A Copyfighter’s Musings

[updated 2/26] 321 Studios lost badly.  As SethF suggests, the worst of it was the complete rejection of the constitutional argument based on Eldred. See here for earlier discussion on that argument. 

As Seth said and I agreed then, unless the courts construe the argument broadly, they will be able to fall back on Corley’s technological-inconvenience-is-not-an-excuse argument, which is precisely what happened here.  The meaning of Eldred as read through MGM v. 321 is that the government may limit fair use (as guaranteed by the Constitution) so long as it advances “significant government interests” and does not unreasonably burden fair use.  Here, the limit was “incidental” because of analog alternatives, and the financial burden was not placed due to the content of the speech.   Judge Illston’s analysis weighed the interests involved with substantial deference to the government.

What’s interesting is that, at first, it sounds like intermediate scrutiny, which is what the Eldred appelants wanted, but it seems very watered down. So it’s better than no First Amendment scrutiny, but only just.

Sigh.  I was hopeful that this would turn out at least a little better, because Judge Illston took an extremely long time (9 months) to render what ended up a rather simple decision.  But, this argument is an uphill battle, and it will likely take something more than a tool primarily used for making back-up copies to make it work.  It’s got to be something more inconvenienced, but also more striking - something like security research, perhaps.  On the constitutional arguments, Kevin makes some other cogent criticisms here.

It also wouldn’t hurt to have a judge who spends less time making conclusory arguments.  The entire opinion is basically citations of Corley and Elcom, but that was somewhat to be expected.  The problem is sections like this:

“Congress enacted the DMCA after evaluating a great deal of information, including testimony from a number of the law professors who filed an amicus brief before this Court. Congress determined that the DMCA was needed to protect copyrights and intellectual property rights; this Court finds that the challenged provisions further important and substantial government interests unrelated to the suppression of free expression, and that the incidental restrictions on First Amendment freedoms are no greater than essential to the furtherance of those interests.”

Before that, Judge Illston mentioned that intermediate scrutiny requires deference, citing Turner I.  But, from what little I know of First Amendment law, it does not mean a complete free pass.  Ward v. Rock Against Racism requires that the regulation “be narrowly tailored to serve a significant governmental interest.”  Narrow tailoring requires only that the interest ”would be achieved less effectively” without the regulation.  Even so, that still requires the court to consider alternatives; if a different regulation could achieve comparable results with less impact on speech, then the present regulation is invalid. 

In Turner I and II (which came after Ward, btw), the SC does set forth a very deferential standard. Rather than weighing competing theories of future harms and benefits, simply asked Congress to provide substantial evidence.  At the same time, the Court examined that evidence to make sure it was indeed substantial.  The Court could then assess whether the “burden imposed … is congruent to the benefits it affords.”  Even though the Court said it need not reject the regulation because another regulation would be “marginally less intrusive,” it did consider several alternatives to make sure they were not “substantially broader than necessary.”

Judge Illston doesn’t bother with any analysis on these scores, and that sadly is a pattern throughout this opinion.  She doesn’t even bother to cite Corley here, probably because it makes similarly conclusory arguments regarding narrow tailoring.  (For more discussion of these cases, see the article linked to here).

Illston also failed to shed any light on the difference between access and copy controls, a favorite subject of Ernest’s.  Skylink is the only case to give a refined, if muddled, definition of access controls.

But 321’s fight is far from over.  Apparently, they’re going to release their software without the decryption component (DeCSS or a variant).  Their opponents claim that this isn’t kosher, but I think 321’s actually got a decent case.  They will have to be very careful, but so long as they don’t tell anyone how to make a circumvention device, doesn’t seem like they’ll be trafficking.  If the DMCA does not apply, then Sony does, and that’s a standard they can surely make.

Here’s the silver lining in Illston’s ruling - she seems to have anticipated this in a way that favors 321:

“The DMCA does not prohibit copying of non-CSS encrypted material, so if 321 removed the part of its software that bypasses CSS and marketed only the DVD copying portion, it could freely market its product to customers who use the software to copy non-CSS encrypted DVDs and other public domain material.” (emphasis added)

Though it might also be used by customers who copy CSS encrypted disks, that is irrelevant under Sony.  Copying non-CSS encrypted DVDs and public domain material will likely count as substantial non-infringing uses.

Filed by Derek Slater at 1:24 am under Big Ideas
1 Comment

With the EFF releasing its P2P white paper on voluntary collective licensing, I think it’s time we started talking terminology. There are a lot of acronyms associated with various copyright policy solutions, many of which are distinct but used interchangeably.  So let’s try to sort them out together.  Here’re some that I’ve picked up on - let me know what I’m missing and what I’ve got wrong.

Compulsory licensing refers to any situation in which, by statute or other governmentally mandated means, a copyright holder must license a particular right to anyone at a given rate.  Current examples include the covers and webcasting license.  Proposed examples for the online downloads market include mandated versions of the terms below, as well as the non-discriminatory licensing provisions in Rep. Boucher’s MOCA.  The latter is similar to aspects of Professor Fisher’s “public utility” model.

Collective licensing, if taken at its broadest meaning, can mean any time one agency is designated to setup and manage licensing arrangements, and then collect royalties for any copyright holders who want to become members.  It is most often used in relation to blanket licensing, where the agency licenses all the rights it has been assigned as one package.  For instance, ASCAP, BMI, and SESAC license the right to publicly perform for all artists in their catalogs.  They may license to anyone willing to pay at a given rate, and they can be forced to do so (e.g., ASCAP and BMI’s consent decree).
Variations on that theme: It could license to absolutely anyone (individuals, P2P providers, ISPs, etc.), but one could conceive of blanket licenses that are only offered to certain entities.  The payment for the license can vary - it could be a percentage of revenue or a flat fee.  As far as the pay-out, ASCAP and BMI count the frequency of use, but one could conceivably pay out in some other manner.
The EFF’s plan is collective, blanket licensing in a fairly conventional sense - one agency licenses to anyone and pays out by frequency.

Alternative Compensation System implements a manner of collective licensing, with some differences, particularly in the mandatory model. It is connected most directly with the models presented by Professor Fisher. In the voluntary version, an artist/copyright holder co-op would offer individuals a subscription to license certain rights - as presented by Professor Fisher, this is the entire bundle of rights, for non-commercial or commercial use, though one can imagine variations with less rights.  Content distributors, like webcasters or download sites, participate without having to pay fees so long as they counted and reported frequency of use/downloading; rather, individuals would pay the fees and thus the services would be authorized to let them download.
The mandatory model would differ in the following ways. First, rather than a co-op, the government would create or designate a particular agency.  Second, the revenue would be generated through taxes, not through licensing fees.

What smells like these but does not actually fit the categories?  P2P service Wippit and Napster 2.0’s deal with Penn State.  In both cases, people are getting all-you-can-eat access to a catalog, so the practical result is similar to some extent.  But the licensing arrangements are negotiated with each copyright holder individually, and those terms only apply to that negotiation.  Like in the EFF’s voluntary collective licensing or Fisher’s voluntary co-op, you have to sign up with a single entity to get access; however, that does not buy you licenses to perform certain activities on any P2P or downloading site of your choice.

Filed by Derek Slater at 11:36 pm under General news
3 Comments

About 5 months ago, Donna wrote “I’ve had thoughts brewing on [how to resolve the P2P wars peacably] for some time. While I cannot yet share details, I am working on something that I hope will serve to 1.) further the collective problem solving and 2.) help clarify various positions within the debate.” That came in response to some discussion of the EFF’s P2P policy solution and strategy.

I don’t know if this is what Donna meant, but it sure seems like part of the puzzle. The white paper is clear, concise, very well thought out - one of the best bullet-points arguments on this complex topic.

I’ll likely have more to say on this, among other things, in the next few days.

Filed by Derek Slater at 11:09 pm under General news
No Comments

Ernest has a nice follow-up to my post yesterday.  From a public policy perspective, I generally agree with him.  I also agree that there are significant practical problems - it will be incredibly difficult and will require more adverse effects (and legislative change) than people often concede.  But I’m still not totally sure I can call it futile yet.  The window doesn’t have to be gigantic - many movies derive their revenue substantially from first week or two of sales, and the same is true to a slightly lesssr extent in music.  The industries could shift even more in that direction.  It won’t be helpful for all artists all of the time, but it could make a difference.  But, like I said, I’m not totally convinced that it’s practically feasible in a way that won’t be substantially harmful on balance.

Filed by Derek Slater at 10:45 pm under General news
No Comments

You might want to acquaint yourself with PaidContent.org - in particular, check out the great notes on the Midem conference.  Forrester presented some research there that got a lot of press - in sum: the online music market will grow and the music industry will recover from recent losses, but this year, many online music services are going to go bust.

And why shouldn’t they?  The a la carte services are not making much if anything right now.  They can make it up in volume, but the market isn’t big enough for that now, particularly with so many services in the market.  Apple has already gobbled up most of it (70% domestic, 30% world), and they can afford to run it as a loss leader because of the iPod.  For the leftovers, Walmart is undercutting everyone else, running it as a loss leader as well.  Meanwhile, the subscription services are off to a slow start; they should come on eventually, but problems involving portability and customers being accustomed to owning discrete units will continue to hold things up for awhile.

These services’ dying doesn’t necessarily mean that the market can’t work, just that it can’t support this many services at this time.  But will it negatively impact the long term market in any way?

Looking just at the prices and catalogs, there doesn’t seem to be a significant effect.  Given the thin margins on the a la carte download services, price doesn’t seem like it’s going to be a major point of competition and differentiation in the near term.  At this point, the catalogs also don’t differ too much in size and scope.  So, after the fall out, there will be less stores, but not necessarily a worse or different situation for the consumer.

Things might be worse because there will be less variety in ancillary services (recommendation systems, discussion boards, music store magazines, et. al). These do differ among the stores, and, moreover, they’re part of how the stores are trying to set themselves apart from P2P.  Because the market cannot support all these stores, it will ultimately not support this variety of services.  Variety is attractive and can help boost the market, but it’s not an end in itself, so this might not be major.

A larger dampening effect will be with subscription services if they go out of business, particularly Napster 2.0 (which lost a 15 million last year).  Napster 2.0 users can’t burn songs to CD, but they can port them to a device, so it feels less like a tethered download and more like actual ownership. If all of a sudden, consumers don’t have access to a collection they’ve built up for a few months, there’s going to be serous frustration. Correction (3/10) - Napster users actually cannot port to a device.  I remember reading it in the terms and conditions when it launched, but the terms have been updated, and this press release makes it pretty clear that porting can’t happen.

But the biggest problem will most certainly come from the balkanization of DRM formats and associated players.  Real is following Apple’s lead with close tying of the DRM to portable players; so might Sony.  People will build up collections from one store.  As if incompatibility with other players wasn’t enough, it’s going to be horrible when someone basically has a defunct format when a store goes belly up.

Many people in the music industry and in the stores consider this format balkanization the biggest obstacle to a growing market.  I’m not sure if it’s the biggest, but it certainly is an important one.

Another is the catalog.  They have to get the hold-out artists on board, as well as deepen the catalog with live performances and other obscure recordings.  According to Phil Leigh citing a Webnoize study, catalog and convenience were generally above price as a key reason why people preferred Napster.

But price is still important - it’s obviously been a key to the initial interest in Apple. Prices won’t change in the near term unless the record companies take a lower cut - I wonder how well Walmart will even do at 88 cents.  Rhapsody’s experimentation gives reason to believe that a lower price can be made up with much higher volume, but right now the record industry isn’t changing their strategy. 

Even if it does, and prices do change, this music bubble will probably still burst. But it might burst differently, and that could have longer term consequences for this nascent industry if it is to survive (if it even can).  Short term gains will matter - how long can the record industry losses continue before it’s basically too little too late?

Filed by Derek Slater at 9:53 pm under General news
6 Comments

One key barrier for the legitimate music services is the size of their catalog.  Among many issues:  the legitimate services can’t convince many artists to offer their music, and P2P has numerous live recordings that you cannot buy anywhere legitimately.  The former will take a little time, as artists adjust - and they certainly will. The latter seems a bit harder.

Nugs.net provides a glimpse of the future.  They’re making high quality recordings of live performances and placing  them online for purchase within 48 hours.

Filed by Derek Slater at 12:13 am under General news
No Comments

Billboard reports that the music industry is pushing Apple and MS to solve their DRM interop problems, and even suggests that there have been private talks between the relevant parties.  The solution: “transcoding” - being able to securely turn a FairPlay-AAC file into a Secure WMA file.  So Apple won’t sell you an actual WMA file, but it will provide you the means to turn their files into WMA.

Huh?

Maybe the RIAA is serious about pursuing a business model that would take advantage of DRM as described below.  But, in the current environment, DRM is not playing that role.  It is dragging down the market and providing no real benefit. Given that, all the compatibility problems can be solved with three letters: M-P-3.

Filed by Derek Slater at 11:57 pm under General news
2 Comments

Mary discusses more of the Digital Media Summit, including Professor Nesson’s talk.  The closest analogue to his idea in the Digital Media Project’s “five scenarios” is actually the Technology Speed Bump scenario, one that’s now been broken out on its own.

This scenario fascinates me for many reasons (its the motivation behind last week’s post on release windows). One interesting part is the role of DRM.  Unlike in today’s digital media market, DRM might have some use in preventing piracy in the speed bump scenario and the biz model Professor Nesson describes. 

Contrary to the content industries’ hopes, DRM does not stop piracy, nor does it even reduce piracy by itself.  Rather, it can only reduce the initial number of uploaders.  Given that one copy can spread infinitely, DRM has been declared basically useless.  Moreover, the speed bump threat model has been declared dead.  The darknet paper basically confirms as much.

But the darknet paper is also careful to note that DRM’s affects might vary “in the presence of a darknet, which is connected, but in which factors, such as latency, limited bandwidth or the absence of a global database limit the speed with which objects propagate through the darknet.”  Consider the plan Professor Nesson describes.  DRM becomes more meaningful because, if you can limit the number of initial uploaders, interdiction becomes much more possible and thus effective. Interdiction is tough if you have to stop myriad uploaders.   Moreover, by limiting the initial uploaders through DRM and their propagation speed through interdiction, spoofing can be more effective, because there are less good copies that need to be crowded out.

These impediments will not stop everyone. But in the short term, they might form an adequate speed bump to create that new release window.  Of course, interdiction and spoofing can be defeated.  And, as I have expressed before, I have significant misgivings about this scenario and am not confident that copyright holders will be able to keep up.  But if they could keep up only for a short period of time, it might have some benefit.  Maybe.  If so, then DRM might finally have an effect on piracy, as it was supposed to, rather than simply preventing competition, legitimate uses, interoperability, and innovation.  Maybe.

Filed by Derek Slater at 11:30 pm under General news
1 Comment

Check out the new Audio Berkman about the iPod.  Nice mix of views from: Fred von Lohmann, Mike McGuire, Cary Sherman, Scott Kirsner, and Urs Gasser.

Filed by Derek Slater at 9:58 am under General news
3 Comments

The Ellison v. Robertson  appeals decision came yesterday.  The district court noted that AOL had no obligation to monitor the actions of users for infringements, terminate repeat infringers, or investigate such users; in fact, 17 USC 512(i) “only requires AOL to put its users on notice that they face a realistic threat of having their Internet access terminated.” The policy is only a “mere threat,” which never needs to be put into action.   If 512(i) forced ISPs to punish people for infringement, then “most if not all of the notice and takedown requirements of the subsection (c) safe harbor would be indirectly imported and applied to subsections (a) and (b) as well.”  (I’d say that that’s already happened.)

The appeals court doesn’t spell it out, but they disagree to some extent.  According to the ruling, the ISP must have a notification procedure much like that for 512(c).  Thus, a service provider cannot simply create a policy and then remain willfully ignorant of notices pertaining to that policy.  Seemingly, an ISP would have to act on the notices to whatever extent its policy requires; otherwise, notices of infringement would still “fall into a vacuum and go unheeded.” 

Neither ruling touches on whether these notices must be of actual infringements. One can strictly interpret 512(i) to say that only repeat infringers, and, as opposed to 512(h), not “alleged” infringers, must be terminated - thus, only people found to have infringed by a court twice must be terminated.  In its more flexible reading of what it means to “reasonably implement” a policy, the appeals court points in the other definition of infringer, but it’s not clear from the ruling.

There are other cases that deal with this matter, too.  See Perfect 10 v. Cybernet (summary here) and In re: Aimster.  The former states that “sufficient evidence of blatant, repeat infringement,” and the latter affirms that the ISP must do “what it can reasonably be asked to do” to prevent repeat infringers from using the system.

Filed by Derek Slater at 9:14 am under General news
1 Comment

Blockbuster is a new ally against region coding, sorta.  This article seems to say they have two beefs.  First, people are importing DVDs and using chipped players, so local Blockbuster’s are getting left out of the distribution loop.  Second, by the time we reach the international DVD release window, the piracy floodgates have already been wide-open from the domestic release window.  So, though not against region coding in principle, they’re against it in practice.

This reconsideration of the release windows is a broader trend in the movie industry.  Matrix Revolutions was released at the very same minute all over the world; X2 was released at the same time in different time zones.  As DVD sales grow and make up more of total revenue, the rental window shrinks.  Indeed, the entire cycle is becoming compressed.

(Cue thinking out loud:) There seem to be three factors at work here:

1.  Big budget, high marketing cost films create increasing incentives to release films broadly and recoup as much as the cost as quickly as possible.

2.  Going digital means going cheaper in manufacturing and cost to consumer.  Maybe we don’t need the jukebox in the sky - the jukebox is becoming priced to own for your home.  If tons of money can be made from those sales, why wait to price discriminate between people who would rent versus those who would get PPV versus those who would wait for cable TV?  As price goes down and convenience goes up, it might make less sense to have many windows spread out over time.  This should only continue as movies are sold online.

3.  Going digital, of course, means piracy.  That was the motivation behind the Matrix and X2 premieres.  It should have a similar impact along the whole timeline eventually.  Even if piracy doesn’t cut into theatre revenues, camcorder copies of theatre showings will cut into all subsequent revenue windows.  The more time between the camcorder copy hitting P2P and the DVD or digital download being released at a reasonable price, the more chance someone will choose to download.

Won’t all this compressed cycle cannibalize some of the revenue streams?   Maybe. But consider how precisely movie studios can track box office numbers. Add into that equation how precisely they could track online downloads and how easy it would be to create the files on a whim.  They could pin down when the box office is starting to cool off and immediately shift gears into the downloads market.

Right now, the movie industry seems to be adjusting relatively well to this - DVDs have been a huge boon.  I’d say we can expect similar compression in the music cycle - in fact, we are already seeing some shfits.  But will we see similar success?  To what extent will they too be able to sufficently recoup costs in a shorter timeframe?

Filed by Derek Slater at 11:36 pm under General news
No Comments

EFF reports “The Hon. Jeremy Fogel indicated he intends to issue a ruling within the next two months on whether Diebold will face the consequences of abusing copyright law.” Maybe it’s just wishful thinking that this could go in our favor.  But even getting this far feels like a win.  Here’s why:

“The plaintiffs liken themselves to modern Galileos persecuted by authorities. I fear that a more apt analogy would be to modern day Don Quixotes feeling threatened by windmills which they perceive as giants. There is no real controversy here.”

That’s from the judge at the final hearing for Felten v. SDMI.  He must have felt very clever. I doubt security researchers felt any safer.

Filed by Derek Slater at 12:05 am under General news
1 Comment

The current trend in digital media services is the Digital Music Store - more or less, you pay to download or stream copies of songs.  Whether you pay by the song or by the month, you’re paying that particular service to give you access to those copies. Though people now have the ability to spread copies far and wide, these services hope that that won’t be a major problem.  To an extent, the Stores are predicated on a combination of technology and law making P2P distribution and downloading significantly costly to the user.  In this way, they imagine a future in which copies can still be controlled.

ACS’ jump off from a different spot.  For one, you wouldn’t really pay for access to a particular service in the co-op or mandatory models (note: this a new draft from Professor Fisher).  Anyone could be a streamer or download service, so long as they counted the relevant downloads.  Also, they try to build on rather than fight against the ability to make copies of digital media.  But they don’t do so by simply accepting that the exclusive right to copy and distribute are dead.  Rather, they try to monetize those actions in alternative ways.

In terms of current dialogue, I’d say these two models dominate the current landscape.  Before, people talked more about a myriad of other alternative models that accepted that copies could no longer be controlled and thus focusing on that revenue stream should greatly lessened if not abandoned in the digital age.  Raymond Ku and Mark Nadel (among others) have fleshed out the ideas well. Mary explains the basics, and many models are listed here by the EFF.  Among them: tip jars and donations; live performance, merchandising, and other complementary products generating revenue; ad-driven models; and charging for the first copy only (see: Digital Art Auction and Street Performer Protocol).  Even Magnatune, which sells copies of albums, follows this pattern to some extent - the sales are almost donations, because the songs are CC licensed, there’s nearly unlimited high qualify try before you buy options, and you choose your own price - basically, Magnatune is banking on fans in niche markets will be highly devoted to the artists they support. 

Two things occur to me about these models.  First, they are both more and less sustainable than some people in the past have asserted.  When I first read about these models, tip jars in particular (I don’t know when), they were talked up a lot - when Stephen King stopped writing his book online, people said that these schemes were pure hype.  The truth is likely somewhere in between.  These models can make money - the question is how much and for whom, and what meaning does that have for our media environment. 

The second point, then, is just that.  How far do these models go towards creating a thriving artistic culture? Start with these general counterarguments (pg 74). Because of cost savings, one need not necessarily replace the entirety of the revenue streams from copying and distribution; however, those are sizeable streams to make up.  Perhaps we don’t need to make them up - we’re content with a lower amount of revenue and thus maybe a smaller, or at least a different, media market.  Fine, many musicians don’t currently make money off of selling copies anyway, but many artists also don’t make much money from the other streams.  Are we expecting a tiered and varied industry, with some nationally popular artists, some regionally, many indies, et al?  Or will we see a ton smaller indies be sustainable but unable to make revenue, exposure, and success go beyond that?  Will we have a lot of people like Magnatune’s artists, focusing on niche markets but not moving above, and are there any negative consequences of that?  And what about the consequences of having artists overly dependent on selling merch or pushing ads - aren’t only certain types of media and artists capable of doing well this way? 

There is no satisfactory or unsatisfactory answer here.  Rather, I want to get a sense of why people think these models will or won’t work, and what it means for them to work.  What does that world of alternative biz models look like?

Filed by Derek Slater at 4:50 pm under General news
3 Comments

Renny Hwang reports in what one will be one of many Digital Media Project papers coming out in the near future.  Highly recommended reading.

Filed by Derek Slater at 12:50 am under General news
4 Comments

1.  “Record Industry Cuts Corners in Crusade Against File-Sharers
Public Citizen, Electronic Frontier Foundation and ACLU Say “File Lawsuits Properly”. The groups do not question the seriousness of the illegal activity alleged by the record companies, but object to the process the companies have tried to use to obtain the file-sharers’ identities.”

I’m glad they are sticking to their guns and ensuring that this entire process is carried out as fairly as possible.  At the same time, the more they push, the more they leave themselves open to attacks that they are really trying to protect infringers.  An unforunate consequence, but one that I expect they will struggle with more and more.

2.  “MacArthur Foundation Awards $600,000 to Electronic Frontier Foundation
“There are a number of influential members of the entertainment industry that use industry standard setting meetings - such as the Digital Video Broadcasting Project in Europe - as opportunities to mandate technological features that control digital media, such as devices for digital television that limit the user’s ability to copy programming for personal use or to skip commercials,” said Jonathan F. Fanton, President of the MacArthur Foundation. “By participating in such meetings, the Electronic Frontier Foundation is working to protect the rights of the user and ensure that future uses of technology are not restricted by the industry.”"

It is impressive to see the MacArthur Foundation throw its weight behind this project, among others, in this field.  One, two, three years ago - would you have expected this?

Filed by Derek Slater at 12:47 am under General news
1 Comment