But if one tries to cooperate while the other cheats, the cheater prospers even more.

The game has a generality that appeals to a political philosopher, but a rigorous specificity that makes it possible to guide computer simulations. As a tool for the mathematical study of human behavior, it is the equivalent of Galileo’s inclined plane, or Gregor Mendel’s pea plants. Do you join the strike, or sneak across the picket line? Rein in production to keep prices high, or undercut the cartel and flood the market? Pull your weight in a study group, or leave the work to others?

Woe to those who did not know the code.

Our simulation was simple: In a virtual world, decision-making machines with limited powers of reasoning played the game over and over. We, as the unforgiving account-keepers, rewarded the ones who prospered and punished the ones who did not. Successful machines passed their strategies to the next generation, with the occasional slight variations designed to imitate the blind distortions typical of cultural evolution.

We also gave the machines a simple language to think with and enough resources to have memories and to act on them. Each generation, paired machines faced each other multiple times. This is how life appears to us: We encounter our trading partners over and over, and how we treat them has consequences. Our model for the world was two Robinson Crusoes encountering each other on the sands.

When we ran these little societies forward, we expected to confirm what many believed to the optimal strategy for playing Prisoner’s Dilemma: tit-for-tat. A machine playing this strategy begins by keeping its promises, but retaliates against an instance of cheating by cheating, once, in return. Tit-for-tat is the playground rule of honor: Treat others well, unless they give you reason otherwise—and be reasonably quick to forgive.

Yet when we looked at the output of our simulations, where the strategies were free to evolve in arbitrary directions, we saw something very different. After an early, chaotic period, a single machine would rise rapidly to dominance, taking over its imaginary world for hundreds of generations until, just as suddenly, it collapsed, sending the world into a chaos of conflict out of which the next cycle arose. An archaeologist of such a world would have encountered thick layers of prosperity alternating with eras of ash and bone.

Instead of an orderly playground ruled by cautious, prideful cooperators, the population produced bizarre configurations that made no sense to us. That is, until one evening, in the office and after filling up pads of graph paper, we stumbled onto the truth. The dominant machines had taken players’ actions to be a code by which they could recognize when they were faced with copies of themselves.

In the opening moves of the game, they would tap out a distinct pattern: cooperate, cheat, cheat, cooperate, cheat, cooperate (for example). If their opponent responded in exactly the same fashion, cheating when they cheated, cooperating when they cooperated, they would eventually switch to a phase of permanent cooperation, rewarding the opponent with the benefits of action to mutual advantage.

Woe, however, to those who did not know the code. Any deviation from the expected sequence was rewarded with total and permanent war. Such a response might take both machines down, in a kind of a digital suicide attack. Because the sequence was so hard to hit upon by accident, only the descendants of ruling machines could profit from the post-code era of selfless cooperation. All others were killed off, including those using the tit-for-tat strategy. This domination would last until enough errors accumulated in the code handed down between generations for dominant machines to stop recognizing each other. Then, they would turn against each other as viciously as they once turned against outsiders, in a kind of population-level autoimmune disease.

As long as the codes lasted we called them Shibboleths, after the tribal genocide recounted in the Old Testament Book of Judges:

And the Gileadites took the passages of Jordan before the Ephraimites: and it was so, that when those Ephraimites which were escaped said, Let me go over; that the men of Gilead said unto him, Art thou an Ephraimite? If he said, Nay; / Then said they unto him, Say now Shibboleth: and he said Sibboleth: for he could not frame to pronounce it right. Then they took him, and slew him at the passages of Jordan: and there fell at that time of the Ephraimites forty and two thousand.

Shibboleths are a common feature of human culture and conflict. Finns who could not pronounce yksi (meaning “one”) were identified as Russians during the Finnish Civil War. Tourists in downtown Manhattan quickly out themselves if they pronounce Houston Street like the city in Texas.

Here our machines had used them to dominate a population so effectively that no others could survive. Even after the era was over, it was their descendants that inherited the ashes. The blind hand of evolution had found a simple, if vicious, solution.

It was a stark and brutal social landscape. But we had given our machines very limited resources to think with. How would two perfectly rational machines act in a conflict, if they each knew the other was similarly perfectly rational? By the very nature of rationality, two completely rational beings, confronted with the same problem, ought to behave in the same fashion. Knowing this, each would choose to cooperate—but not out of altruism. Each would recognize that if it were to cheat, its opponent would too, making them both losers in the game.

The two endpoints establish a spectrum. At one end are our minimally-calculating machines, parochial zero-points of culture that naturally, we found, distilled down to a vicious tribalism. At the other end is the inevitable cooperation of the perfectly rational agent.

On this line between beastly machines and angelic rationality, where do we find the human species?

If we humans are super-rational, or at least on our way there, there is reason to be optimistic. Francis Fukuyama might have been thinking along these lines when he penned his end-of-history thesis in 1992. Though Fukuyama’s argument was rooted in 19th-century German philosophers such as Friedrich Nietzsche and Georg Wilhelm Friedrich Hegel, we might rewrite it this way: A sufficiently complex simulation of human life would terminate in a rational, liberal-democratic, and capitalist order standing against a scattered and dispersing set of enemies.

Fukuyama’s argument was based not just on philosophical speculation, but on a reading of then-current events: the collapse of communism, the flourishing of electronic media, the apparently frictionless opening of borders, and a stock market beginning an epic bull run.

Today his thesis seems like a monument to the dreams of an earlier era (one chapter was titled “The Victory of the VCR”). Our cultures are evolving today, but not, it seems, toward any harmony. The chaos of the 21st century makes our simulations feel immediately familiar. Two decades after 9/11, even the Western liberal democracies are willing to consider dark models of human behavior, and darker theorists than Fukuyama.

Carl Schmitt, for example, who saw the deliberative elements of democracy as window dressing on more authoritarian forms of power. Or Robert Michels, whose studies of political inequality led him to see democracy as a temporary stage in the evolution of society to rule by a small, closed elite. As intellectuals at both political extremes increasingly see the possibility of a rational political order as a fantasy, Shibboleths take up their role in defining racial, national, and religious boundaries and appear once again to be ineradicable features of political life.

There is a great, and rich, valley between these philosophies, and another between the computer models that match them—between the simple, violent and less-than-rational agents that John Miller and I simulated, and the super-rational cooperators that Fukuyama might have considered to be waiting at the end of history. The models, at least, encourage a guarded optimism.

Researchers associated with meetings at the Machine Intelligence Research Institute (MIRI) in Berkeley have studied the behavior of rational but resource-limited machines who could inspect each other’s source code. Such transparency might seem to solve the problem of cooperation: If I can predict what my opponent will do by simulating his source code, I might decide cheating is not worth the cost. But what if my opponent’s code includes a simulation of what I will do as a consequence of running that simulation, and tries to exploit that knowledge? Without the symmetry of perfect rationality, this problem leads to some extreme mental contortions.

Some of the machines in MIRI’s bestiary might remind you of people you know. “CliqueBot,” for example, simply cooperates with anyone who shares the same source code. It only cares about codes that match its own letter-for-letter. “FairBot,” on the other hand, tries to look beneath surface differences to prove that an opponent will cooperate with someone like itself. Informally, FairBot says, “if I can prove that my opponent will cooperate with me, I’ll cooperate with him.”

How do these machines get along? While the full solution is a paradox of regress, studies of predictive machine behavior in a Prisoner’s Dilemma standoff provide the comforting answer that mutual cooperation remains at least possible, even for the resource-limited player. FairBot, for example, can recognize similarly-fair machines even if they have different source code, suggesting that diversity and cooperation are not impossible, at least when intelligence is sufficiently high.¹

Even the genocidal machines at the violent end of the spectrum may carry a heartening lesson. They emerged from the depths of a circuit board, simulated on a supercomputer in Texas. They had no biological excuse to fall back on. Maybe we, too, shouldn’t make excuses: If a behavior is so common as to emerge in the simplest simulations, perhaps we ought neither to fear it, nor to idolize it, but to treat it, the same way we do cancer, or the flu.

What if we saw tribalism as a natural malfunction of any cognitive system, silicon or carbon? As neither a universal truth or unavoidable sin, but something to be overcome? 

Simon DeDeo is an assistant professor at Carnegie Mellon University, where he runs the Laboratory for Social Minds, and external faculty at the Santa Fe Institute.

The author would like to thank the Alan Turing Institute for their summer hospitality while this article was written.

References

1. Barasz, M., et al. Robust cooperation in the Prisoner’s Dilemma: Program equilibrium via provability logic. arXiv 1401.5577 (2014).

Mastodon is the new social media platform, a decentralized alternative to Twitter that is currently blowing up. This is a step by step guide on how to run your own Mastodon instance on Digital Ocean.

Set up a Droplet

Create a new docker droplet:

This droplet has almost everything we will need preinstalled.

You will receive an email from DO with the credentials you can use to log in to start setting up the server.

Connect to the server as a root user, using ip and password from the email:

You will be prompted to change the default password, so do that.

Then create a new user with the username you like, and grant him the sudo powers:

adduser ray
gpasswd -a ray sudo

Connect domain name

Let's also immediately point your domain name to the droplet. After buying the domain(I recommend using namecheap), change the Custom DNS settings to look like this:

Then, in DO's networking tab, create a domain, and add an A record pointing to the droplet:

Now you will be able to ssh into your server using your new username and a domain name:

Install and configure basic stuff

Update and upgrade all the software:

sudo apt-get update && sudo apt-get upgrade

Install nginx(we'll use it to serve our droplet on the right port), and your favorite text editor:

sudo apt-get install nginx emacs

Now let's add docker to the sudo group, that will allow us to run all the docker commands without sudo:

sudo usermod -aG docker $USER
sudo service docker restart

Clone and configure Mastodon

Clone the repo and cd into it:

git clone https://github.com/tootsuite/mastodon.git
cd mastodon

Now let's configure some settings. First, rename the file .env.production.sample into .env.production and open it.

Set the database username/password settings:

DB_USER=your_username
DB_NAME=your_databasename
DB_PASS=your_password

Set your domain name:

LOCAL_DOMAIN=hackertribe.io

And enable https:

Run docker-compose run --rm web rake secret to generate PAPERCLIP_SECRET, SECRET_KEY_BASE, and OTP_SECRET.

Configure the email server

Create a SendGrid account, go to Settings > API Keys, and generate an API key.

Then set up the config like this:

SMTP_SERVER=smtp.sendgrid.net
SMTP_PORT=587
SMTP_LOGIN=apikey
SMTP_PASSWORD=
SMTP_FROM_ADDRESS=youremail@gmail.com

(for SMTP_LOGIN literally just use "apikey")

Configure the site info

Open the file /mastodon/config/settings.yml, and enter the information about your instance(title, description, etc).

Build the containers

Before we can build the containers, we need to add a swap file, without it my $10/month droplet was running out of memory during the build process. To add swap, execute these commands:

sudo fallocate -l 1G /swapfile
sudo chmod 600 /swapfile
sudo mkswap /swapfile
sudo swapon /swapfile

(you can read more in depth about it here)

Now let's finally build our containers! (It will take a few minutes)

docker-compose build
docker-compose up -d

(the -d flag means that we want to run it in the background mode. You can try running it without this flag, and you will see the log of everything that's going on on the screen)

Create the DB and migrate

Now we need to run several commands in the db container to create the database.

SSH into the container:

docker exec -i -t mastodon_db_1  /bin/bash

Switch to the postgres user:

Create a user for your db(use the username and password you've just set in the .env.production)

createuser -P your_username

Create a database, giving the ownership rights to the user:

createdb your_databasename -O your_username

Now you can get back to your own user, and run the migrations:

docker-compose run --rm web rails db:migrate

Precompile assets

Now you can precompile the assets:

docker-compose run --rm web rails assets:precompile

After this has finished, restart the containers:

docker stop $(docker ps -a -q) && docker-compose up -d

And now your mastodon instance will run on yourdomain.com:3000!!

Setting up nginx and SSL

First, follow this guide to generate SSL keys and set up the basic nginx configuration.

Then, because the docker containers are serving the application on the port 3000, we will need to use nginx to proxy all the requests to them.

Create the file /etc/nginx/sites-enabled/mastodon_nginx.conf, and copy the settings from here.

Now, after you restart nginx:

sudo /etc/init.d/nginx restart

It will serve your Mastodon instance!

Conclusion

Congratulations =) Create an account, test things, and invite some people to use your instance!

I also recommend submitting a link to your instance to this list to make it easier for people to discover it.

If you're looking for some help with deployment - send an email to contact@startuplab.io, and I will setup mastodon for you($100).

Also, always feel free to toot at me at @startuplab@mastodon.social, I will be happy to answer your questions =)

Has your working day become one long battle to wade through a to-do list?  

The multiple distractions of the modern workplace – digital overload, open offices and constant interruptions, to name a few – can make it near impossible to achieve your goals, or even get anything done at all.

But, what if you’re going about things the wrong way? Perhaps you should be thinking more about what you shouldn’t be doing instead.

You may also like:

- A novel trick to beat procrastination
- Why you should manage your energy, not your time
- Why paper is the real ‘killer app’

That’s one of the strategies employed by Canadian entrepreneur and investor Andrew Wilkinson, who has come up with a list of “anti-goals”.

Topsy-turvy thinking

Wilkinson noticed his day (and that of his business partner) was filled with things he didn’t want to do. He was feeling stretched, doing business with people he didn’t like, with a schedule dictated by others, he wrote recently on Medium.

'Inversion’ is a strategy that looks at problems in reverse, to minimise the negatives instead of maximising the positives

He wanted to figure out how to improve his day and make it more enjoyable. So, he followed the lead of Charlie Munger, right-hand man of famed investor Warren Buffet, and a proponent of ‘inversion’ – a strategy that looks at problems in reverse, focusing on minimising the negatives instead of maximising the positives.

To put it in practice, Wilkinson came up with his worst possible workday: one filled with long meetings at the office, a packed schedule dealing with people he didn’t like or trust. Then he came up with his list of ‘anti-goals,’ which includes no morning meetings, no more than two hours of scheduled time per day and no dealings with people he doesn’t like.

These ‘anti-goals’ have made his life “immeasurably better” he wrote in the blog.

“I think people always try to think about where they want to go. ‘What will make me happy?’ is such an open-ended question, and it’s surprisingly much easier to figure out what makes you miserable,” he wrote in an email to BBC Capital.

What not to do

Wilkinson and Munger aren’t the only ones using anti-goals to help them cut out distractions and realise ambitions.

Don’t let people ramble, don’t agree to meetings with no clear agenda, and work shouldn’t fill a void that should be filled elsewhere

Tim Ferriss, author, podcaster and investor believes in the power of a ‘not-to-do’ list. Why? “The reason is simple: What you don’t do determines what you can do,” he writes. On his not-to-do list? Don’t let people ramble, don’t agree to meetings with no clear agenda, and work shouldn’t fill a void that should be filled elsewhere.

Another fan? Angela Ceberano, founder of Flourish PR, a public relations firm in Melbourne, Australia. She uses the ‘traffic light system’ to list things to ‘stop, start or continue’ doing. Stopping unproductive activities is crucial for goal attainment as it allows a clearer direction, she says.

Why it works

Many productivity experts promote forward-looking thoughts and actions, so how can focusing on the negative work? By helping us reflect on and cut out activities that don’t align with our broader goals, says Repa Patel, an Australia-based executive coach and director of leadership development firm Leading Mindfully.

Focusing on the negative helps us reflect on and cut out activities that don’t align with our broader goals

Anti-goals, says Maurice Schweitzer, a professor of operations, information and decisions at the Wharton School at the University of Pennsylvania, “Give us a step-by-step process for thinking about things a little differently.”

Wilkinson’s list is specific to him, says Schweitzer. They are a set of crisp, clear, guidelines that are broken down into actionable steps and, therefore, attainable. It’s about prioritising that which is important.

Anti-goals can give us a different perspective “in a way that helps us identify an underlying issue,” he says.

“Goals narrow our focus and motivate us in a specific direction.”

To comment on this story or anything else you have seen on BBC Capital, please head over to our Facebook page or message us on Twitter.

If you liked this story, sign up for the weekly bbc.com features newsletter called "If You Only Read 6 Things This Week". A handpicked selection of stories from BBC Future, Culture, Capital and Travel, delivered to your inbox every Friday.

Copenhagen-based startup UIzard Technologies trained a neural network to automatically generate code from a graphical user interface screenshot.

Currently, a UI designer mocks up the interface and then the front-end developer takes the design and translates it into code – but, what if an AI system can do this for you? The company’s slogan is “code less, create more.”

“Implementing GUI code is time-consuming and prevent developers from dedicating the majority of their time implementing the actual features and logic of the software they are building,” mentioned the founder and CEO Tony Beltramelli in the related research paper. “Moreover, the computer languages used to implement such GUIs are specific to each target platform; thus resulting in tedious and repetitive work when the software being built is expected to run on multiple platforms using native technologies.”

Using CUDA, Tesla K80 GPU and cuDNN with the TensorFlow deep learning framework, their trained model takes a screen grab of the UI design, assesses the picture— various icons, features, and the layout— and then generates lines of code based on what it sees.

As of now, their able to generate code targeting three different platforms (iOS, Android and web-based technologies) from a single input image with over 77% accuracy, but they say it can drastically improve by training their networks on larger datasets.

Read more >

“Aren’t you tired of just nodding along when your friends starts talking about morphisms? Do you feel left out when your coworkers discuss a coproduct endofunctor?

From the dark corners of mathematics to a programming language near you, category theory offers a compact but powerful set of tools to build and reason about programs. If you ever wondered what’s a category or a functor and why care, this series might be just what are you looking for.

But don’t wait! If you call now, you’ll get this explanation of dual categories!

Next time, you too can be the soul of the party and impress your friends with category theory!*”

*(results may vary)

Intro

Category theory is a branch of abstract math. Why it gets so much attention from (functional) programmers?

As it happens, modeling programs using category theory allows us to apply theoretical results directly to our code, explore new approaches to existing problems, and increase our confidence on the solutions. At first, category theory might seem impenetrable, but one can go far by learning the basic vocabulary

But let’s go to the beginning

What’s one of the most important technique for programming?

Removing unnecessary detail and keeping the essence is an extremely powerful tool for programming.

What if we dial it to eleven?

Let’s abstract over all the characteristics of the things we want to model, and just end with “things” (called objects) and the connections between them (called arrows, or if you want to get really fancy, morphisms).

Just things and the connections between them:

To make a category, we are going to require only two things: every object is connected with itself (identity) and if object A is connected with object B which in turn is connected with object C, we can consider that object A is connected with object C (composition)

If we formalize the definition:

A category Cat is structure consisting of:

Obj(Cat): collection of objects.

For each A,B ∈ Obj(Cat), there’s a set C(A,B) of morphisms from A to B

f:A→B means f ∈ C(A,B)

(In other words, for every pair of objects A and B, there’s a bunch of arrows connecting them… or not)

A composition operation between arrows:

if f:A→B and g:B→C, then g∘f:A→C

(I can make a “new” arrow connecting the end of f with the beginning of g )

For each object X, exists an identity arrow:

Id_X:X→X

We’re going to have only two requirements (laws) for the identity and composition of a category:

Identity as unit

For any arrow f:A→B,

f∘Id_A = f = Id_B∘f

(f composed with identity on A is equal to f and is equal identity on B composed with f)

Composition is associative

f∘(g∘h) = (f∘g)∘h

(it doesn’t matter if I compose f and g first or g and h first, the resulting composition is the same)

Some mathematical examples of categories are:

Set:  the category where the objects are sets and the arrows are functions from one set to another

Pfn: the category of sets and partial functions

How is related to programming?

Yes, most examples in category theory are from math, but what about programming? If we consider the types of a program and the functions between those types, we can form a category: function composition will be our arrow composition and the identity function applied to each type will be our identity morphism. (with the caveat that we have to consider all our functions total and ignore non-termination [infinite loops, exceptions, etc],  also known as bottom ‘_|_’ ).

This is a toy program that takes a String, parses it as Int, divides by two, and gets the byte value of the result.

In Scala:

def toInt(s: String) = s.toInt

def divByTwo(i: Int): Float = i/2f

val program: String => Byte = (toInt _) andThen (divByTwo _) andThen (_.byteValue)

Scala provides an identity function,  so we know that identity[String] , identity[Int], identity[Float], and , identity[Byte] exists. Also, andThen acts as function composition in Scala

Now, is easy to see that (toInt _) andThen (divByTwo _) andThen (_.byteValue) == (toInt _) andThen ((divByTwo _) andThen (_.byteValue)) == ((toInt _) andThen (divByTwo _)) andThen (_.byteValue)

Since identity is defined as def identity[A](x: A): A = x , we can verify identity andThen f == f == f andThen identity is true for any f

So, if we squint and pretend the functions are total, we have:

Objects:  String, Int, Float, Byte

Arrows: toInt _,  divByTwo _ , _.byteValue

Id: identity

composition:  andThen

identity is neutral and andThen is associative.

So we can model our program with category theory, and take advantage of it.

That’s where the applicability of concepts like Functor, Monad, Natural transformations, etc.  come from.

In our next articles we’re going to expand on why that’s useful… stay tunned.