Closing the Video Pipeline

UPDATE: See above for more.

I will get to Bunner soon – I have to get through a couple of trade secret docs first. See Matt’s list of links for the scoop and have fun parsing the victory spins from each side.

For now, I want to get in the wayback machine and talk about the Video Pipeline case.  Matt and Frank were both a little up in arms about this case with good reason. Not that it came out wrong – given caselaw, it’s not necessarily wrong. But it is a bit troubling for several reasons.

I agree with Matt that, at least for the sake of clarity, the judge should have dealt with the clip previews (Video Pipeline’s own trailers) and the copied trailers separately.  I also found it hard to discern to which copyrighted work he’s comparing Video Pipeline’s materials.  The original film or Disney’s trailers?

This is especially disconcerting because of the weird reasoning regarding Disney’s entering the trailers market. In both the first and fourth factors of the fair use test, Judge Simandle focuses on how Disney has already created trailers that serve the same purpose as Video Pipeline’s. Ruling against fair use in regards to Video Pipeline’s copied trailers makes sense for this reason because they do not complement Disney’s trailers. (Yes, there is the registration issue here, but I don’t think that’s a huge deal. Since they do already own the underlying elements, it makes some sense that they also own the trailer they actually created.)

But if we apply that to the clip previews the reasoning starts to fail. Though they supplement Disney’s trailers, they complement the original work. I don’t get how, just because the clip previews do not explicitly “add additional or critical evaluative information,” they aren’t complementary to the movie. I don’t understand why Disney’s entering the market first makes the use non-transformative and without a separate purpose.

Part of this confusion stems from the conflating of the first and fourth factor. The ruling makes sense to me as far as the fourth factor goes given current caselaw. As we saw in, entering any derivative market that a copyright holder normally occupies will go against a fair use claim. But this shouldn’t factor into the first factor, the purpose and character test. That test has to do with what type of use, not what the market impact will be. [added:] Indeed, commerciality does impact the burden for the fourth factor (even though commerciality should not imply actual damage to the copyright owner), but that’s ultimately beside the point when analyzing the purpose of the fair use in relation to the original work. Since the potentially fair use is of the original movie, the proper comparison is to the original movie, not the trailers. So when you put conflation of the factors together with the conflation of the original movie and the trailer as well as the trailers with the clip previews, you end up with a total mess.

Look at it this way: let’s say Disney hadn’t entered the market yet and that Video Pipeline won the case. So Video Pipeline is now distributing original trailers, and Preview Maker comes along and starts distributing original trailers which are substantially unlike Video Pipeline’s. Now, as we know Disney would lose to Preview Maker just as it lost to Video Pipeline. But, Video Pipeline now brings suit against Preview Maker. Video Pipeline would most certainly win given Judge Simandle’s reasoning.

Does that make any sense? We want to determine who makes trailers based solely on who moves into the market first? Is the proper comparison for Preview Maker to Video Pipeline, not Disney? (Matt says that he has since decided that this wasn’t the basis for the court’s ruling: “I have to acknowledge that that reasoning makes more sense than what I was originally thinking, which was that the mere fact of making trailers precludes anyone else from also making trailers.” Given the judge’s reasoning in the first factor section, I actually think Matt’s original sentiment is somewhat correct. I think the judge might have been saying both.)

Again, this doesn’t mean the ruling was totally wrong as a matter of caselaw. Hardly anyone survives the fourth factor when the copyright holder has already entered the market. And since it was  a commercial use using key scenes, I can see why the judge would rule against. He just didn’t make his reasoning as strong as it could be.

As Matt suggests, he would have been better off ignoring Kelly v. Arriba Soft. In many ways, Kelly is an outlier in today’s fair use doctrine. It looks at how the copyright holder would be actually damaged and carefully weighs the interests involved.  The court clearly recognized how Arriba’s images have a dissimilar purpose (first factor) and a complementary relationship (fourth factor) with Kelly’s. The court found little reason to believe Kelly would be harmed and much reason to believe that Arriba’s use would promote creativity:

“The Copyright Act was intended to promote creativity, thereby benefitting the artist and the public alike. To preserve the potential future use of artistic works for purposes of teaching, research, criticism, and news reporting, Congress created the fair use exception. Arriba’s use of Kelly’s images promotes the goals of the Copyright Act and the fair use exception….[T]his first factor weighs in favor of Arriba due to the public benefit of the search engine and the minimal loss of integrity to Kelly’s images.“

Had Judge Simandle actually read this case and used its logic, his approach to Video Pipeline would have been far different.  Fair use cases like Video Pipeline simply treat fair use as a solution to market failure. If you’re making money that could be made by the copyright holder, it’s not a fair use. I hope, if this case is appealed, the judges actually read Kelly.

Is Uploading Infringement?

For those of you who followed this, you’ll know why I can’t resist reposting this brief discussion from pho.  Really interesting technical debate.

Update: Just want to point to this article again, which cites the relevant caselaw that has sometimes said that it is distribution. See this article that counters those cases.

Back and Brown

Back for now. I hope to catch up over the next week, but things are going to be a bit busy over the next 3 weeks or so as I see family and return to school.

Speaking of which, Brown has updated its computer use policies. As you might remember, Brown received many DMCA notices last year and warned its students about its copyright infringement policy.  The new policy is superior to Harvard’s, though it seems a bit harsh. I am still troubled by the requirement of removing files from student’s computers. That might make sense if a file were downloaded, but uploaded files might have been acquired legally.

I hope other schools mimic Brown’s fair use faq (more succinct than Stanford’s) and its explanation of copyright infringement.