Some legal non-answers with respect to online ad-blocking

My purpose here is not to discuss the ethics or economics of online ad-blocking, nor even to provide definitive answers as to the legality of online ad-blocking, but rather to identify a few of the moving pieces. It’s called “issue spotting,” and it’s how lawyers like to pass the time.

Copyright

As an initial matter, why would copyright have anything to say one way or the other about ad-blocking? Copyright provides rightsholders a fairly limited set of exclusive rights. It’s not immediately obvious how you’d fit ad-blocking into one of those buckets.

Ok, I’m hiding the ball. Copyright is probably relevant if you need a license to read a website (more on this below), in which case violating the license creates a copyright violation.

But ad-skipping per se doesn’t implicate copyright. As the 9th Circuit Court of Appeals held in a case involving skipping TV ads:

[C]ommercial-skipping does not implicate Fox’s copyright interest because Fox owns the copyrights to the television programs, not to the ads aired in the commercial breaks. If recording an entire copyrighted program is a fair use, the fact that viewers do not watch the ads not copyrighted by Fox cannot transform the recording into a copyright violation. Indeed, a recording made with PrimeTime Anytime still includes commercials; AutoHop simply skips those recorded commercials unless a viewer manually rewinds or fast-forwards into a commercial break. Thus, any analysis of the market harm should exclude consideration of AutoHop because ad-skipping does not implicate Fox’s copyright interests.

In other words, a user has no obligation under copyright to watch ads. A user is allowed to record a TV show to watch it later, and the recording doesn’t retroactively become not a fair use if the user skips ads during playback. This is partly the case because the program owner usually doesn’t even own the copyright to the ads, which means that the program owner lacks any kind of plausible standing to invoke copyright. But even if the program owner did own the copyright to the ads, there’s no right under copyright where you can prevent people from skipping over parts of your work. You’re not prevented by copyright from fast forwarding through the boring parts of movies. Changing the manner in which you watch a copyrighted work doesn’t invoke copyright.

This should all apply to online ad-blocking as well. To the extent you’re allowed to view a site, blocking ads only matters if it’s a violation of a license term (again, below). Otherwise you’re just changing the way you view something, which is fine.

Now, a creative lawyer might argue that blocking ads modifies a work and thus is an unauthorized “derivative work.” This usually runs smack in the ownership problem mentioned above. And it’s hard to see how simply changing the way in which you consume something is “preparing” a derivative work (whatever “preparing” means). But ignoring those points, users traditionally have been permitted to modify copies of works that they own. (See pages 3-6 in Sherwin’s white paper on first sale.) Users can copy portions of a book to repair it, tear ads out of magazines, and so on. None of this is considered to be preparing a “derivative work,” which should probably be understood to mean preparing a derivative work (such as a translation or a movie adaptation of a book) and then actually doing something public with it. In any event, the derivative work line of thinking seems pretty weak compared to the “license” theory, and I would admire any lawyer with the courage to claim that ripping the ads out of a magazine was an unauthorized infringment of a derivative work, which I think you’d need to believe if you also think that ad-blocking is the preparation of a derivative work.

But you’re violating some kind of “license”?

What license? A license just means “permission,” and you only need “permission” to do something you otherwise are not permitted to do. Permission can be conditioned–like “I give you permission to enter my property as long as you don’t get rowdy.” That means that if you violate the conditions of the license, it no longer applies to you, and you’re back to the default state. (If you get rowdy, you’re trespassing.) If you violate the terms of a license to do something you’re allowed to do anyway: big whoop. Doesn’t matter.

This is where it’s important to note that license is not a contract. A contract is a binding agreement between parties that involves an exchange of value. A license is just permission, and a license might be one of the pieces of “value” that is exchanged in a contract, but it is not a contract itself. (But, such a contract is often called a “license agreement” or just a “license” instead of a “contract” since legal ambiguity benefits lawyers.) In many cases whether something is one or the other might not matter: Maybe there’s no difference between “You agree to do X, I agree to show you content Y” (contract) and “Viewing Y is conditioned on your doing X” (license). But there are other times when it very much does matter: The contract version would be binding even if I could otherwise view Y legally without it. But to get to a contract you’d need to actually show consent of some kind. If there is no contract but only a license, and I could legally view Y without getting permission, then I can violate without breaking the law (here, copyright law). But licenses can be more powerful than contracts in other ways. A contract is only binding on the parties that agree to it–where there is “privity of contract.” But think of something like the GPL, which expressly allows people to do something (modify and distribute software) that would be illegal without a license. No one is allowed to do those things without following the terms of the license. So in that sense the license is binding on anyone who wants to do those things, even if there was no explicit agreement between the user and the software copyright holder.

Ok. As an initial matter, do you need “permission” to read a website? You don’t need permission to read a book, or to listen to music or watch a movie. So if you don’t need a license to do any of those things, why would websites be any different?

This is where copyright law’s stupidity with regard to all things digital comes to the fore. With software, simply running software is considered to be making a “reproduction” of it, which means that you can’t do it without some kind of permission to do it. (The “reproductions” in question relate to RAM copies or, in the old days, the copy of software you install to your hard drive from a floppy disk.) The law graciously says that if you own a copy of software, you are automatically granted a license to make whatever copies you need to make to use that software. But this, in turn, is why software companies and their lawyers are so keen to claim that users don’t actually “own” their copies of software–which means that users do need permission to run the software, which also means that conditions can be put on that permission. Which means that if you violate those conditions you’re back to the default state of making reproductions of a copyrighted work without permission, which violates copyright.

Side point: I would argue that software companies shouldn’t be able to do this. The fact of ownership can’t be changed with flimsy legal fictions, and if you own the actual physical media that software is on–whether it’s a hard drive, flash storage, or an optical disk–then you “own” a “copy” of the “software.” Because that’s both common sense and the plain meaning of the relevant statutory definitions. But some courts insist on giving software companies more power than they should have.

A better approach to any of this would be for the law to expressly exclude all transient, “essential step,” buffer, and RAM copies from the definition of “reproduction” so that copyright law stays out of the mess entirely. But that’s not where we are.

This sort of bizarre thinking where all digital “copies” no matter how utilitarian or etherial require a license can be applied to websites. Under this stupid way of looking at the world, reading a website is not like reading a book, because of the nature of digital communication. After all, “copies” of the site are being made on various servers, in your browser, cached on your local disk, and so on.

Anyway, this kind of crap is why it’s not totally beyond the pale to claim that you need a “license” to read a website even though you don’t to read a magazine. But I would argue that putting up a website at all creates an implied license for users to read it. The question then is whether part of this implied license is that “You must view all my ads.” I’d say that hinging a legal argument on the violation of some term of an unwritten, implied license is pretty weak.

It is more interesting to consider whether an express license (that is, one that is actually written down with words) can override whatever the implied license “says.”

Side point: I don’t think much would change if you say that any user reproductions of a website are fair uses rather than done under an implied license. I mean, viewing a website probably would be a fair use if you insist that legally-relevant reproductions are happening and there was no implied license. But there probably is one.

But you’re violating some kind of contract?

What contract? A website operator can indeed structure things so that you have to agree to a contract before using the site. Remember that a contract requires agreement between the parties and an exchange of value. The “value” you give the website is your agreement to the terms. The “value” the website gives you is their content. “Clickwrap” contracts that actually make users expressly agree to terms are often upheld. But most websites don’t do this. A site that requires users to log in and to agree to terms to use it might be on somewhat firmer ground in saying that users are in trouble if they violate those terms.

What about terms of service that are just buried away somewhere? These are sometimes found to be enforceable, as long as they are on a page that a user loads, or linked to from such a page. But they often are not. Most of the litigation around these “browsewrap” contracts (or attempts at contracts) tends to involve sites that do more than just present articles along with ads. It is probably not reasonable to assume that a reader is obligated to read the terms of some random newspaper site linked to from Facebook or Google News, though it might be “reasonable” to expect users who are transacting some sort of business with a site to know that there are probably terms somewhere and to hold them to them, even if empirically no one reads them. That said, even those sorts of browsewrap agreements are on very shaky ground legally compared to clickwrap agreements.

By the way, this -wrap thing is by analogy with “shrinkwrap” licenses, which are also kind of bogus.

What about liability for the makers of ad-blocking software?

Probably not happening. If there is indeed copyright infringement, maybe there would be secondary liability. But making a tool can’t lead to secondary liability if the tool is capable of substantial noninfringing uses, and ad-blocking technology is.

And tortious interference with a contract–forget it. That is a very hard claim to prove, and you basically have to show that the software developer is interfering with a contract between a reader and a site using means that are themselves wrongful (that is, not “privileged”). One classic example is that a competitor is allowed to encourage a customer to break a contract with another company, since competing to win business is “privileged.” Also, the claim usually requires that the alleged tortious interferer act “maliciously”–deliberately to harm someone. (In the previous example, trying to win new business is not trying to hurt your competitor for the hell of it. It’s just trying to win new business. Unless you’re Larry Ellison.) Ad blockers are just trying to get people to use their software or maybe harm the ad industry in general, but it’s unlikely they are motivated by specific malice toward individual websites, and even if they are, good luck proving it.

Also: maybe there’s a tortious interference claim against the user for interfering with the contractual relationship between the ad network and the publisher? You’d have to show malice against the specific publisher (since ad blocking doesn’t harm actual advertisers: they pay by impression)–and that viewing a site with an ad blocker harms it more than simply not viewing it at all. Seems like a tough claim.

CFAA

The Computer Fraud & Abuse Act is bad, and if you’re liable for other reasons, the CFAA could make you super-liable. It should be repealed, but in this context, it doesn’t change the underlying analysis much.

Random Common Law Stuff

Sure, maybe there’s a misappropriation, unjust enrichment, or a trespass to chattels claim here. I do not want to research these things.

Conclusion

There’s too much wiggle room to offer a firm conclusion. I’d say that a site with a clickwrap contract is on pretty firm ground if its terms say that readers are not allowed to block ads. To the extent that you need a “license” to read a site at all, you probably have an implied license. (If you don’t need such a license, a browsewrap is probably not worth the pixels it’s printed on.) Text buried away somewhere on a website is probably not enough to change the terms of that implied license, but courts come to different answers on this in different scenarios. Finally, when all else fails, there are loose common law claims.

My 2¢: ad blocking should be legal, since the default state is “things are legal,” and since the law is a terrible tool for dealing with complicated and changing ethical, technical, and business issues like ad-blocking.