Ninth Circuit Appeals Court Decision On Fair Use And Right Of First Sale Fails To Budge The Needle On Either Issue
from the we-hold-that-Adobe-litigated-itself-into-defeat dept
A ruling on fair use, the right of first sale and the limits of trademark protection has been handed down by the Ninth Circuit Court of Appeals panel. Normally, I’d proceed the word “ruling” with an adjective like “important,” or “terrible,” or “wonderful.” But this ruling is none of those. It’s a ruling, and I suppose it does set some sort of precedent, but thanks mainly to Adobe’s inept handling of the case, it does very little to clarify any of the above issues.
What it does do, however, is at least reaffirm burden-shifting for companies who insist they haven’t sold software, but “licenses.” This is the preferred method of undermining the right of first sale: tying up purchasers with legal language that claims they’re not making purchases, but rather “renting” the goods until the manufacturer says otherwise. This allows companies to stipulate all sorts of restrictions on purchased goods, including the resale of software.
Adobe tried to claim Joshua Christenson had never “purchased” any of its software, and therefore could not resell it. Christenson had in fact purchased software, but from third parties rather than directly from Adobe. Adobe didn’t like being cut out of the loop, so it took him to court for alleged copyright and trademark infringement.
Six years later, very little of it has stuck. (Christenson was sued in 2009.) The appeals court notes that Christenson successfully managed to shift the burden of proof on the licenses v. sales issue back on Adobe — which then failed to produce evidence that the reseller possessed only worthless licenses rather than resellable goods.
Noting that if construed broadly enough, software licenses could “swallow the statutory right of first sale defense,” the panel says that the burden of proof must be shifted back to the plaintiffs when alleging copyright infringement of licensed software. The reseller only has to prove he or she acquired the software legally. Adobe — being the manufacturer — is in a “superior” position to demonstrate whether or not the software in question was merely licensed.
As the district court held, it was uncontroverted that Christenson “lawfully purchased genuine copies of Adobe software from third-party suppliers before reselling those copies.” Christenson offered invoices to document his purchases of legitimate Adobe software from various suppliers. Nothing on those invoices suggests that he was other than a legitimate purchaser of the software. According to Christenson’s sworn statement, “[n]either [he] nor SSI have a contract with any of the suppliers that supplied SSI with software… SSI asked them if they could supply SSI with a product at an acceptable price, and if they could, payment was negotiated.” This claim is consistent with Christenson’s inability to produce something more than invoices from his suppliers: He cannot produce records that do not exist. Christenson discharged his burden with respect to the first sale defense.
Adobe countered that it never sold software. It only licensed it. Burden of proof shifted to Adobe, which came up empty.
Adobe’s problem is that it did not produce those licenses or document the terms of contracts with specific parties. Because of the state of discovery at the time of the summary judgment motions, the district court excluded virtually all of Adobe’s late-offered evidence of licenses. Adobe challenges this ruling in its appeal. The district court and magistrate judge had a long history with the parties and their discovery efforts. After a careful examination of the rather tortured discovery process, we conclude that the district court did not abuse its discretion in granting Christenson’s motion to strike and excluding evidence purporting to document the licenses…
Adobe’s effort to substitute general testimony and generic licensing templates in lieu of the actual licensing agreements does not withstand scrutiny under Vernor. Under Vernor, the precise terms of any agreement matter as to whether it is an agreement to license or to sell; the title of the agreement is not dispositive. And here, in the end, there is no admissible evidence that Adobe “significantly restrict[ed] the user’s ability to transfer the software” at issue here.
So, there’s not much in the way of a “win” for the right of first sale. While it’s heartening that the burden shifts to the “licensing” party to prove infringement, all this really does is encourage companies like Adobe to write more restrictive licenses. It does almost nothing at all to treat licenses as actual purchases, which can be resold without the manufacturer’s permission.
Other claims raised were similarly bungled by Adobe. The trademark infringement allegations were dead in the water because Christenson never used Adobe’s logos for anything else than selling actual Adobe products.
While it does appear there may have been some shady software sales by Christenson (purchasers complained about received academic versions rather than the full OEM versions they were expecting), Adobe failed to raise its false advertising claims until the case was already in the appeals court — far too late for them to be addressed.
In the end, we have a decision on fair use and the right of first sale that does nothing to push either issue forward. Fortunately, this stasis also does nothing to damage either of these defenses, something the Ninth Circuit Appeals Court has done in the past. One gets the disheartening feeling that if Adobe had managed its case better, the right of first sale would be worse off, rather than bolstered by the Appeals Court’s final opinion.
Filed Under: copyright, fair use, first sale, joshua christenson, licensing, ninth circuit, sale, software
Companies: adobe
Comments on “Ninth Circuit Appeals Court Decision On Fair Use And Right Of First Sale Fails To Budge The Needle On Either Issue”
This seems very similar to the Autodesk Autocad case from a few years ago. Guy there was buying used copies at flea markets and stuff, and reselling them.
Seems though, adobe’s case should be against the suppliers, not him. Buy not using the software, he never agreed to any of the licensing terms.
Someone should tell Adobe that they didn’t actually give them their money, they only licensed it to Adobe and can therefore restrict what they do with it.
Licenses and sales are not mutually exclusive
Vernor v. Autodesk was horrible in that it created a false dichotomy of “licensed OR sold” with respect to “software”. Adobe and Microsoft are big fans of claiming infringement of their copyrights in “software,” but the Copyright Act does not even use that word. What is does talk about are “computer programs,” which are literary works. Just like “books,” these works can be licensed (e.g., I license you the right to make a copy of my literary work, and recite it in public), or sold (e.g., I bough a copy of your literary work). Copies are, of course, material objects in which the work is fixed. It is perfectly sensible to sell a disc along with a license — e.g., I sell you this disc in which my work is fixed, and I license you the right to make two copies onto computer hard drives and one copy onto a backup DVD.
Christenson unfortunately failed to clearly state when it was referring to “software” as the intangible literary work and when it was referring to the tangible copy. But at least the context made that clear enough. It certainly seems to trash the Business Software Alliance’s mantra that “software is licenses, not sold.”
It would be nice for the Ninth Circuit to finally acknowledge the existence of Section 202, which makes perfectly clear that copies and copyrights are two different things. Ownership of a copy has no bearing on copyright ownership. Ownership of a copyright has no bearing on ownership of the copy.
My view is that the Ninth did indeed move the needle. There are active cases in the district courts in which Adobe is trying to claim that it only sells licenses for its “software”. It muddies the waters, and some judges seem to take a first glance at Vernor as standing for the proposition that software is always licensed, not sold (though Vernor does not go that far, and Augusto certainly focused on ownership of the copy — uh, phonorecord).
But I think Adobe knows its charade is on shaky ground. I mean, seriously, do its corporate assets include millions of discs and computer hard drives in millions of homes all over the world? Does it own my hard drive because I downloaded (reproduced the work onto my hard drive under license) it onto my hard drive, converting it into a copy of Adobe’s literary work? For that reason, Adobe is placing more emphasis on a “license to use” the literary work — a right found nowhere in the Copyright Act. It uses “serial number license keys” to purport to control who may “use” (privately perform?) its works, and where.
Our courts really need to step it up, and not let Adobe and its BSA cohorts redefine the Copyright Act. A vibrant first sale doctrine is essential to prevent those who depend on secondary markets from being disenfranchised from the promise of progress of science and the useful arts.
Re: Licenses and sales are not mutually exclusive
It is perfectly sensible to sell a disc along with a license — e.g., I sell you this disc in which my work is fixed, and I license you the right to make two copies onto computer hard drives and one copy onto a backup DVD.
No it’s not.
If you sell me the disc, then I own that copy of the computer program, and I can rely on 17 USC 117, which makes it non infringing for me to make copies and adaptations of the software so that I can use it, and to make backup copies. I don’t need your stinking license, unless, perhaps, I want a site license, which I often will not.
I’ve yet to find anyone who can make a good argument in favor of EULAs (outside of site licenses) as providing any value to users where physical copies are involved.
The best way to treat these transactions IMO is as ordinary UCC sales, under Klocek v. Gateway.
Re: Re: Licenses and sales are not mutually exclusive
You are partially right. No license is needed to do what the law allows. But still, it is perfectly normal — for centuries — to treat sales of copies and licensing of works as not mutually exclusive. The common law on that point was codified in 1909 (and currently appears at 17 U.S.C. § 202). Another example, outside the scope of 117, would be, “I sell you this disc in which my work is fixed, and I license you the right to authorize two non-owners of this disc to reproduce the work onto their respective hard drives.”
Also, in my prior example, a license to make a copy onto a backup DVD “for archival purposes” converts that into a lawfully made copy free from the limitation in 117(a)(2), so such EULA-licensed archival copy may be sold without infringing the copyright, wheras the 117(a)(2) archival copy may not.
Law vs. Facts.
The law hasn’t shifted, but the facts have. In Verner vs. Autodesk, there was a valid contract, signed by a firm of architects in exchange for a $15,000 discount. When it sued Verner, AutoDesk had all its ducks in a row. It had got an agent to purchase the software from Verner, for $600; looked up the serial number; identified the original purchaser; and retrieved the relevant signed contract.
However, in the present case, it seems that Adobe is relying on OEM copies and student copies, and is apparently having difficulty producing actual contracts with actual purchasers. As the value of each copy becomes less, the legalities inevitably become more sketchy. Lawyers and legal procedures cost money. I think you will find that Adobe did not make a representative available to negotiate with students purchasing the student edition, nor were the students legally advised. Obviously, a “click-wrap” license, which manifests itself only after sale, is worthless as a contract. The purchaser is within his rights to treat a click-wrap license as attempted fraud, and set his own price for accepting a refund. A refund from a company, which in many cases can only obtain credit at loan-shark rates, is an enforced loan. Beyond that, there is the whole concept of “contracts of adhesion.” A judge can go over a contract of adhesion, and strike out terms which seem inequitable to him. Likewise, there is the question of whether students in public institutions were sold software under duress or coercion, and the extent to which Adobe might have conspired in this.
Another relevant fact is that GIMP is very good, and will become better. Open-Source represents a greater and more immediate threat to Adobe than it does to AutoDesk. Graphics is not as much of a niche market as architecture. The final output of a graphics program is usually a JPEG or PNG file, or a piece of paper. The compatibility issues are much less complicated than those applying to architectural CAD/CAM.
The next road block is children. The law is very solicitous to prevent children from being exploited. Suppose it becomes the case that all persons who are disposed to learn to use graphics software do so at the age of eight. Obviously, they have access to GIMP, and if Adobe just does nothing, the children will grow up knowing how to use GIMP. The only way Adobe can keep pace is to simply give away vast numbers of no-strings-attached full versions of its software.
Re: Law vs. Facts.
If you want to avoid the haloing effect, use File-> Export to -> and save as a .png file.
GIMP does raster images. For vector images (you need that for logos, etc.), use Inkscape, which is also OS and generates both .bmp and .png files, which you can then pull into GIMP to make further adjustments to. If you need to resize the image go to View and untick “Dot for Dot” before you scale it.