Monday, May 26, 2008

Never Confuse Corporations With Law Enforcement

Ariel Waldman Twitter harassment: the overview, @ Ars Technica

Online harrassment of women is pervasive and horrible, so I hate to say this, but Twitter is right and Flickr is wrong.

Since Twitter and I had an open dialog started, I would periodically report cases of continuing harassment (some of which spread between Flickr and Twitter). Twitter would take no action while Flickr would immediately ban and remove all traces of the harassment.

Here we've got two distinct approaches to handling harassment: Flickr plays the role of cop, and Twitter stays out of it.

Twitter's legal reasoning - fear of a lawsuit if they ban somebody - is totally incomprehensible to me. It almost seems insane. But the conclusion they reached, that it's not for them to play law enforcement, is totally reasonable.

Harassment on the Web is a problem, but attempting to solve it by handing individual corporations the role of law enforcement creates several additional problems.

First, it's not in every case actually law enforcement - insulting terms which carry no specific factual allegation are legal. I can call any random person a douchebag; but if I were to claim a person did some specific thing, with no actual evidence to suggest it occurred, or reason to believe it occurred, that would be illegal. Waldman's harasser called her a "cunt," which is legal, but also referred to her as a "crack-whore" and made up unfounded allegations about her as well. Both of these are (as far as I understand it) illegal under libel laws.

However, these are fine legal distinctions, and they're not for a random blogger with no formal legal training to make. They're not for corporate lawyers to make, either. Such lawyers have only their corporation's interests to protect. They are under no obligation to protect users, and giving them that role without also giving them that responsibility is foolish.

Because companies are under no legal obligation to protect our reputations, we have no protection under the law should they fail to do so; nor would we have any protection under the law if they were to conspire in damaging those reputations, either actively or through negligent application of their terms of service.

In a case of harassment and slander, you go to the courts, you don't place a phone call to Twitter's CEO. If you were robbed in a 7-11, you wouldn't demand justice from the cashier. You'd call 911. Likewise, if one of your users calls you up, asking you to play the role of law enforcement, you should decline. Banning or not banning users, and policing their behavior, is an inappropriate and legally risky role for corporations to play.

By playing the role of cop, Flickr encouraged people in thinking of them as authorities who protect, rather than as a corporation, whose primary interests are business interests. The problem there, of course, is that a corporation is what they actually are. Twitter made the same mistake, and it could bite them on the ass.

Twitter is taking the position that it "is a communication utility, not a mediator of content." In theory, this is a completely true and accurate position to take - but Twitter may not be able to take it. They've already, before this situation, banned users for harassment and for "inappropriate" content. By policing their content in the past, they have taken on a role they cannot sustain. They've created an expectation of protection, both through the actions they've taken in the past and through their terms of service. Having mediated content in the past severely undermines their position that they are not a mediator of content. In theory it should be true but in practice it has no credibility.

I have no idea if this thing is headed for the courts or not - I have nothing to do with it and taking a legalistic perspective is often overly dire and serious - but some case of this nature will end up in the courts sooner or later. It's only a matter of time. If you want to make sure you're not the Web app that gets to make legal history, the only responsible choice you can make is to categorically refuse to police your users in any way at all.

From the users' point of view, it's important to remember that social software is about shaping communities, and when you deal with rules which shape communities, you deal with politics.

Fascism should more properly be called corporatism, because it is the merger of state and corporate power.
- Benito Mussolini
(Fascist dictator of Italy and enemy of the United States in WW2)

Separation of state and corporation is as important as separation of church and state. Assigning your protection to a corporation means subordinating your protection to somebody else's business needs. The convention on the Web of terms of service "agreements" being site-specific equivalents to the law will probably one day be considered a short-lived historical curiousity, a footnote demonstrating the Web's Wild West origins. It is very much unscalable.