We had just mentioned the infamous decision in the Vernor v. Autodesk case last week, in discussing the Psystar decision. If you don't recall, the court in the Vernor case effectively decimated the concept of "first sale" in software, making it questionable if you could ever resell software that you'd bought. To make that work, the court argued that software sales (even though it's "bought" in stores) are really "licenses," similar to rentals, rather than product sales and, thus, you can't resell. Of course, as someone noted in our comments, then why does Apple have a "How to Buy" page for its software, in which it is entirely described as a product you are purchasing.

In the Vernor case, the court gave a recipe for effectively destroying first sale. All anyone has to do is claim that they're licensing you something, even if it has every indication of being a full purchase. This seemed to contradict with the entire First Sale doctrine (and numerous other cases), but apparently the Supreme Court doesn't want to be bothered with this. It refused to hear Vernor's appeal, meaning the existing ruling sticks. This is one case where I could definitely see another appeals court coming to a different conclusion, meaning that, hopefully, the Supreme Court will revisit this issue at some point in the future. In the meantime, the first sale doctrine is severely limited to the point of near non-existence in software.

Last year, we had noted that the 9th Circuit appeals court was set to hear three separate cases, all revolving around the first sale doctrine, which allows you to resell copyrighted works that you possess. The first ruling of the three, back in September, was bad news: overruling a good district court ruling, in Vernor v. Autodesk, saying that anyone could effectively wipe out your first sale rights by simply putting a "license" on it. The second ruling, in MDY vs. Blizzard, was more of a mixed bag. It accepted the basics of Vernor but said that just because you violate a "license," it doesn't automatically mean you violate the copyright.

Now, the ruling in the third case, UMG vs. Augusto, has come out and it looks pretty good. It upholds the first sale rights of people who get "promotional" CDs (pdf of the ruling, which is also embedded below). Basically, the court seems to agree with the lower court's ruling, which suggested that being able to overrule first sale rights with a couple of sentences stamped on a CD, which the labels clearly never intended to get back, would undermine the entire principle of the first sale doctrine (though, it did so for different reasons).

So, how did the court square this ruling with its own decision in Vernor, which essentially said something different? It basically comes down to the fact that Universal Music gave out these CDs without expecting them back or without getting the original recipient to agree to anything specific. That is, the text stamped on the CD doesn't count as a true license agreement. But all the crap included with Autodesk software does count as a license agreement (rather than a true transfer of ownership):

It is one thing to say, as the statement does, that "acceptance" of the CD constitutes an agreement to a license and its restrictions, but it is quite another to maintain that "acceptance" may be assumed when the recipient makes no response at all. This record reflects no responses. Even when the evidence is viewed in the light most favorable to UMG, it does not show that any recipients agreed to enter into a license agreement with UMG when they received the CDs.

Because the record here is devoid of any indication that the recipients agreed to a license, there is no evidence to support a conclusion that licenses were established under the terms of the promotional statement. Accordingly, we conclude that UMG's transfer of possession to the recipients, without meaningful control or even knowledge of the status of the CDs after shipment, accomplished a transfer of title.

The main difference between the lower court ruling and this new ruling is that the appeals court focused on the lack of any actual "agreement" in the license, while the lower court focused on the lack of expected return of the promotional CD.

Not surprisingly, I think the court got this right but since I also believe that the earlier Vernor ruling was very, very wrong, it's not surprising that I think this ruling does a nifty little tap dance to pretend that this ruling and the Vernor ruling are consistent. It basically says that it all depends on "the means of distribution," in that recipients of promo CDs did not ask for them, while purchasers of software did. But that seems to be besides the point and somewhat unrelated. The same issue that seems to drive this ruling for Augusto should apply to Vernor as well. The court notes that there's "no acceptance" of a license and you can't assume acceptance via no response. And yet, that's how most such software licenses work as well.

While I'm happy about this particular ruling, it still seems to conflict with itself and I would imagine we haven't heard the last of these three first sale cases...

Earlier this year, we noted that the 9th Circuit appeals court was facing three separate important cases about the "first sale doctrine." We also warned that the 9th Circuit can be wacky at times, and its first ruling of the three confirms that, by effectively wiping out the first sale doctrine and stating that buying a piece of software is not really buying a piece of software -- and that, in many cases, you've merely licensed it and cannot resell it. The key point:

We hold today that a software user is a licensee rather than an owner of a copy where the copyright owner (1) specifies that the user is granted a license; (2) significantly restricts the user's ability to transfer the software; and (3) imposes notable use restrictions.

The full ruling is here: The case involved a regular eBay seller, who bought some copies of Autodesk at an "office sale" from a company who was no longer using them. He put four of the copies up on eBay for sale, and in each case, Autodesk sent a DMCA takedown notice. The guy, Timothy Vernor, counternoticed each time and Autodesk failed to reply each time, leading the auctions to be reinstated. However, despite all of this, when the 4th takedown came, eBay automatically suspended Vernor's account, and it took about a month to get it reinstated. So, he sued Autodesk, and claimed that he had the right to sell these copies that he had legally obtained.

Last year, the district court sided with Vernor, saying that Autodesk had no right to restrict his first sale rights (which allow you to resell a legally owned copyrighted product without first getting permission from the copyright holder), and that Autodesk's "license" was really a license in name only, since at no point did the possessor of the software have to return it. Instead, the court pointed out that when you bought AutoCAD, there really was a transfer of ownership.

Tragically, the appeals court disagrees with that, and effectively opens the door to effectively killing off the entire concept of the first sale doctrine, by highlighting the rather simple steps anyone needs to take to make sure any "sale" is really considered a "license," and thus removing the first sale rights. Basically, you just have to say there's a license and that "license" has to have a few rather simple things in it. You know all those stories about video game companies hating the used game market? Well, video game companies just got their "kill the secondary market free" card from the 9th Circuit. Of course, as we've pointed out, having a secondary market tends to increase the value of the primary market, so software companies (and others selling copyrighted works) may wish to think carefully before wiping out the secondary markets.

The court does note that this ruling could have quite a negative impact, but says that it's ruling based on what the law says, and suggests that perhaps Congress may wish to revisit this issue (fat chance of that happening, of course). While it is true that a court must rule within the law, even if it thinks the end result is bad, I think that the court here is using that as an excuse. The more you look at the details, the more you realize this is a license-in-name-only, and (as the lower court realized) the court could easily make that point and stay within the law, protecting these important first sale rights.

That said, this case is hardly over. Vernor's lawyer on the case, Greg Beck from Public Citizen has already announced plans to ask the court to reconsider the case en banc (i.e., with the full panel of judges, rather than just the three judge panel who heard the case), and even if that fails, this case will certainly be appealed to the Supreme Court. Given the two other first sale cases soon to be heard, combined with what appears to be (even if the court denies it) a circuit split with at least the 2nd Circuit on first sale issues, hopefully the Supreme Court will set the record straight and reinstate first sale rights for software. It would be great, of course, for Congress to just step in and fix things, but it's difficult to remember the last time Congress got something right when it came to copyright law... and with lobbying giants like the MPAA siding with Autodesk on this one, you can bet that no one in Congress will be able to secure enough votes to fix things.

This ruling is pretty depressing if you actually believe in property rights. It shows, once again, how copyright is not a property right, but often quite the opposite: restricting what people can do with their own property.

Zusha Elinson has noted that a trio of important cases concerning the "first sale doctrine" in copyright law will all be heard on appeal in the (at times wacky) 9th Circuit in early June. We've written about all three cases here before. There's UMG v. Augusto, which questions whether or not it's legal to resell "promo CDs" that record labels stamp "not for resale." In that case, the judge ruled that reselling was perfectly fine, and preventing such sales was a violation of the first sale doctrine (and would, effectively, give record labels a way of creating perpetual and all-controlling copyright, if the decision went the other way). Then there's Autodesk v. Vernor, similarly involving the right to resell software. Again, the district court ruled that this was allowed, noting that software is really sold, not (as Autodesk claimed) just licensed.

The third case is the troubling one. MDY v. Blizzard is the one case that went the other way, in a ruling that left many copyright experts scratching their heads, noting that it seemed to go against everything that the first sale doctrine stood for -- and that nothing the guy did (he made a bot that worked in World of Warcraft) actually violated copyright law.

Hopefully, the appeals court upholds the first two cases and reverses the third... but these days, you never know how courts are going to rule on these sorts of issues. And, the 9th Circuit is often notoriously... weird in some of its rulings. Either way, these are three cases worth watching, as they could have a pretty big impact on the question of whether or not you have the right to do what you want with products you bought.