You are viewing a read-only archive of the Blogs.Harvard network. Learn more.

DRM, the DMCA, and IP as Property

Read Ed Felten’s recent post on DRM and the comments – make sure to read Cypherpunk’s comment and follow-up with this Unlimited Freedom post. 


Cypherpunk clearly has a point here that you at least have to take these questions seriously in the abstract: If DRM prevents piracy, do its potentially socially harmful effects make it not worth it on balance?  If DRM+DMCA prevents piracy, same question.


It is imaginable that we’d all be, relatively speaking, better off with lessened fair uses, for instance, but also less piracy.  In the abstract, the answer to the questions on DRM are not so clear.  This question could really be further abstracted to: is copyright’s role in incentivizing authors worth it on balance given its socially harmful effects (e.g., monopoly-like effects, harm to follow-on creators, etc.)?  If you think this question is worth considering, then so is the above.


At the same time, when looking at the question in practice, I think the answers get a bit clearer, particularly with respect to the DMCA.  Today, DRM and the DMCA do roughly zero in preventing infringement.  I have discussed elsewhere (in draft form) two scenarios in which DRM might play an effective role. In the “Speed Bumps” approach, DRM and the DMCA might be ineffective or simply not meaningfully beneficial while causing harm.  The “Technological Lockdown” approach might also be ineffective, and it would be grossly more harmful and draconian.


In any case, I doubt Dan Burk would disagree with Unlimited Freedom/Cypherpunk on the abstract point.  In fact, he and Julie Cohen have considered it in great detail.  They suggest that we should place legal limits to curb the socially harmful uses, while allowing the identified socially beneficial uses to continue.


But: does DRM really do much harm?  Unlimited Freedom/Cypherpunk and Branden Cox don’t speak directly to this, but they do implicitly make some judgments on this score.  Branden Cox seems to say that it would be preferable to let copyright owners control uses through DRM and license them as they choose.  In his comparing IP to physical property and the workings of contract law to DRM, Unlimited Freedom/Cypherpunk implies that granting these broad entitlements is beneficial. Tom Bell has made such an argument with regard to DRM’s impact on fair use, and we saw glimpses of this argument in the Bridgeport decision. 


Though Unlimited Freedom/Cypherpunk writes that with regard to DRM, the “blogging world … all too often betrays a complete ignorance of economic effects,” this is hardly an open and shut economic question.  As Mark Lemley explains in great detail, there are many things economically wrong with treating copyright just like physical property.  Contra Unlimited Freedom/Cypherpunk, copyrighted products are public goods, and they don’t stop being public goods when wrapped in DRM; we should recognize how the public good character might change our approach.  Neil Netanel also cogently attacks the “neoclassical” approach to copyright, arguing that it ignores copyright’s important role in contributing to a democratic society.


If you accept that DRM has socially harmful uses, we could take many approaches to curtailing them.  Again, this is not in conflict with taking seriously the questions this post began with.  We could, as Burk and Cohen do, condition use of DRM on allowing certain uses. Tom Bell even suggests that copyright holders be forced to exit copyright and only receive contract law protections for DRM and contracts that control more than copyright allows.   Or, we could simply allow DRM but remove the DMCA, as Felten suggests and I agree with. John Mitchell, a keen DRM-critic, suggests numerous beneficial uses for DRM, and I think those should be allowed to develop in the marketplace; however, the DMCA provides so little in terms of meaningful benefits and so much clear harm that it should be removed.  That would go a long way to protecting legitimate uses impaired by DRM.  Also, I agree with Unlimited Freedom/Cypherpunk that arguments to ban certain technologies because of their potentially harmful uses may be quite knee jerk, but we ought to consider arguments about how the market might not function, like in instances of concentrated market power and limited competition.


In sum: yes, we need to consider DRM and the DMCA’s costs and benefits in their totality; however, they aren’t necessarily beneficial on net, their harm depends on your viewpoint regarding copyright’s function, their benefits in practice may differ from their theoretical ones, and these factors must also be assessed.

Downhill Battle Sampling Project

Downhill Battle has launched “3 Notes and Runnin’,” a new project to protest the Bridgeport decision (via Techlawdvisor).  They’re soliciting 30 second songs that use the George Clinton sample.  Wonder if Clinton and Bridgeport will get upset – that suit was not the first time they’ve gone after someone. 

Tivo-Netflix and Some Legal Questions

Rafat Ali dropped me a line with a question about the Tivo-Netflix deal.  Pointing to this post by Jeremy Allaire, he wanted to know about how first sale might affect licensing for online movies.  Here’s my response:


Good question. 17 USC 109 (the first sale doctrine) is even stronger for movies. Music and computer programs have limitations regarding rental; movies (and books) have no such limitations. While some video stores (e.g., Blockbuster) have entered into licensing and revenue agreements with the studios, any video store can simply buy copies of DVDs and rent/sell them to customers without getting the approval of copyright holders.


However, Jeremy is basically right in his statement: they can’t just take the content off of the DVDs and put it online. Just like we discussed with eBay, first sale only applies to the distribution right, and it only applies to lawfully made copies. Putting the songs online would implicate the right to copy, and the only clearly lawfully made copy is the DVD that Netflix owns. Also, distributing the movies via streams might implicate the public performance right.


As we have seen in the music industry, these definitional issues (copying v. distribution v. performance) can be a huge drag on license negotiations. Just as the first sale doctrine protected rental outlets in the analog world, the mechanical compulsory aided distributors of music. For the digital world, Harry Fox Agency, the record labels, and online distributors have long fought over what requires a mechanical and what falls under the compulsory. See the Section 115 hearing testimony.


We’ve seen some licensing problems with the online movie industry. Beyond the anti-trust suit against Movielink, see the Video Pipeline case, in which a licensed distributor of movie trailers was barred from putting those same trailers online.

Copyright Office Induce Draft Now Available

Better, but still incredibly, superbly rotten.  Look for Ernest to give it the full treatment later.  Short story: send in the accountants! Let’s ask all businesses to open up their portfolios to see where their revenue comes from and base everything on the product’s current uses without paying any attention to potential uses. Oh, and btw, let’s fold non-commercial services into the mix by looking at current uses too.  Note how so much of the analysis deals with the “business model” approach, and then they slip in the line about 501(g)(1)(c) without any further discussion. 

Court Says Banning Sampling Poses No Threat to Creativity

In Bridgeport Music v. Dimension Films, the Sixth Circuit has ruled that there is no de minimis defense to sampling a sound recording and the substantial similarity test does not apply.  The court stated throughout the opinion that a bright line rule that rejected all sampling, no matter how small, should be adopted.  The court did not reach the issue of fair use, however.  (Note that this is different from the Beastie Boys case.  That involved the copyright in the composition exclusively.  This decision focused on the sound recording.) 


In its gross oversimplification, this decision rivals the Biz Markie case, which began “Thou shalt not steal.” Let me give you the summation up front: To this court, the issue is merely a matter of competing market actors, and all we need to do is create an efficient market.  Any threat to creativity is just the market at work.  No attention is paid to the public’s interest.  (BTW, I’m going to treat this a little out of order, to align related parts of the argument.)


The court emphasizes at the outset and throughout that a bright line rule is far and away preferred.   This predetermines the entire opinion, for it would be very difficult to create a bright line rule allowing sampling.  What, would you use the number of seconds sampled? The number of instruments?  A rule based on the de minimus or substantial similarity defenses certainly couldn’t (and shouldn’t) allow all sampling. 


The court provides many reasons for going with this bright line rule, but the most important lies in this paragraph: “To begin with, there is ease of enforcement. Get a license or do not sample. We do not see this as stifling creativity in any significant way. It must be remembered that if an artist wants to incorporate a ‘riff’ from another work in his or her recording, he is free to duplicate the sound of that ‘riff’ in the studio.”


For the moment, let’s get over the shock at arguing “We do not see this as stifling creativity in any significant way.”  For now, the important point is that the court is arguing in favor of creating a clearly defined property right, which will lead to the efficient allocation of resources. The court makes this clear in its next statement: “[T]he market will control the license price and keep it within bounds.  The sound recording copyright holder cannot exact a license fee greater than what it would cost the person seeking the license to just duplicate the sample in the course of making the new recording.”


Once a market is created, an efficient result will be produced.  Apparently, creating a property right in this way will lead to a sufficiently creative environment.  First of all, this argument about the license fee’s ceiling assumes that the “riff” is de minimis, insubstantial, or otherwise a fair use, because the covers compulsory expressly states that “the arrangement shall not change the basic melody or fundamental character of the work.” Thus, as long as the composition is under copyright, sampling a subsection of the composition might infringe.  More importantly, all this assumes that the applicable tests and defenses that usually apply are merely ways around market failure and not a way to provide breathing room for other creators.  This also ignores the non-economic incentives not to license (e.g., sampling is a threat to the artist’s integrity) and the socially inefficient pricing for uses with social benefit outside what the licensor (and licensee) can capture in monetary form (e.g., satire) (See Mark Lemley, “The Economic of Improvement in Intellectual Property Law,” 75 Texas Law Review 989). 

In this way, the purpose of copyright – incentivizing creativity for the public’s benefit – doesn’t easily square with this approach.  Indeed, the fact that copyright law uses a fact-specific, balancing approach, rather than a bright line rule, in other derivative works contexts means nothing to this court.  Sure, as the court notes, balancing tests are costly and time-consuming for the judicial system.  And, sure,  the court’s right that it’s the first to consider digital sampling of a sound recording and thus it’s creating a new rule.  But it’s not as if courts have never had to deal with these complicated questions. In fact, when treating this thorny area of copyright law, courts have figured out a way to deal with these questions in the form of the de minimis and substantial similarity tests. The court does not fully consider the illogic in requiring this ease of enforcement for sampling a recording, while using the fact-specific tests for sampling from a composition.

To try to explain this illogic, the court makes a less-than-satisfactory statutory construction argument.  The court takes the statute (“The exclusive right of the owner of copyright in a sound recording under [the section 106 right to prepare derivative works] is limited to the right to prepare a derivative work in which the actual sounds fixed in the sound recording are rearranged, remixed, or otherwise altered in sequence or quality.”) and argues:


“The import of this language is that it does not matter how much a digital sampler alters the actual sounds or whether the ordinary lay observer can or cannot recognize the song or the artist’s performance of it. Since the exclusive right encompasses rearranging, remixing, or otherwise altering the actual sounds, the statute by its own terms precludes the use of a substantial similarity test.”


It’s a possible reading, I suppose, but not the only or best reading.  Congress was expressly limiting the right, and it didn’t necessarily abrogate any typical limitations to infringement actions.  Literal copying can be insubstantial or de minimis.  Consider the rights of a book author.  Though the Copyright Act does not explicitly state this, the derivatives right implicitly gives him the right to rearrange, remix, and otherwise alter the actual words he wrote.  Does that definitionally mean quoting a single sentence would be infringing?  Not necessarily. (See 4-13 Nimmer on Copyright

A Little More on Skylink

You know, it’s still a lot to wrap my head around.  I would really like to first go back and read many DMCA articles and all the cases a little closer.  But this tentative assessment is going to have to do for now.


Prelim Points


First, I’m just going to assume you’ve read at least Ernest’s and Ed Felten’s posts – I’ll get to some of what they bring up, but not all of it.  The test that comes out of the court, more than simply allowing access devices or circumventions with purely non-infringing uses, created a test that examines whether the device or circumvention has a reasonable relation to a copyright-right. 



“A copyright owner seeking to impose liability on an accused circumventor must demonstrate a reasonable relationship between the circumvention at issue and a use relating to a property right for which the Copyright Act permits the copyright owner to withhold authorization-as well as notice that authorization was withheld. A copyright owner seeking to impose liability on an accused trafficker must demonstrate that the trafficker’s device enables either copyright infringement or a prohibited circumvention.”


Second, I can imagine someone arguing that, in reality, it’d be better to let the DMCA be as draconian as it is on its face; that way, it’ll be more likely that Congress will be forced to confront the DMCA in all its horribleness.  I guess it’s a fair argument. On the other hand, if the examples of the DMCA’s furthest overreaching and most unintended consequences become the basis for Congressional reworking, we might just end up with a slightly narrower statute.  In my view, it’s better to have this breathing room than nothing at all.


Problems in the Opinion


There are several ways in which the decision’s argument is lacking, even putting aside the clever, and sometimes misleading, ways they find this consistent with Reimerdes. Ernest notes a point of poor crafting, like deeming Reimerdes about an injunction for illegal copying, but a lot of the other stuff is just the judges being a little too clever.  I’ll get back to this piece more below.


As discussed by Ernest, the decision lacks a clear distinction between copy and access controls. Indeed, it seems they’re now almost one in the same.  The trafficking provisions now relate to devices that enable “copyright infringement” (1201(b)) or a “prohibited circumvention” (1201(a)).  Well, a prohibited circumvention is now “a use relating to a property right for which the Copyright Act permits the copyright owner to withhold authorization” – which is basically another way of saying a potentially infringing use. 
What the court tried to do was ensure that in all cases there’s a nexus between the device or circumvention and infringement.  That’s a good thing, on some level.  On the other hand, it seemingly renders meaningless the separate, non-infringing related meaning of 1201(a).


Is there an alternative way they could have limited 1201(a) and kept the distinction intact?  Ernest has argued as much in the past.  I think the Skylink court put so much focus on the DMCA’s purpose of preventing piracy that, they missed the sort of “piracy,” e.g., black boxes for stealing cable, that 1201(a) was aimed at.  With Skylink’s reading of the DMCA, I’m not even sure when 1201(a) would reach that sort of act.


Also, if the DMCA was about rebalancing, why do they draw so much weight from 1201(c) and so little from all the statute’s exemptions, the Copyright Office triennial review, and the lack of provisions banning circumvention of copy controls?  Didn’t Congress try to keep the balance intact that way?  Why is this test required on top of that given Congressional intent?  I’m not saying that they’re wrong in drawing on 1201(c) this way, but this is a decent counterargument, I think.


A smell test with some broader applicability


Like the District Court, this court wanted to figure out a way to correct for Congress lack of clarity and foresight.  Unlike the District Court, this court actually came up with a test that makes some sense, is decently reasoned, and has applicability beyond this case.  No, it does not solve all the problems with the DMCA, but it does open the door to dealing with many of the most abusive, unintended consequences.   They focused on the DMCA as targeted at impeding copyright infringement.  When discussing the district court opinion and the potential for bringing a sort of misuse view into the picture, Doug Hudson pointed  out it’s not completely uncommon for courts to try to come up with balances when statutes are unclear like this.


Some people have put a lot of focus on the consumer expectations issue, saying that was the key. I disagree.  I think the court agreed with the district court’s assessment there and kept that door open for future cases. But, as the court clearly states, the key issue was the relationship between the circumvention at issue and a copyright-right.


Anti-trust and Misuse are back in the picture


With the importance of the balancing test, this section has been overlooked:



“Chamberlain’s construction of the DMCA would allow virtually any company to attempt to leverage its sales into aftermarket monopolies-a practice that both the antitrust laws, see Eastman Kodak Co. v. Image Tech. Servs., 504 U.S. 451, 455 (1992), and the doctrine of copyright misuse, Assessment Techs. of WI, LLC v. WIREdata, Inc., 350 F.3d 640, 647 (7th Cir. 2003), normally prohibit.  Even were we to assume arguendo that the DMCA’s anticircumvention provisions created a new property right, Chamberlain’s attempt to infer such an exemption from copyright misuse and antitrust liability would still be wrong…. Because nothing in Seventh Circuit law contradicts Data General, we similarly conclude that it is the standard that the Seventh Circuit would most likely follow. The DMCA, as part of the Copyright Act, does not limit the scope of the antitrust laws, either explicitly or implicitly.” 


That leaves open the door to further arguments against particular uses of the DMCA.  Misuse, esp., is pretty vague, but particularly relevant with regard to tying content to devices.


Also, though the court doesn’t reach the contracts issue, it’s worth noting that this is the same circuit that decided Bowers.  While having no bearing on the issue directly, the Skylink court’s general sentiments were very similar to the the Bowers dissenter’s:



“There is, moreover, no logical stopping point to the majority’s reasoning. The amici rightly question whether under our original opinion the first sale doctrine and a host of other limitations on copyright protection might be eliminated by shrinkwrap licenses in just this fashion. See Brief for Electric Frontier Foundation et al. as Amici Curiae 10. If by printing a few words on the outside of its product a party can eliminate the fair use defense, then it can also, by the same means, restrict a purchaser from asserting the “first sale” defense, embodied in 17 U.S.C. 109(a), or any other of the protections Congress has afforded the public in the Copyright Act. That means that, under the majority’s reasoning, state law could extensively undermine the protections of the Copyright Act.”


Now consider Skylink’s words:



“It is unlikely, however, that the Second Circuit meant to imply anything as drastic as wresting the concept of “access” from its context within the Copyright Act, as Chamberlain would now have us do. Were

BMI (not BMG!) Has Record Year

The movie industry had a record summer at the box office, despite a decline in ticket sales.  That’s pretty predictable, since broadband is not pervasive and watching a DivX download is not nearly as high quality as going to the theater.  So it’s not worth making too of this. 


It’s also not worth making too much of recent news that BMI has had a record year.  Maybe people are getting BMI confused with BMG.  The latter is a record company, the former is a performing rights organizations.  The former’s revenue is pretty small relative to the total revenue of the record companies.  These results are consistent with last year’s growth, spurred on in part by the ringtone bubble.  This revenue stream will likely be hit more slowly; although radio will lose relevance, licensing for TV will not fall off as quickly, and restaurants, live concerts, and other public places will still pay fees.


Which is not to say I agree with the “sky is falling” rhetoric of the music industry.  But we should keep these stats in perspective.

MS: On Second Thought, Put On These Handcuffs

Farhad Manjoo has a nice article up about the new MSN Music Store (via Frank).  At the very end, he notes a somewhat predictable turn of events: MS changed the language, noted by Fred, that advocated burning DRM-locked songs onto CD and then re-ripping into another, possibly unencrypted format.  It now reads simply:



“Unfortunately Apple refuses to support the popular Windows Media format on the iPod, choosing to only support their own proprietary DRM format. If you are an iPod owner and are unhappy about this, please send feedback to Apple and ask them to change their policy and interoperate with other music services. There are more than 70 portable audio devices that support MSN Music today, and we hope that someday Apple decides to join with the industry and support consumer choice.”


70 devices: wow! I mean, Microsoft has picked them all out for me ahead of time, probably put a little red bow on them, too.  What would I need interoperability and independent consumer choice for when Microsoft can shepherd me to approved devices?


Apparently. the businessmen had a chat with the techies.  I wonder if anyone raised their hand and reminded others that Windows Media DRM is also proprietary. Manjoo contacted Rob Bennett, the senior director of MSN Entertainment, who wrote back:



“I’m reviewing the language on the preview site now…. We absolutely don’t want to encourage people to circumvent the usage rights for music downloads. It is unfortunate that Apple still disables Windows Media support in the iPod (the firmware they license from PortalPlayer actually supports WMA but they turn it off), restricting their customers’ choice of where they download music. Our approach is very different, encouraging broad choice of many music services and many portable audio devices with the Windows Media format.”  

Apple v. Real v. MS

I had the pleasure of appearing on Ernest’s radio show with Jason Schultz, Rafat Ali, and Mike Mesnick this afternoon.  The topic: the competition between Digital Music Stores and control over platforms using DRM.

Skylink Linking

Jetlagged and busy, I can’t do a thorough enough job on Skylink yet.  Apologies.  I agree with many of Ernest’s points, though I think I’m happier, or at least more optimistic, about the door being opened for companies to successfully defend themselves under the DMCA.


For now, I want to direct your attention to a couple noteworthy docs.  First, read Dan Burk’s “Anti-circumvention Misuse,” which suggests an approach somewhat broader than the Skylink court’s.  Second, check out R. Anthony Reese’s article on the merging of access and copy controls.


Third, go back and read the discussions after the district court ruling.  In particular, follow the back and forth between Ernest and Doug Hudson (doogieh).  Hudson provides important counterpoints to Ernest’s take on the consistency of the opinion, and his points are even more germane now.

It’s a Kind of Magic

Numerous posts out there about Valenti’s latest interview silliness.  What’s most sad and most enlightening about the interview are the technical non sequiters that Prof Felten points to.  The security design process is apparently just about “stuffing … algorithms,” really cramming them in there, and then hurf, thunk, clatter, out pops DRM – “technological magic.”


Valenti is always colorful and uses simplifying rhetoric.  What I hear most clearly, though, are the DRM vendors speaking through him.    As I have experienced, DRM vendors speak just like this about security.  They go around to conferences and board meetings, hocking magic pixie dust and hornless unicorns.


So Valenti didn’t invent this view of DRM, but he’s not just innocently following the vendors – that is, he’s guilty of a sort of “willful blindness.”  The DRM vendors give Valenti the excuse not to listen to the actual experts.  DRM vendors will always tell him what he wants to hear.  Real security experts will not.  Thus, the latter are just dirty hackers, while the former are wonderful magicians.


When people choose not to listen, there is no convincing them, and thus I’m not optimistic about convincing Valenti’s kind.   For awhile, they won’t listen to these arguments about DRM.  They will, however, listen to their bottom line.  As they continue to give their money to DRM vendors, and piracy does not decline, they’ll listen to the fly buzzing out of their wallet.  Then, instead of an argument between magic and reality, we’ll just confront the reality, and argue from there.  That’ll be the day.


(This is not to minimize the potential importance of other reasons the content industries use DRM.  But I do think that they actually believe in DRM’s efficacy in preventing piracy, and that remains a major impediment in their thinking and in the larger debate.)

Viva La Revolution!

Some new potentially illegal art, designed by a friend of mine: The Che Moore T-shirt.  Let’s count the rights it might violate.

With DRM and the DMCA, Nothing “Plays For Sure”

PaidContent points to reports that the new Microsoft Music Store will debut September 2.  It seems like we’ve been hearing rumors of this forever (or, at least as far back as February).  Apparently, the Store will integrate with Microsoft search and IM and will debut with over 600k songs at 99 cents each.  Given the coming of Janus, a subscription service is no doubt on the way as well.


Microsoft is also about to start a massive Windows Media marketing campaign, centered around its new “Plays for Sure” compliance mark.  Paidcontent puts it, “Microsoft is planning to tout its interoperability (some would say it is just about ubiquitous monopoly).”  Well, monopoly may be a little strong, but ubiquity really is the key.  Go back to Cory’s response to Scoble.  To rephrase SethS’s comment, interoperability is not the same as ubiquity.  You still have to ask MS’s permission to create interoperable software and hardware. 


With DRMed digital media, backed by the DMCA, nothing plays for sure.  Please, somebody start THAT campaign, rather than playing these silly games.  Your digital media is forever tethered to the DRM owners and relevant copyright holders.  Your digital media plays the way they say it can be played, that’s for sure.  But you will never truly be able to use your digital media however you want on whatever device you want – we will never see true interoperability.

If the FCC is going to lock down the airwaves…

…then Ernest is going to rock web radio.  He’s got a new series on ITConversations called The Importance of the Law and IT.  The first edition is about MGM v. Grokster, starring C.E. Petit, FvL, Denise Howell, and Tim Wu.  I haven’t heard it yet, but I don’t need to to recommend it.

Too Many Great Posts

Lots of news.  Ernest and Copyfight have great coverage of the Don’t Induce Act as well as JibJab’s victory.  Prof Felten has two nice posts on innovation.  Frank has plenty of links up, with some particularly interesting ones involving colleges and digital music services.  And Judge Posner is blogging up a storm over at Lessig’s.

« Previous PageNext Page »