Autodesk, EULAs and Games, Oh Boy

You may have recently heard of a court decision out of the Ninth Circuit involving horror stories about EULAs banning the right to resell games. There has been a lot of misinformation and fearmongering surrounding the case, with people shouting how it is the end of the world. It really isn’t, and I’d like to take the opportunity to go over the actual decision, as well as the existing law behind it, to explain why this will have minimal, if any, effect on gamers.

Background on First Sale

The case deals with two major fields of law: contracts and copyright. Put simply, a properly-written EULA is a legally enforceable contract. As long as there is nothing flawed in the drafting of the EULA (for instance, if it contains clauses that the court has deemed "unconscionable" because they violate public policy) and nothing flawed in the act of agreement (your consenting to the EULA, typically when you scroll down through all the legalese and click "I Agree" on the license while installing the game) it will be held to be enforceable.   What the EULA ends up doing for most software is saying "These are the rules under which you can use this program."  This argument has been shown to hold up in courts: Blizzard successfully used it to sue a bot manufacturer in World of Warcraft.

However there is a catch. Copyright has a doctrine called "First Sale", which acts as a limitation on the power of copyright holders to limit your ability to resell a work. The doctrine which has been around for over 100 years in common law (well before computers existed — keep that in mind), and was made part of federal statutory law in the current version of the Copyright Act, says that if you are the owner of a copy, you are entitled to sell it if you choose.  The key phrase here is owner. In order for the doctrine to apply, you must have lawfully received it through a sale — not a license. In a license, you are not the owner; you merely have permission to use the software.

So what happens when you have something that looks like a sale, but is being called a license? Well, the courts have traditionally held that "if it looks like a sale, it’s a sale; if it looks like a license, it’s a license." It was a very vague and relatively simple test, but one that’s been around since before computers existed and has long been accepted as the legal standard.  The courts would say "It doesn’t really matter what you call it. We’re going to determine whether it is a sale based on what the agreement looks like."

Until the late 2000’s, that’s where we were. The world wasn’t ending, software licensing proceeded as normal and case law developed around it, but the test remained that same vague "looks-like-a-duck" test.

The Facts Leading Up To Vernor

Autodesk, most famous for making "AutoCAD" and perhaps more famously in the games industry, making the modeling program Maya, has a big problem with piracy. Most of the products they make are very expensive (typically in the hundreds to thousands of dollars per copy), and are typically sold on a "per-seat" license basis to large companies, who then use the software to make games.

Tim Vernor ran an eBay store, where he sold a used copy of AutoCAD.  Autodesk, claiming this was in violation of the license agreement which required that the software not be resold, sent a number of DMCA takedown demands, and eventually Vernor sued them. Vernor claimed that he never agreed to the EULA because he never opened the box or installed the software. Autodesk countered that it didn’t matter; the person who sold it to Vernor had no right to sell it, and therefore Vernor had no right to use it. Now, a note for everyone who just flipped out and said "BUT HE PAID FOR IT." Under Autodesk’s argument, Vernor basically got scammed; he would have the right to get his money back or sue the person who sold it. Courts aren’t always fair, to put it bluntly. In any event, the case came before the District Court in Washington.

The Vernor Decisions

The District Court ruled squarely in favor of Vernor, who had backing from the coalition of NPOs that represent free software interests: the EFF, the ALA, Public Knowledge, etc.  The court examined the law and determined that there were two conflicting sets of cases. Out of the four major cases on point, three were in favor of Autodesk, and one (the earliest case) was in favor of Vernor. However, the earliest case in support of Autodesk’s argument (called MAI v. Peak) had a major point overruled by Congress, and the other two cases relied heavily on that argument. The court concluded that the most solid rule to follow was the one in favor of Vernor, and ruled that he had made a sale, and was entitled to the protection of the first-sale doctrine.

Autodesk appealed, and the Ninth Circuit heard the case. Instead of saying that the four cases conflicted, however, the Ninth Circuit found a way to tie all four cases together and said "No, in fact, they outline a set of scenarios that act as a test for whether there is a license." The Ninth Circuit held that there was a three-pronged test: 1) does the copyright owner call it a license? 2) does the copyright owner restrict the end user’s ability to transfer the software? 3) Does the copyright owner impose use restrictions? Under this test, the Ninth Circuit reversed the lower decision, and found that there was in fact a license.  Autodesk sold the software as a "nontransferable license", required written consent to transfer or lease the software, imposed global use restrictions and restrictions against "modifying, translating, or reverse-engineering the software, removing any proprietary marks from the software or documentation, or defeating any copy protection device", and provided that the license would terminate if these restrictions were violated.

Now take a step back, and look at the Vernor three-prong test, compared to the prior test. The Ninth Circuit in their decision said "this looks like a license, and we’re going to call it a license." That’s the exact same thing it was before.  The only difference is now, post-Vernor, we have a clearer set of rules as to what "looks like a license".  In other words, the test is still "does it look like a duck", but now we have a written description of what a duck looks like, to compare with.

There’s a common misconception in this case that stems from the Ninth Circuit’s holding. A disturbingly large number of media outlets have been saying that the court was "required to follow precedent” or "had no choice" on this case. This is in fact, misstating the words in this decision. What the court actually said was that they were "bound to reconcile precedent." What this means is that when the District Court saw two competing sets of case-law they said "Oh Shi-….we better throw out one of these, because both can’t be right." The Ninth Circuit, on the other hand, said "WAAAAAITT a minute. You have to try and make the sets work together. You don’t just walk out because you had an argument with your spouse — you try and find a way to reconcile the differences."

The other major misconception is from irresponsible media outlets claiming that this decision means that the so-called "magic words" that something is a license will make it so under Vernor. As stated above, that’s not the case. Calling it a license is only one of the three parts. You still need to have some actual facts behind it that shows that there are enough restrictions to justify it actually BEING a license.  So that’s Vernor in a nutshell.

And here we are today.

Applicability To Games

So now we are in this post-Vernor world. Cue the mass hysteria of "OH GOD I WILL NEVER BE ABLE TO SELL A GAME AGAIN, THAT’S IT I’M MOVING TO MEXICO" nonsense. But as I said earlier, there really is no major change. The test still largely hinges upon whether the EULA is really just a sale in disguise.  If it was never remotely a sale to begin with, the outcome would have been a license whether you looked before or after Vernor.

The Ninth Circuit noted that there are some serious policy questions in play here, and that Congress has the opportunity to simply pass legislation making its views on the topic clear (as it did when it amended the Copyright Act to include first-sale, and it did when it overruled MAI v. Peak). For instance, if the court had ruled in favor of Vernor, that would have been a major blow to things like educational pricing. Think about it: If you buy a copy of Microsoft Windows on an educational discount for $99, and can then freely resell it to whoever you want, why would anyone buy the non-educational version at $500?  And regardless of this case, it is very likely that the issue will at least be brought before Congress for clarification, as both sides have very powerful advocates behind them.

The first obvious difference between the software from the case, and games, is that games generally don’t cost hundreds of dollars. AutoCAD and other similar programs cost ridiculous amounts per license, and often are sold in enterprise versions, where you buy a single copy of the program, and a certain number of "seats" that it can be installed on at once; or a single copy comes with multiple CD keys. That’s not standard practice in the games industry.  Games are generally sold on a one-to-one basis, which tends to weight towards being a sale, not a license.

Second, there are very few applicable restrictions on use that a copyright owner can impose on how you play a single-player, offline game. When you play say, Super Mario World, there is no online component, there is no legal restriction from Nintendo on what you can or can’t do, and to the extent that they tried to put any in place, it would be laughed out of court as unenforceable. Single-player games and offline games simply weigh more towards being a sale than a license.

MMOGs, however, are a different story, and are unclear.  In MMOGs, there is a very real interest to the copyright owners in determining how precisely you use their game. Sure, you can buy and install the software but if you break their terms of service you’ll be banned and your license to play revoked. Similarly, you only get the game for a limited amount of time — as I found out painfully with Tabula Rasa, you only get to use it until their servers shut down. You install them on a CD-Key basis which is tied to a paying account and can be activated only once, and most limit the way you can transfer items or accounts to other people. MMOGs, therefore, have a lot of similarities to a license.

Post-Vernor, nothing about MMOG’s changed. They still have a lot of similarities to a license, and those similarities are nearly identical to the characteristics outlined by Vernor — it’s called a license, it imposes notable restrictions on use, and it significantly restricts the user’s ability to transfer.  In other words, they would have been a license before Vernor, and they’re a license after Vernor. So really, nothing much changed for MMOGs.

But what Vernor DID do is helped to exclude other forms of software that don’t have those kinds of restrictions. In fact, I’d argue that now it is more clear than ever that single-player games and offline games are clearly intended to be sales protected by the first-sale doctrine, simply because they don’t past the Vernor test. Before, you’d have to go through a series of arguments on each side, and the judge would have to weigh each one, and different judges could come to different conclusions. Now, the uncertainty is gone. Prong one, prong two, prong three. Bang bang bang. It either has these restrictions, or it doesn’t, and the restrictions are either "notable" (i.e. significant and enforceable) or they aren’t.


So there you have it folks.  There’s really not a whole lot that has changed, but I hope you can see through some of the fearmongering and misinformation that have unfortunately been spread about this case. Hopefully the first-sale doctrine is now a bit more clear to you, and you’ll think a little bit more about what you’re agreeing to when you scroll past that legalese on your game installations.  For another really good take on the Vernor decision, I’d suggest you read Copyright and Technology Blog’s take — they’re an excellent source of information written by some very talented legal analysts. There’s also one bright spot in all of this for Vernor — in overturning the District Court, the Ninth Circuit allowed him to have his claim of "misuse of copyright" be heard, which if successful would basically prevent Autodesk from enforcing the infringement claim against him.

Dan Rosenthal is lawyer and analyst for the video games industry

Tweet about this on TwitterShare on FacebookShare on Google+Share on RedditEmail this to someone

Comments are closed.