Minority Report for trolls

A new paper suggests that it might be possible to identify potential trolls before they do their worst. Researchers at Stanford and Cornell have pulled out patterns of behaviour exhibited by the approximately one in 40 users of three news sites—CNN, Breitbart and IGN—who were subsequently banned for abuse. These include trolls’ unwillingness to mould their conversation to the slang of an online community; their propensity to swear; and the volume of contributions they make to a debate. Making an algorithm of these patterns, the researchers believe they can be 80% confident of identifying those likely to cause trouble within five posts online.
 

From The Economist. I wonder what percentage of online trolls are men. It must be a majority. Men are terrible.

Now we just need Tom Cruise and a team of soldiers crashing in through people's windows before they even turn into trolls: “Sir, we're arresting you for the future abusive tweet directed at Sarah Marks that was to take place later this evening at 11:09pm.”

I need me some precogs to help me pre-block people on Twitter before they troll me.

Altruistic punishment

Steve Randy Waldman on the rioting in Baltimore.

Politically motivated riots are a form of altruistic punishment. Look it up. Altruistic punishment is a “puzzle” to the sort of economist who thinks of homo economicus maximizing her utility, and a no-brainer to the game theorist who understands humans could never have survived if we actually were the kind of creature who succumbed to every prisoners’ dilemma. Altruistic punishment is behavior that imposes costs on third parties with no benefit to the punisher, often even at great cost to the punisher. To the idiot economist, it is a lose/lose situation, such a puzzle. For the record, I’m a fan of the phenomenon.
 
Does that mean I’m a fan of these riots, that I condone the burning of my own hometown? Fuck you and your tendentious entrapment games and Manichean choices, your my-team “ridiculing” of people you can claim support destruction. Altruistic punishment is essential to human affairs but it is hard. It is mixed, it is complicated, it is shades of gray. It is punishment first and foremost, and punishment hurts people, that’s its point. Altruistic punishment hurts the punisher too, that’s why it’s “altruistic”. It can’t be evaluated from the perspective of winners or losers within a direct and local context. It is a form of prosocial sacrifice, like fighting and dying in a war. If you write to say “they are hurting their own communities more than anyone” you are missing the point. Altruistic punishment is not a pissing match over who loses most. The punisher disclaims personal gain, accepts loss, sometimes great loss, in the name of a perceived good or in wrathful condemnation of a perceived evil.
 
So you want to evaluate riots, then, as tactic. Surely these rioters can’t imagine that this — this — will reduce the severity of policing, bring jobs to the inner city, diminish the carceral state. By the way, have I told you, fuck you? Altruistic punishment is generally not tactical. Altruistic punishment is emotional. The altruism in altruistic punishment is not pure, not saintly. The soldier takes pleasure even as he takes wounds exacting revenge for a fallen comrade on another human who was not, as an individual, his friend’s killer. The looter takes a pair of shoes, because why the fuck not? If you perceive the essence of the riots in the shoes you are an idiot. Altrustic punishment is not tactical, it is emotional, and it is sometimes but not always functional. It functions, sometimes, to change expectations about what is possible or desirable or acceptable. In economist words, people’s propensity for altruistic punishment changes the expected payoffs associated with nonaltruistic behavior by those punished directly and, more importantly, by third parties who observe the unpleasantness. Changes in expected payoffs change the equilibria that ultimately prevail, in ways which may be beneficial for some groups or for “society as a whole”, however you define the welfare of that entity. Of course, there are no guarantees. Changes in expected payoffs can alter equilibria in undesirable directions as well. Drones anybody? This is a risky business.
 

Economists, game theorists, and psychologists come up with some of the most elegant names for various phenomena. I'd never heard of altruistic punishment, but I'll never forget it now. That is, sometimes when you bear the costs of the world unjustly, it can feel like the only way to set things right is to force a more even distribution of the costs across the population.

Housing and wealth inequality

In the last 50 years, the two societies have become even more unequal. Although a relatively small black middle class has been permitted to integrate itself into mainstream America, those left behind are more segregated now than they were in 1968.
 
When the Kerner Commission blamed “white society” and “white institutions,” it employed euphemisms to avoid naming the culprits everyone knew at the time. It was not a vague white society that created ghettos but government—federal, state, and local—that employed explicitly racial laws, policies, and regulations to ensure that black Americans would live impoverished, and separately from whites. Baltimore’s ghetto was not created by private discrimination, income differences, personal preferences, or demographic trends, but by purposeful action of government in violation of the Fifth, Thirteenth, and Fourteenth Amendments. These constitutional violations have never been remedied, and we are paying the price in the violence we saw this week.
 
...
 
Nationwide, black family incomes are now about 60 percent of white family incomes, but black household wealth is only about 5 percent of white household wealth. In Baltimore and elsewhere, the distressed condition of African American working- and lower-middle-class families is almost entirely attributable to federal policy that prohibited black families from accumulating housing equity during the suburban boom that moved white families into single-family homes from the mid-1930s to the mid-1960s—and thus from bequeathing that wealth to their children and grandchildren, as white suburbanites have done.
 

A must read on all the policies that have entrenched and amplified segregation in Baltimore. If we learned nothing from The Wire we should have learned that the forces aligned against the black population in Baltimore was structural, built in to our institutions, and not just the work of a few pernicious people. The whole environment is hostile, toxic.

Baltimore, not at all uniquely, has experienced a century of public policy designed, consciously so, to segregate and impoverish its black population. A legacy of these policies is the rioting we have seen in Baltimore. Whether after the 1967 wave of riots that led to the Kerner Commission report, after the 1992 Los Angeles riot that followed the acquittal of police officers who beat Rodney King, or after the recent wave of confrontations and vandalism following police killings of black men, community leaders typically say, properly, that violence isn’t the answer and that after peace is restored, we can deal with the underlying problems. We never do so.
 
Certainly, African American citizens of Baltimore were provoked by aggressive, hostile, even murderous policing, but Spiro Agnew had it right. Without suburban integration, something barely on today’s public policy agenda, ghetto conditions will persist, giving rise to aggressive policing and the riots that inevitably ensue. Like Ferguson before it, Baltimore will not be the last such conflagration the nation needlessly experiences.
 

This ties closely with a 2014 paper (PDF) from MIT grad student Matthew Rognlie that argues that what's responsible for the rise in wealth inequality is not, as Piketty argues, r - g, but instead the ability for the wealthy to capture outsized returns on housing and to pass that wealth on to future generations through that real estate. 1

  1. Rognlie's paper, fascinatingly, began as a comment on economics blog Marginal Revolution on a post about Krugman's review of Piketty (search for “Rognlie” on that post to see his two comments that launched, if not a revolution, at least an intellectual salvo). Here is a great profile of Rognlie and the story of how his comment turned into a paper that was the talk of the economics world. ↩

At the heart of Rognlie and Piketty's debate is how easy it will be for capital to substitute for labor in the future. Today it is still difficult, and thus Rognlie's prediction that the marginal returns to capital is still sound. Until robots can replace humans in more tasks, that should hold. In that environment, our best bet to decrease wealth inequality is to fight NIMBY-ism, loosen development codes, and build more housing.

WIP

James Somers wrote a Chrome extension called Draftback that allows you to play back a Google Doc so you can see every keystroke and edit that went into the writing. Chadwick Matlin spoke to Somers about Draftback for FiveThirtyEight:

Somers started all this because he thinks the way we teach writing is broken. “We know how to make a violinist better. We know how to make a pitcher better. We do not know how to make a writer better,” Somers told me. In other disciplines, the teaching happens as the student performs. A music instructor may adjust a student’s finger placement, or a pitching coach may tweak a lefty’s mechanics. But there’s no good way to look over a writer’s shoulder as she’s writing; if anything, that’ll prevent good writing.

As a result, finished work has become a writer’s currency. It’s what she hands in to her editors, what she publishes as a book, what she’s assessed on. The process of writing — the masochistic act of choosing what to put down on the page — is merely what she complains about to friends. It’s a hidden act, and self-conscious writers (as if there were any other kind) prefer it that way.

Somers wants to use Draftback to peek over somebody’s shoulder — ideally somebody really good. His personal goal is to get A.O. Scott, the film critic for The New York Times, to write a review or essay in Draftback. “He’s a beautiful prose stylist (diction, cadence, etc.), his writing is accessible and unpretentious but world-class, and he seems to always put his finger on the essence of whatever it is he’s talking about.” Somers is curious about whether all that comes naturally to Scott.

It's an interesting experiment into demystifying the craft of writing, but I tried watching a playback of Somers' writing of his post and I gave up after a few minutes. It's playback at too granular a level.

Much of good writing does occur at the editing or revision stage, call it post-production, and perhaps watching the fits and starts in Somers' piece is testament to that. Some writers sketch out the skeleton of a piece first, almost an outline, and then flesh in the actual prose. Others just dive in and start writing and find the structure later, or they have an intuitive sense of structure that just emerges sentence by sentence, like a sculpture being revealed from a block of stone.

Rather than seeing a piece played back one note at a time, I'd prefer to see a piece at a few discrete stages, perhaps the most valuable being the first rough draft, then the draft the author sends to an editor, and finally the draft marked up with the editor's proofs, especially if the editor were someone like the legendary Eleanor Gould. 1

The valuable lesson to be learned, in the end, may be the same one to be learned from watching anyone who is good at anything. Craft takes dedicated practice and lots of hard work. Sometimes viewing just the final product can hide all of that, and unearthing all the discarded drafts on the cutting room floor can be instructive.

  1. I've written before about my adoration for Gould; a collection of some of her most renowned proofs would be a national treasure. Let's hope The New Yorker has kept some and releases them in book form someday.