Regulating The Internet? Not the Tubes Themselves…

If Net Neutrality is an argument about economics (and federal administrative law), Content Regulation is an argument about ethics and culture.

Net Neutrality is becoming an old hobby horse for a lot of people. It gets a lot more attention than most telecommunications policy issues. Even though questions about copper wire lines vs fiber optic cables actually affects more people, the internet is generally united by the fact of its own existence.  This is about regulation at the highest level, determining the equality and/or equity of access to content. No one online is indifferent to the internet—the only debate about net neutrality is which policies are best for the consumer and the telecommunications marketplace (or, in the United States, “telecommunications marketplace”).

But there is another layer of regulation that is quickly gaining attention. If Net Neutrality is about the form of the internet (its structure and broad organization), there is a growing need to consider questions about the regulation of the content of the internet. Over the years, the internet has been a vector for some amazingly good and amazingly bad actions by humans. The differences in the kind of regulatory concept at play are hard to understate. Rather than comparing it to different video games, I would compare it to the difference between a video game and a tabletop game.

1) I’ve always been fascinated by the dawn of the computer age. My childhood was the tail-end of a world in which homes did not have internet access. By the start of law school, everyone looked up famous cases and Latin phrases on Wikipedia during class (except for the people who did the reading the night before- they looked it up before class). I’ve often compared the early days of the internet to a kind of Wild West setting: a lawless frontier where fundamental questions about the mold of civilization were not yet settled. I thought most of those questions would be settled by 2015. We are not close to a consensus on rules. Indeed, we are still testing what types of rules are feasible or desirable.

Video games are literally made of rules: the computer code that constitutes the game itself. Tabletop games are made of… usually cardboard, or some kind of paper. (Occasionally, they have some plastic – or even metal if you got the collector’s edition.) This may sound like a silly or vacuous distinction, but it has important ramifications for the kinds of problems that can happen in a game, and the kinds of solutions that will (or won’t) be effective.

2) Lawlessness can lead to problems. This was probably not known until 2 decades of unfettered internet, but now we know. Free to do anything, people have tried very hard to do everything. Every app, platform, hosting site, game, or program online that gets big enough eventually starts to experience just about every problem type that humans can present. From intellectual property disputes to death threats, from fraud to manslaughter, the internet has been a way for people to discover criminal behaviors that past generations could never have the opportunity to access. The unethical choices of both multi-national companies and village simpletons are available for repeated viewing.

In a video game, the code can sometimes glitch and create problems for players. The code can also execute perfectly, but there may be complaints about the design of the game itself (a level being too difficult or some power or tactic being of an unsuitable level of power). With some difficulty, players can cheat by actually breaking the code, but more games can detect this (and especially so in professional e-sports settings). In a tabletop game, anyone can cheat, the rules may be wrongly applied (or not applied at all), and all manner of chaos can ensue. DDoSing an opponent during a game might be a little bit akin to literally flipping a table during a game of Monopoly or checkers,

3)  YouTube’s takedown system is already an example of an effort to regulate content, and it already shows some of the challenges with instituting a content regulation system: people will find ways to game that system. Any system of regulation will have two negative outcomes: it will penalize the innocent, and it will be dodged by the guilty. The most you can hope for is that it will protect most of the innocent and it will penalize most of the guilty. The US justice system, even when working as intended, will sometimes produce undesirable results: a guilty person will go free, and an innocent person will go to prison. The hope is that this happens very infrequently.

The most common reaction to bad behavior online has been for authoritative parties to do nothing. The most common reaction by authoritative parties to actually do something has been to ban the bad actor. The most common reaction to this ban is to come back with a different username or account.

In video games, cheaters are often banned (if they are making the game worse for other players). But in table top games, people who ruin the game are just not invited back. No one will play with them anymore. People might hang out with someone less if they behaved in a wildly unacceptable way during a casual weekend game of Risk or Werewolf. In a video game, bad behavior has very limited consequences. In a tabletop game, bad behavior can have lots of meaningful implications.

 

4) What would it look like to regular content? Getting it wrong is easy — which is the primary reason that’s what’s going to continue to happen. Whether trying to penalize criminals or regulate behavior online, creating a fair and ethical system that consistently produces more good results than bad ones is difficult. One problem is that incentives are at odds: most platforms want to turn a profit, and if bad behavior yields a net gain, the platform needs a solution that will actually make more money than the current bad behavior (plus the cost of implementing the remedy). Another problem is that platforms tend to think of regulating their content the way that most Americans think about regulations: an appointed governing authority (or combination of authorities).

 

Conclusion

You can’t make people be good, but you can keep deleting all of their manifestations of their behavior on the internet: You can suspend or ban accounts, and eventually IP addresses. You can automatically censor strings of characters, and continually update the list of banned strings. These will continue to be the solutions offered, and they will continue to mostly fail while they almost half-succeed.

Over a decade ago, Lawerence Lessig asserted that laws are of four types: market, cultural, legal, and architectural. It turns out that enforcing the legal type of law in a digital space is very difficult. But cultural norms practically enforce themselves. And architectural laws are always already enforced. Market rules can be fickle, but persuasive. A lot of efforts to regulate content will fail because they will hinge on the concepts of legal enforcement.

The lack of rules and regulations is what made the internet a place where amazing things could happen. Without rules to stop imagination and creativity, people created art, solved problems, built positive communities, and enriched themselves and each other. In that same landscape: without rules to stop hate and anger, people created harassment and bullying, invaded privacy, ruined lives, occasionally killed people, and destroyed a lot of good in the world. Lawless frontiers are the best opportunity for the most beautiful, important, and inspiring expressions of humanity. They are also the best opportunities for the most despicable, dangerous, and damaging expressions of humanity. What the internet becomes will be decided—has always been decided—by what people bring to it.

Advertisements

Software [Non-]Ownership: EULAs and Thinking About Property

We don’t really own most of the software we buy. As one writer put it a few years ago, “the software on my computer may as well be tied to a long piece of elastic, just waiting for the publishers to give it a tug.” That “piece of elastic” is a license, as in “End User LICENSE Agreement.” Almost all of the software we buy- especially what we download rather than physically purchase- is licensed to users by publishers and developers. These licenses vary from one piece of software to another*, but for a lot of games, the licensor (publisher or developer) has the legal right to take the game away from the licensee. Usually, the licensor will include specific reasons why they might do this, but will often round out the list with something like “or for any other reason.” There are not many limits on what this license cannot contain, must include, or how it has to be structured.

This model has been around for a long time, but I think it is fast becoming a serious problem. The core of the problem is that almost all users think, feel, and act as though they do own the software they have purchased. The American concept of property is still fundamentally rooted in John Locke’s Second Treatise of Government (a text that was a tremendous influence on the Founding Fathers and early US statesmen, politicians, and writers): if you work on something (or pay for it, or both), you have a claim of ownership on that thing. It is how we understand all of the other ownership paradigms in our society, and makes it easy to determine where ownership begins and ends. I cannot think of a single instance where one might purchase a physical object and not have full ownership rights over that object. Any arrangement where something is transferred with some kind of “elastic string” still attached is not called a “sale.” It is called “renting,” “leasing,” “borrowing,” or possibly “putting under mortgage.”

For most American minds, the concept of a “sale” includes the concept of a complete ownership transfer. I think it is easy to consider this one of the central reasons why there is so much resistance to copyright law, digital piracy laws, and other abstract controls on ownership of non-physical property. The laws that guide physical property do not apply to digital property, even though the only model we have for thinking about digital property is our history of thinking about physical property (Locke). A key point of contention is whether the legal concepts underpinning physical property must be different from the legal approach to digital property. The fact that US law does take these two different approaches does not mean that US law must take these two different approaches.

It may turn out that changes in net neutrality will affect how publishers and developers rethink this business model. As internet use becomes a worse experience, especially for data-intensive games, it is more important that companies ACTUALLY sell the product, not just license it. A data-choked internet will increase the need for offline gaming. It will create a whole new level of challenges for always-online DRM and increase the potential for server-crashing launch fiascos (e.g., Diablo3 and SimCity). A big reason that companies went to online-DRM models was to combat piracy. If net neutrality slips away, the ISPs might make piracy difficult enough (intentionally or unintentionally) to make developers feel more comfortable with moving to a sale-not-license model.

The only games currently practicing this kind of model are “abandonware” games: games whose developing companies have closed up, or have simply allowed their works to pass into the public without fuss. It is not clear that all presumed “abandonware” games are actually “orphan works” according to actual US copyright law. Indeed, a lot of games I’ve seen peddled under this banner are decidedly NOT orphan works and are not subject to the same freedom of transfer that the seller implies. But the videogame world operates on the legal principle of developers declining to enforce their civil rights against consumers as heavily as its programmers rely on the principle of “last in, first out.”

*I think the analysis is different for different software. Stricter license make sense for, say, reduced-cost versions of editing, publishing, or creative suite software for “Academic Use Only.” My position in this post is certainly not that “all licenses are bad,” or anything remotely close to that sentiment.

 

EDIT/ UPDATE:  Another recent blog post on this subject, from a slightly different angle.