Picker MobBlog

Felten: Diverse Users, Diverse Uses

Hey, where is everybody?  I guess I missed the party.   I would have been on time, see, but I put this new CD in my computer and it turned out to be copy-protected, and I had to spend the rest of the week in the lab.

Most of the brilliant things I was planning to say have already been said by others, so I'll be brief.. 

Julie is right to argue that users' interests deserve more attention.  A big part of the problem, I think, is that copyright arguments tend to underestimate the diversity of users, and the diversity of uses of copyrighted works.  The tendency instead is to postulate some notional average user, who makes some limited set of uses, and to attend only to that average user's interests.  We see this, for example, when broadcast flag advocates argue that the flag isn't harmful to consumers because it still allows some limited forms of time-shifting.  We see it, too, when a judge rules that the DMCA isn't a fair use problem because the first few fair uses he can think of are still possible with analog media.

The diversity of uses has grown vastly with the advent of digital technology, and it will continue to grow.  So continuing to ignore diversity will only lead us more badly astray.   Ignoring new digital uses is quite common in copyright discussions (present company excepted, of course), which helps to explain why copyright has such a bad name among technologists and early adopters.

Which brings me to Tim's suggestion that we should treat users' choices as presumptively good.  Who knows the most about the universe of uses?  Individual users, that's who.  As usual in a free society, we should start with the presumption that citizens can make choices for themselves.

The Economics of Situated Users

So if Joe Liu is late to the party, I'm at the afterparty.

I enjoyed Julie's wonderful essay and the many comments. Julie is on to something that many of us think -- that there is something missing, and something important, in how copyright thinks of its users.

Here's how I'd state some of what Julie is talking about. I see the copyright law is preoccupied with battles between distributors, with the occasional authorial interest thrown in. It is sometimes reminiscent of 1960s antitrust -- a law trying to protect favored, yet dying, entities, with little or no attention to user welfare. Consequently, the public is often disserved by the law's preoccupations.

But where I may part company slightly, is in two areas. First, I think I agree with Larry S. -- on first reading, the concept of a "situated user" yielded, to my perhaps too-rational choice-infected brain -- a unclear way of thinking about these problems. For me, the contribution of the "situated user" essay is to capture real user interests missing from copyright's calculus. So while Julie points us in the right direction, but I would use different tools.

What tools? In short, I think the economic criticism of copyright's approach to users is powerful and (at least to me) clearer. There are two basic reasons. First, if you think consumer welfare matters, what a user or consumer wants to do ought presumptively be what users want. Second, economic doctrine is not so narrow as to be unable to capture the user interests Julie discusses.

Consider the following economic critiques of copyright's approach to users:

1. Copyright often neglects the user as our best proxy of what serves the public interest. For example, the best steward of IP within the users' control is often (surprise) the user. That's the instinct behind the first sale doctrine, for example, but the point is deeper. What the user decides to do with property in their control, I think we should presume, is what will in the aggregate be in the public's interest. (Note to Jim -- this, I think, is also the strongest argument for the principle of NN). That's a rebuttable presumption -- rebutted, in the main, by collective action problems. Mass copying might, for example, ultimately hurt users. But that what the user wants to do is good should be the presumption.

2. Copyright is also too insensitive to the link between users' freedoms and innovation (as Fred has already said). Decentralized innovation and product development has a far better track record than centralized. Copyright ignores the extent to whichusers' interests can be necessary to decentralized innovation.

3. In addition to the non-transformative uses discussed above, Copyright is also insensitive to a range of productive uses that seem to only marginally affect the central problem of market failures for authorship. This is related to point #1 -- that what the user wants to do is usually good for all of us. But it reminds us that have long thought that, for example, the economic case for the broad derivative work doctrine is weak and perhaps non-existent.

Stated otherwise, I have more faith than Julie that the economic user is a useful and not narrow concept, and that a better economic view captures more than simply the fact of consumption.

Tim Wu

Cohen: Many Thanks

I have really appreciated everyone's comments.  It has helped me think in greater detail about additional work to be done in framing and explaining the argument.  Thanks also to Randy for hosting the discussion.

I am leaving town tomorrow a.m. and won't have reliable Internet access tomorrow, so I'm signing off.

best wishes to all, Julie

Tushnet: Users and Consumers

Rereading the introduction to Julie’s paper, with its discussion of the choice of the term “user” over “consumer,” made me think again about Pamela Samuelson’s recent fascinating talk at the Washington College of Law, where she gave a keynote address celebrating the Glushko-Samuelson IP Clinic. She discussed copyright’s consumer protection functions, both current and prospective – the ways in which copyright is formulated not just to encourage and reward authorship, but to protect consumer freedom in small (first sale) and large (anti-monopoly, copyright misuse) ways. She convinced me that adding “protection” to “consumer” has rhetorical benefits in insisting on the dignity and legitimate interests of consumers. This would be more useful if we were in a political climate that was friendlier to consumer protection, but you can’t have everything.

As Samuelson’s talk went on, she ended up defining almost every limit in copyright as a consumer protection measure, including provisions that privilege libraries and schools. The major class of limits that Samuelson didn’t mention was the compulsory licenses that regulate business-to-business relations. Yet if, as she suggested at the talk, Sony is a consumer protection case because there Sony stood for end users’ interests and the library exceptions are consumer protection exceptions because the libraries help end users get access, I’d think the compulsory license provisions that allow mechanical reproductions, the antitrust decrees under which ASCAP and BMI operate, and the compulsory cable and satellite licenses would also qualify as consumer protection measures, since they also are intended to aid in both greater access and more equitable access to copyrighted works.

So, while I’m now convinced that we (copyright skeptics or minimalists) shouldn’t abandon the term “consumer” to the extent that it helps us make links with other doctrines that explain why infinite rights expansions for copyright owners are bad, I also think that it doesn’t commit us to specific rights, limits on rights, or privileges. Indeed, the question of what rights are good for users or consumers or what-have-you can always be answered with a copyright maximalist story, too: the market will provide, and the incentives of ownership will cause so much production that consumers/users will have more choice and more utility overall. (I think of this as the copyright version of the Laffer curve, but maybe that’s just me.)

There’s a structural parallel with consumer protection law – false advertising law finds nothing objectionable about suppressing information that’s useful to some consumers if it’s misleading to a substantial number of other, less sophisticated consumers. In other words, groups of consumers (and users) can have competing interests, something that Julie points out in the last paragraph of her piece. One argument for competing interests of users as users has been made by Landes and Posner, and earlier by Justin Hughes, that some audiences may have interests in fixed and non-reinterpreted versions of copyrighted works that are harmed by fair use, limits on the derivative works right, and entry of works into the public domain. I think this is silly, for reasons explained here, but the point is that identifying a particular type of person about whom you’re concerned, whether it’s author or user, is just the beginning of the analysis – at least for a consequentialist.

It’s here that Julie’s focus on process might help us skeptics and minimalists. By emphasizing the uncertainty and contingency of the processes that lead us to find and produce meaning – even if some parts of the production are not accidental, as David McGowan correctly points out in the first comment here – the concept of the situated user can push us to think about copyright’s dynamic effects in more than the basic “create incentives but don’t close off so much that future authors can’t create” way. But I’d also love to have Julie say more about the ways in which she sees users’ interests conflicting – is it just the Posner/Landes/Hughes story of conflict?

Romantic Users and DRM

Julie's paper raises numerous interesting and important issues, but let me focus on one that has yet to be discussed--how to think about and critique DRM in light of Julie's analysis.

The genius of analog fair use cases (Julie, sorry for belaboring this point), and the missing ingredient in most DRM approaches, is that it focuses on the particulars of the user at hand.  In the DRM context, however, courts--such as the Second Circuit in the DeCSS case--struggle to identify some archetypical user and judge the importance of circumventing copy protection systems accordingly.  In the DeCSS case, Lessig and Benkler made their First Amendment argument based on the case of a romantic user--a ten year old student using a snippet of Schindler's List in digital form to examine the horrors of the Holocaust.  Hollywood countered with the worst form of an economic user--the digital pirate.  In the end, the Court adopted only a temporary answer--that with the availability of analog fair use opportunities, romantic users still had an opportunity (albeit with an inferior technology) to do their thing and thus there was no constitutional harm to precluding digital copying of otherwise protected content.

For DRM solutions to begin to approximate the flexibility of analog fair use opportunities, they will need to recognize the implications of the situated user--distinguishing, for example, between the user who edits a movie to place herself in the title role for her friends and the person making that same movie available for widespread copying.  As many have emphasized, stopping all copying would stiffle creativity and cultural discourse.  Not to mention, it also, as Sony recently learned, pisses off customers of all kinds.

Cohen: Performance; Profit?

Rebecca's point about performance is important, and links up to the comment I made in my opening post about the dimension of physicality.  An example that I mentioned was manipulating copies (so, the situated user asserting control over the cultural artifact, whether a physical object or a digital file).  Performance might be understood as inserting oneself *into* the "work" - what's important here is not the physicality of the artifact but the physicality of the self, and the ways in which performance alters the experience of the work and becomes another way of working "through culture."  In either case understanding the "work" (there's a reified construct for you) as the primary unit of analysis misses something vitally important about the process of working through culture.

If one accepts the premise that copyright should respect performance as a mode of experience even when it's "public" in the technical sense, the question remains whether it should matter that there was a charge for admission to the Buffy show.  (I suppose one other possibility is that the admission charge for Buffy covered only expenses, much as one can make photocopies in the library and still have to pay for paper.)  This seems conceptually related to the question whether it should matter that the creator of the SAT (the Seinfeld trivia guide) later turned a profit on it.

I haven't yet reached a comfortable resolution on the profit question, but I'm inclined to think that (although a for-profit cutoff might be politically more palatable) profit shouldn't matter.  I suspect there may be more disagreement here.

Rebecca Tushnet here.  I'm still working on some other comments, but Fred shamed me into acting now.  Also, I should disclose that I had already read and commented on an earlier draft of Julie's paper before this workshop started, so my comments may be more marginal.

The Buffy musical performance is a great example for me, because it is all copying and all innovation, like any performance. Fred suggests that there is nothing “remix culture” here, but I would argue that the relatively long pedigree of the performance right, combined with our now-default image of a “performance” as a recorded performance, has obscured the extent to which any performance is a new version and thus can be identified as a product of the romantic author/user. There’s a reason the counterPULSE theater folks decided to stage “Once More With Feeling” (OMWF) rather than project the DVD onto a big screen, the same reason people snapped up the tickets – because the performance would have put a different spin on the well-loved words, possibly offering new insights and certainly offering much fodder for comparison. As a friend of mine who teaches a Buffy studies class points out, staging what was a teleplay is particularly helpful for teasing out important differences between theater and film; for example, where the TV Buffy could show parallels between characters using cuts and other editing techniques, the stage Buffy cannot, but can use space – having one character stand exactly where another character did in a previous scene – in a way that’s less available to a filmed version. Likewise, fan fiction is often about retelling the same story, only different this time – a mix of the familiar and the new made interesting by being together. (Side note: this fan video (.avi), may not be entirely accessible to non-media fans, but it contains some wickedly funny commentary on copy-protection schemes and what a fan might do in response.)

What does that mean for the situated user? I’m not sure, but I’m increasingly convinced that “performance” is a relevant concept. Sonia Katyal is working on a paper using performance theory to show how copyright doctrines (particularly – sorry, Julie! – fair use, with its privileging of the original and the savage parody and its intolerance for the valley between the two) have distributional consequences for various user groups, including fan fiction writers. Performance theory could be used to talk about how people create coherent narratives for themselves out of the bits and pieces they find around them, or, directed more specifically at copyright’s exclusive rights, it could be used to help justify limits on things like the performance right.

Going back to Buffy, as I often do, I want to make two final points: First, Fox was apparently unprepared to license the theater’s performance; that is, not only did it not have a price for a public performance that didn’t involve playing the DVDs, it didn’t have a mechanism set up to set a price, because the theater was doing something so odd in the context of television. Second, Fox does sell the Buffy script books to the mass market (there are novelizations and spin-off novels and comics too, but here I’m speaking of the straight script books, spelling errors and all). As I recall, the OMWF script book includes musical notation as well, so that one could play the music on one’s own instruments. The theater simply proposed to use those script books as script books rather than as fan artifacts, the same way the original TV producers used the scripts. I think both of these things are related to the fact that Joss Whedon is one of TV’s few auteurs; he has been recognized as such by the fans themselves/ourselves.  There is a market for script books and new performances of OMWF the way there is not a market for the same things for, say, "CSI." And that to me confirms that the romantic user is structurally dependent on the romantic author, the person who creates something popular enough that fans can find each other as audiences by identifying as fellow fans.

von Lohmann: More Situated Users

Randy asks for concrete cases where the situated user concept yields insights distinct from traditional doctrinal approaches. Joe suggests fanfic. But I worry that fanfic bows more in the direction of the romantic user than the situated. Bill (obliquely) suggests that "guidebooks" for obsessive fans might be something we would understand better from the perspective of the situated user. I agree with him -- I think the concept of situated users best distinguishes the two cases he discusses.

But let me try two additional concrete examples.

First, there was a recent incident here in San Francisco involving an effort to stage "Once More With Feeling," the one-and-only musical episode of Buffy the Vampire Slayer. (Here, of course, I am chumming the waters for as-yet-silent Mobber Rebecca Tushnet, a fellow Buffy fan.)

It's vaguely embarrassing to admit that there was a general buzz of excitement at SFist HQ when SFist Jon found out that counterPULSE Theater was staging a live-action version of the Buffy The Vampire Slayer musical episode, "Once More With Feeling." Set for performances this weekend and next, tickets sold out pretty much immediately and geeks of all stripes (musical theater geeks, comic book geeks, Star Trek geeks, Joss Whedon geeks, and your everyday run of the mill geeks) began lovingly sheathing their precious original tickets in durable plastic covers to preserve forever.

Well, where geeks go, lawyers are soon to follow -- last week, counterPULSE got an angry letter from Fox TV's attorneys, demanding that they cancel all the shows or face a copyright and trademark infringement lawsuit. Feeling as if they had no choice, counterPULSE has now cancelled all the Buffy shows (which they are now referring to as "the Halloween shows," or the "Uffy" shows).

Traditional copyright doctrine views this as largely an open-and-shut case of infringement. The economic user has little to add, beyond noting that there is a market for this sort of thing (see, e.g., the Gilligan's Island musical toured for 12 years!) and that the market belongs to the rightsholder (not the creator -- the creator of Buffy, Joss Whedon, gave his blessing). The romantic user doesn't much care, as there is nothing "remix culture" here. This was fans trying to be slavish to the original (though we all wondered how they would handle the oblique lesbian sex reference that was in the episode!). Ditto the post-modern user. I suggest that each of these accounts is missing something important, something that the fans who put on the show and who bought the tickets understood viscerally about participatory culture, about creative play.

Example #2: We have recently witnessed a flurry of new announcements about authorized, for-fee, "on-demand" ways to watch current television shows. Apple is selling last night's episode of Lost for $1.99. Comcast and DirecTV have announced partnerships with CBS and NBC to do similar things.

What does this mean for the future of unauthorized home time shifting using TiVos, Windows Media Center PCs, and MythTV boxes? The economic user says that what was once a fair use should gradually be restored to the rightsholders as transaction costs dissipate. The romantic and post-modern users only care indirectly, to the extent that the DRM associated with these new offerings may impede "remix culture" or "semiotic disobedience" (again, tip o' the hat to Sonia Katyal for the latter coinage).

I'd suggest that if you tried to take away people's TiVos and force them all to pay to watch last night's episode of the their favorite program, you'd see a backlash that was not simply about an unwillingness to pay. I'd also suggest that this development would have important implications for the future course of innovation in the home media technology realm, as it would in effect return us to a "pre-Hush-a-Phone" world where only approved devices were entitled to connect to our television culture.

Liu: Participating in Shared Culture

I’m a little late to the party, but I want to build on Randy’s comment on “Implementing Shared Culture,” because I think it highlights a distinctive facet of Julie’s paper. The first two examples in the post (Dr. Seuss and Seinfeld SAT) fit into our standard author model, and I’m not sure the idea of a situated user adds much more to the analysis (though Julie seems to disagree, as noted in the comments to the post). As Randy points out, the situated user plays a potentially greater role in the third example (fan fiction).

And on that score, I think Julie’s paper usefully directs attention to the potential role that smaller-scale users play in implementing shared culture. Alice Randall and Tom Stoppard play a role in implementing shared culture. But so, too, do the folks writing fan fiction, as well as those putting up personal web sites with appropriated materials, sharing music clips with friends, creating short star wars parodies, etc. Individually, the contributions may be trivial, but in aggregate, they play a significant (and I would argue increasing) role in implementing shared culture (particularly the diversity of that culture).

With these smaller-scale appropriations, the issue is not so much fair use (these uses are individually so trivial they will never be litigated), but rather the more systematic barriers to users implementing shared culture. For example, the prohibition on anti-circumvention tools raises the costs of participating in shared culture. If I want to post a digitally-altered clip from the latest Star Wars movie, the DMCA makes it more difficult to do so. (Not impossible, just more costly). Reflexive enforcement of shrinkwrap agreements might also raise such costs.

So, to take Julie’s suggestion to expand beyond fair use, perhaps the situated user perspective would lead us to be more concerned about the impact of the DMCA on the ability of individuals to participate in shared culture. We might ultimately conclude that the impact is not that significant, or that it is outweighed by a greater need to prevent large-scale infringement. But the broader point is that right now, the situated user perspective is largely missing from the debate. Having said all that, I do have some questions about the situated user construct, but I’ll leave those to another post.

Speta: A Telecoms Analogy (and a Question)

Picking up on Fred von Lohmann’s comments, the important thing for me is the way in which the paper advances the conversation about how users actually act, and of the value (economic and otherwise) of what they do.  I see an analogous debate in telecoms.  In the past, one could see the Hush-a-Phone revolution as a parallel to what Julie is talking about.  Prior to this, AT&T controlled  both the telecommunications network and the equipment attached to the network -- and, by extension, the uses of the network.  And regulation was almost entirely focused on the carrier -- were the rates reasonable but sufficient to allow a return on investment (etc.).  The idea of the “user” was someone receiving the service the carrier offered.  Regulation ensured (or tried to ensure) that the user got this at a reasonable price, but that was (largely) as far as the user was considered.  The FCC's Computer II rules codified what the D.C. Circuit began in the Hush-a-Phone case:  a rule that the user of the telecommunications network could attach to the network anything it wanted (such as a fax machine or a radio extension), as long as it didn't harm the network.  This injected the "user" into the telecoms debate and allowed the user to make use of the network for new purposes.  In Julie's terms, it allowed users the freedom to "play" with the network (or at least the bandwidth provided by the network).

Today, the analogous debate is over "network neutrality" rules -- rules that would codify the rights of broadband users to use their connection for any application they wished to run (for example).  (See Tim Wu's paper.)  Advocates of these sorts of rules talk about play and innovation and speech – things very similar to the values Julie is talking about.

Now, I'm not an advocate for network neutrality rules, because I think, in the current environment, they are more costly than beneficial, and because network owners should have lots of incentives to allow users to create new ways to use the networks (because that increases demand for the network).  But, and here's the interface with Julie's paper, I tend to think that network neutrality rules make more sense the more one is convinced of the persistence of market power among network providers (although even then there are good arguments against).  So, the predicate question (for both Julie and Fred) is why does the copyright system systematically interfere with users' ability to play?  Copyright may sometimes confer market power vis a particular work.  But there is lots of copyrighted and non-copyrighted material that might be substituted and lots of market practices that might allow play.  To take only one, well-known example, Lucasfilms seems to have come to the place where they allow and encourage fan films, which are clearly inside the derivative works right, so long as they aren’t commercialized.  For another, think of all the ways that on-line games (worlds) allow users to customize their persona and even environments.