Boing Boing Staging

Why are license "agreements" so uniformly terrible?

In the simplest terms, a license is a grant of permission to engage in some behavior that would otherwise be prohibited. You need a license to drive a car, a radio station needs a license to broadcast over the public airwaves, and James Bond needs a license to kill, all for the same reason. Without permission, those activities are against the law. Sometimes that permission comes from the government, and other times from private parties. If you enter your neighbor’s property without a license, you are a trespasser; with a license, you are an invited guest.

But modern license agreements have evolved into something else altogether. They create private regulatory schemes that impose all manner of obligations and restrictions, often without meaningful notice, much less assent. And in the process, licenses effectively rewrite the balance between creators and the public that our IP laws are meant to maintain. They are an effort to redefine sales, which transfer ownership to the buyer, as something more like conditional grants of access.

So what do licenses actually say? Most of us have no idea, and for good reason. License agreements are long, inscrutable, and full of bad news. They are the Lars von Trier films of legal documents. The form and substance of license agreements discourage consumers from reading them, which perversely reinforces their worst attributes.

Let’s start with their length. The current iTunes Terms and Conditions are over 19,000 words, translating into fifty-six pages of fine print, longer than Macbeth. Not to be outdone, PayPal’s terms weigh in at 36,000 words, besting Hamlet by a wide margin. The demands of these prolix legal documents are jaw-dropping. Take Adobe’s Flash, a software platform installed on millions of computers each day. Assume the average user can read the 3,500-word Flash license in ten minutes—a generous assumption given the dense legalese in which it is written. If everyone who installed Flash in a single day read the license, it would require collectively over 1,500 years of human attention. That’s true every single day, for just one software product. Imagine what would happen if you tried to read every license you encountered.


Not surprisingly, the vast majority of us simply throw up our hands and ignore licenses altogether. One recent study shows that as few as one out of a thousand software shoppers even glanced at the text of license agreements. And most who did spent only a few seconds perusing the terms. Even Chief Justice John Roberts, hardly known for his casual disregard for legal obligations, can’t be bothered to read EULAs. It hardly seems fair to expect more from the average person.

License terms are not negotiable. So there’s little to gain from a careful reading. Suppose you carefully examine the Flash license and find some objectionable term. Perhaps it limits you to a single installation of the program or disclaims liability for damage to your computer. What exactly are you going to do about it? Adobe is not going to negotiate a new license with you. They won’t even entertain the idea. So your choice is simple. Either use the product—and live with the license—or don’t. Take it or leave it.

Intentionally or not, rights holders and retailers have managed to nearly universally dissuade their customers from reading the terms that purportedly govern their purchases. And if the public rationally avoids investigating licenses, there is little marketplace incentive to offer more consumer-friendly terms. Better terms would simply go unnoticed. When software maker PC Pitstop included language in its license offering a cash prize to the first user to notice the clause, it took nearly four months before someone collected the $1,000.

When high-quality products are indistinguishable from poor ones, we get what economists call a market for lemons. Even though car buyers would pay more for a vehicle with no mechanical issues, they often can’t tell a reliable used car from a clunker destined to break down in a steaming heap in a week or two. And since they can’t sort the good deals from the bad ones, they are only willing to pay a price corresponding to a low-quality car. But if buyers aren’t willing to fork over extra money for a high-quality car, used car dealers have every reason to stock their lots with the cheapest cars available. So despite the fact that buyers would pay a premium for high- quality cars, the market fails to supply them.

For the same reasons, most EULAs are lemons. Licensors have lots of information about what their licenses say. They drafted them after all. But the average person has very little information. This information asymmetry breeds disengagement and distrust. And if companies don’t gain any advantage in the marketplace from more consumer-friendly licenses, that only serves to further entrench unfavorable terms. Once license terms are adopted, they have a way of spreading. In part, their viral nature is about saving time. Few licenses are drafted from scratch. Lawyers copy and paste liberally.

Somewhat less innocently, the uniformity in license terms is partly about safety in numbers. Once a term becomes standardized, its inclusion becomes a strategy for reducing competitive risk. A company that adopts industry standard terms guarantees that it is no worse off than its competitors. Combined with the lemon problem, this sort of soft collusion helps ensure that we don’t see robust competition on the basis of consumer-friendly license terms.

Instead, we see a growing list of standard terms, almost none of which add to a product’s value from the perspective of users. Some restrict what you can do with the products you purchase. These include limits on making backup copies, prohibitions on bad reviews, provisions permanently tying a product to a particular device, and bans on reverse engineering—the process of discovering how a product works through observing it in action. Other terms eliminate legal rights and remedies. These include limitations on liability, bars to class action suits, and mandatory arbitration clauses. And if the licensor neglected to include some one-sided term it later deems useful, many licenses give the drafter the option to change the terms of the EULA at any time.

But for our purposes, the most important license provisions are ones that try to redefine ownership and limit the transfer of the products we purchase. Across the board, nearly every license agreement for digital content—software, games, music, movies, and books—declares that the product is licensed, not sold. As Apple informs its customers, “the software products made available through the App Store … are licensed, not sold, to you.” Microsoft says the same thing: “We do not sell our software or your copy of it—we only license it.” Amazon’s Kindle store follows suit as well: “Kindle Content is licensed, not sold, to you by the Content Provider.” Sony’s PlayStation license states: “All Software is licensed, not sold, which means you acquire rights to use the Software … but you do not acquire ownership of the Software.” The same sort of terms are increasingly attached to hardware devices with embedded software. Most of these licenses preclude you from reselling, lending, renting, or otherwise transferring their purchases.

There are two ways to interpret these kinds of terms. They might stand for the uncontroversial—and frankly obvious—proposition that when you buy a product like Microsoft Office software or an iPhone, you are not acquiring all of the copyright, patent, and trademark rights in that product. To which we say, duh. But these licenses typically mean something beyond that. They mean that you don’t own the thing you buy. You don’t own and can’t transfer the plastic disc, the digital file, or the physical copy of the code embedded in your phone. So when retailers and record labels tell you that the song you purchased is licensed, not sold, they mean two things—you don’t own the copyright in the song and you don’t own the file you downloaded.

The End of Ownership: Personal Property in the Digital Economy [Aaron Perzanowski and Jason Schultz/MIT Press]

Exit mobile version