12 September 2010

Suspended License

A couple of fascinating decisions in the last week on the extent of copyright protections — and purchaser/user privileges — expose a couple of holes in current publishing practices. Well, not holes, exactly; more like gaping breaches.

Over across the pond, a court in Switzerland ruled that, as interpreted by my only feline colleague the IPKat, an internet actor

may not search peer-to-peer networks for copyright infringing files, register the IP [internet protocol] addresses of the uploaders and forward them to the copyright owners, which then, based on the data, file criminal complaints against unknown (the prosecution, unlike private parties, can force the ISPs to hand over name and address of the person/entity behind the IP address).

(German-language judicial press release) (PDF) This is inconsistent with both German and Dutch rulings. The key point is not the interference with antipiracy efforts per se, though, but the primacy of data-protection law over copyright law.

Similarly, the Ninth Circuit has asserted the primacy of contract law over copyright law... and missed most of the anticompetitive implications of its reasoning. In Vernor v. Autodesk, Inc., No. 09–35969 (9th Cir. 10 Sep 2010) (PDF), an individual purchased AutoCAD disks and requested a declaration that he could resell them on eBay. Autodesk opposed, arguing that the disks were mere physical artifacts, and that a user who obtains a copy of AutoCAD obtains only a license; and since Mr Vernor was not an "authorized dealer," he did not have the right to transfer that license via eBay at a price and on terms of his own choosing. The Ninth Circuit agreed:

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. Applying our holding to Autodesk’s [Software License Agreement], we conclude that [the eBay seller] was a licensee rather than an owner of copies of Release 14 and thus was not entitled to invoke the first sale doctrine or the essential step defense.

Slip op. at 13879–80 (footnote omitted). Notice, though, that this is based upon a unilateral declaration of license by the transferor (AutoDesk), which is inconsistent with ordinary custom for software. Although most software copies are transferred under a "license," those transfers are only so as to maintain control over the copyright in the software; they do not restrict later transfers of the copies of the software to third parties.1

Together, these decisions raise some really, really interesting questions. In no particular order:

  • Consider, for the moment, an individual who properly purchases AutoCAD software and then moves to Switzerland (EU law would disfavor enforcement of this kind of license). Now assume that this individual tries to sell his copy of the AutoCAD software on a Swiss equivalent of eBay, which is being probed by AutoCAD in the same way as the Swiss court prohibited Logistep AG from doing. No, this isn't just the civil procedure/conflicts of law nerd in me coming to play; it's a question that just sitting there, waiting to be asked on an exam. Or in practice.
  • DVD and videotape distributors have, for years, asserted a "license-like" restriction on every DVD. Does that fall within Vernor? What about a more-restrictive version of the existing "license"?
  • Can a publisher turn a printed book into a mere "license" by printing a license agreement on the copyright page of the book, and thereby restrict used sales of the book? (n.b. This is not a hypothetical.)
  • Virtually all publishing contracts are written under the premise that an author is selling certain rights to the publisher, which the publisher may then reuse by publishing the book. Leaving work for hire aside, this is inconsistent with the premise of the Berne Convention and the US Copyright Act of 1976, which (as evidenced in the US by such terms as § 203 and § 304(c)) now treat the author-publisher transaction as a license rather than a sale. Does Vernor represent some kind of tacit acknowledgment that the party who is offering copyrighted material in the stream of commerce has the authority to unilaterally declare that the transaction is a license?
  • More to the point, consider the way that e-book "libraries" are stored for most proprietary e-book readers, such as the Kindle. Under the reasoning of Vernor, the Big Brazilian River was in the right last summer when it deleted infringing copies of 1984 from Kindle-users' "libraries" without notice, because "license" implies "right to unilaterally terminate" in a way that "sale" does not.

In short, these opinions — especially when considered for possible conflicts-of-law issues — raise far, far more questions than they answered. Sarcastically, one might remark that that's the expected result in the common law; whether that enables citizens (and businesses) to comport their conduct to the law's requirements, though, is another question entirely. And the less said about how Vernor and Logistep might influence the pending ACTA negotiations, let alone the enforceability and/or reach of the GBS settlement, the better... as I have a football game to watch, and thinking about that sort of thing is going to inhibit my analysis of the battle in the trenches.


  1. I'm leaving the question of OEM licenses and versions of software aside for the moment; that's a much closer question, because the initial license is not for general use, but as part of a tying agreement that may — or may not — raise antitrust implications of its own.