Yesterday a federal judge issued an injunction against the file-sharing network LimeWire, causing it to stop distributing and supporting its software. Meanwhile, the Choruss initiative to implement “flat tax” file-sharing on college campuses appears to have died, despite the effort that Warner Music Group and longtime flat-tax ideologue Jim Griffin put into it.
Judge Kimba Wood’s injunction against LimeWire, which follows her summary judgment against the company back in May, is a milestone in the company’s four-year legal battle with the music industry, and by all accounts, it’s the end of the war. It’s also the first time that the Supreme Court’s 2005 Grokster decision, which created a new category of liability for “inducing copyright infringement,” has manifested itself in a major court action.
LimeWire CEO Mark Gorton claims that he intends to figure out a way to turn LimeWire into a legal paid service. Good luck with that strategy, Mark; it worked so well for Pirate Bay, iMesh, and Kazaa, just to name three.
Gorton’s attempts at emulating Michael Robertson in the music-industry-nose-thumbing business have failed; neither his blatantly disingenuous legal strategy nor his PR charm offensive have worked. And whereas Robertson made tens of millions selling his original MP3.com to Universal Music Group, Gorton may find himself personally liable for that much in fines.
The failure of Choruss is a different story, but there are common elements to both. Choruss was an attempt by Warner Music Group CEO Edgar Bronfman to experiment with a music licensing model that the copyleft has been advocating for years: a flat monthly fee per user in return for unlimited rights to use content on a network. Bronfman had hired Jim Griffin, a respected digital music pioneer as well as a vociferous advocate of what I call the flat tax approach, to lead the effort.
Griffin quickly found that the industry was not going to adopt a US-wide flat tax model voluntarily; it would require getting Congress to enact a statutory license. That wasn’t going to happen in any reasonable timeframe (if ever), so he turned his attention to college campuses.
College campuses had several advantages in terms of openness to the flat tax model: legal, technical, and economic. The legal advantage was that college administrations could have been held liable for the copyright infringements of their students — which are typically massive — so they had incentive to adopt a legal music service of some sort. The technical advantage was that colleges control their campus networks and are capable of filtering them (at least to some extent) so that file-sharing can be done freely inside the firewall but not through it to the outside world.
The economic advantage was that college administrations could easily charge students a monthly fee for use of content on the campus network. Such a fee could be added to existing student activity or IT fees without too many people noticing or caring, especially alongside the five-figure tuitions many colleges charge. Subscription services like Napster had been offering this arrangement to colleges. But unfettered file-sharing with access to a large, DRM-free music library would be better for users, all else (such as price and music selection) being equal.
In other words, Choruss was offering a value proposition to colleges that was actually superior. Unfortunately, Choruss failed because it couldn’t get the licenses to music from at least one of the major music companies and several publishers. (See my article from earlier this week for a possible reason why.)
Choruss had intended to partner with a file-sharing service called Audiogalaxy, which would have provided the infrastructure and user interface. Here’s where the commonality with LimeWire comes in. Audiogalaxy was a file-sharing service that operated during the original Napster era. Like LimeWire, Audiogalaxy used patently ineffectual technology to block sharing of copyrighted works, and like LimeWire, it was shut down in music industry litigation. (Yet in some ways, Audiogalaxy was ahead of its time: it was the first file-sharing service to incorporate what we now call “social” features in its user interface.)
Now that Choruss has failed, Audiogalaxy is relaunching as a service that lets users sync their own music files across their devices and stream it from a server — in other words, it now looks a lot like other current services such as Catch Media and DoubleTwist. In a post last week, Digital Music News’s Paul Resnikoff noted — as I did for Catch Media — that most of the files that would be “synced” with such a tool are most likely not legally obtained. In calling this “ironic,” given Audiogalaxy’s history and Jim Griffin’s intentions with Choruss, Resnikoff was right on the mark.
Copyright and Technology 
I’m confused by the approach (as reported in the linked article). Jim Griffin is quoted as saying:
“We couldn’t even find half of the rights holders. Without locating these rights holders, it was impossible to secure the necessary licenses.”
Does this mean that they weren’t negotiating with any Collective Rights Organisations then?
Ben,
I found this comment a little mystifying too, especially given that Jim Griffin knows the collecting society world extremely well. I think it depends on what content Jim wanted to license for Choruss. A typical set of licensable music has developed – essentially it’s the four majors plus the major indie aggregators such as Merlin and the Orchard. To go beyond that requires negotiation with individual rights holders.
The ones that Jim “couldn’t even find” would have to represent relatively obscure music, and/or local repertoire from countries outside the US and western Europe.
However, I suspect that Jim’s statement is a bit of a… how shall I put this politely… red herring. I think that many rights holders are fundamentally opposed to the flat-tax idea, for a number of reasons, including philosophical ones that I discussed last week plus a lack of trust that the royalties will be distributed equitably under such a scheme.
Despite its advantages on college campuses, I personally think flat tax is a horrible idea, but let’s not get into that here….
Good insight – I can only agree. But if you have time, go on! I’d be interested in knowing why you think a flat tax on ISPs would be a bad idea…
OK. I’ve written about this at various times in the past, but here are my top three reasons why flat tax is a terrible idea:
1. The flat tax system sweeps under the rug the issue of proper royalty allocation. Right now, the collecting societies apportion royalties using somewhat arbitrary formulas for setting royalties to downstream entities and/or consumers (e.g. European levy systems; you don’t have these in the UK, and we have very limited levies in the US) and by a range of dubious and inaccurate methods to determine whose content got seen or played how many times. This system is bad now – it’s particularly unfair to “long tail” rights holders, such as indie artists – and a flat tax system would only introduce opportunities to make it worse.
Jim Griffin is open to the possibility that in a closed system like a college campus, you could use something like audio fingerprinting to count file transfers, but this is not the same thing as counting plays; you’d still be relying on statistical proxies, just as radio stations do now with ASCAP and BMI. The only systems that do this accurately are the streaming services like Napster, Rhapsody, MOG, and Spotify — all of which would go away under a flat tax system.
2. Flat tax penalizes those who don’t consume digital content and favors those who consume a lot of it. It’s patently unfair. It’s like making everyone pay the same electricity bill each month. ISPs could possibly mitigate against this by paying the flat tax themselves and placing a layer on top that apportions fees more equitably, but then all the heavy users will just go somewhere else, where this isn’t done.
3. Flat tax plays havoc with incentive to create by keeping the size of the pie constant, or at least dependent on a factor that is almost totally irrelevant (number of subscribers to network services). There would be no incentive to do anything that involves promoting or developing artists, whether by a record label or some other entity.
I could go on; there are other reasons, but that’s enough for now.
Thanks for setting that out, really interesting.
I’m supposing that when we say ‘flat tax’ we’re both talking about the same thing. For me, this means Blanket License. Is there a difference for you?
The beauty of legitimate networks (e.g. the Broadcast network) is that there is no incentive for the potential copyright licensee to infringe copyright – the fee has already been paid by a blanket license, the only problem remains (as you rightly point out) the accuracy of usage reporting. You mention fingerprinting but perhaps watermarking is a more pertinent technology within legitimate networks, since in this case one is not interested in tracking any possible infringing usage just legitimate usages. One would imagine that a combined approach of watermarking plus blanket license on ISPs could work very well…No?
I get point two and think that your objection plays strongly in the US but maybe less so in the EU, where we have public service broadcasting paid for by a flat tax – Everyone’s happy to pay it so long as we can see quality output and feel like we still punch above our weight. Going even further into that mindset there are in the EU (as I’m sure you know) CROs that support subsidised cultural causes from the gross revenues, effectively reducing the outflow to foreign Rightsholders. I don’t think any of this will change in the EU for many decades to come…
As regards incentives…Well, there is great competition within a closed flat tax economy to get onto distribution networks that pay better than others, which is a kind of incentive, albeit one within a rather nasty zero sum game…Still at least the overall market would grow if the Griffin approach was adopted and then over time the Internet market could grow and remunerate equitably over time.
Ben,
Yes, we’re talking about the same thing. I call it a flat tax because that’s what it is. An even more correct term to use in the European context would be levy, but many Americans don’t know what that means.
Watermarking is a way of identifying files, not tracking play counts. The currently predominant way of identifying files — and the one that Choruss would probably use — is fingerprinting, which is “look at the bits and guess what the content is.” Fingerprinting is widely used, but it has two disadvantages in this context: first, it’s not 100% accurate, especially for long-tail content; second, it can’t distinguish between different editions of the same music (e.g. US vs. German or track from original album vs. track from anthology).
Watermarking solves these problems but has the insurmountable problem of having to insert watermarks into every file that’s out there in the wild. The technique that Nielsen uses for US network and cable TV content is to watermark as much broadcast content as they can and rely on fingerprinting for the rest — a sort of “belt and suspenders” approach.
You still have the problem of play count reporting. Any system that relies on file-sharing, especially unprotected files, can’t solve that problem in a secure manner. And that’s only the tip of the iceberg. The part of the iceberg that’s below the water line is having a complete and reliable database of rights holders. This is what Jim was lamenting. You can’t do any of this without that. There is an EU-level project now called the Global Repertory Database (GRD), which I wrote about recently and which purports to solve that problem. It’s nice that they are working on a solution but it is a very daunting challenge on several levels and meaningful progress will be years away.
We do have public broadcasting here in the US, albeit on a smaller scale than in the UK. We just don’t have explicit license fees as you do. Instead, a small portion is funded through the income tax, a bigger portion through individual donations (“voluntary license fees” if you will), and the biggest portion through corporate philanthropy — which in a sense is also government funding because the corporate funders get tax breaks.
Finally, regarding the incentives: I disagree. The incentive alignment is all wrong. Incentives for greater Internet use are only tied to increased use of music and other copyrighted content in a very minimal way. People don’t sign up for ISP or mobile accounts to get music; they sign up for email, Facebook, web browsing, etc. Economic incentive misalignment is a recipe for failure — as it has been for DRM in music. But the zero-sum game is what really kills the idea. The market needs investment to grow. Investors (of one kind or another) are never attracted to zero-sum markets.
Off the bat, I agree with what you’ve said here – entirely. But out of interest could we tease these arguments out a bit more?
Starting at the end and then working to the front:
My point about the zero sum game is that any flat tax economy becomes this, on this we can both agree. The music economy is comprised of many micro markets that contribute to the total. Some of these markets (and I’m generalising globally here) are zero sum, others are not (such a live music). My stress was simply that perhaps a zero sum flat tax environment would be a better place to rethink the Internet space than the current mess. I don’t think this would deter investors looking to build in various parts of the value chain because the ‘Internet and Music Mix’ is only ever one constituent in the mix. There’s always ad revenue, subscription services and so on. I can hear your objections as I type but I really think the key to what I am saying is that the answer is not free, freemium, premium but get the market sorted with a basic level playing field (flat tax) and then add value.
As you mentioned with the Nielsen example, where a network is legitimate (like Broadcasting) watermarking is perfectly good for tracking all usage. Isn’t it? Any legitimate network simply has to have software looking for the watermark and that’s that – doesn’t matter how many times it has to record it. If the Internet was made legitimate by a flat tax, what incentive would there be for people to bother with files that were unwatermarked? They’d have to be more interested in no one knowing what they were listening to than listening to it in the first place…That’s an incentive I don’t understand – so over time all those unwatermarked files would become unused.
I share your appraisal of the mammoth task to watermark the entire back catalogue of music rights but it could be done. And more efficiently than fingerprinting which can only be done in real time (I think) whereas watermarking can be done offline (I think)…
Ben,
First of all there is no such thing as a level playing field in this case. My #4 objection to Flat Tax is that there are those network service providers who won’t pay it, including those outside the US. They won’t be licensed for content use, which means that the legal posture towards them would be the status quo ex ante. Of course, that won’t stop millions of people from sharing content files on these “unlicensed networks.” Funnily enough, some of the copyleft advocacy types to whom I try to explain this never seem to have thought of this problem.
If there were a level playing field, then I agree with you that you’d be able to create a market for value-added services (even though you’d still have the problem of proper rights holder compensation). This would make it easier for the market to define the true value of services such as curation, streaming access, social recommendations, and so on, without having to intertwine it with the value of “bits.” But there is no level playing field.
Regarding watermarking: it is possible for record companies to watermark their existing catalogs. But it is impossible to insert watermarks into all files that are in the wild. Even if this were possible (which it isn’t), you’d have to have standards for insertion and extraction as well as the watermark payload for it to work. The RIAA has proffered a standard for the latter, but the former — watermarking algorithms — is “secret sauce” to watermarking vendors and covered under patents, and therefore would be very difficult to standardize. It’s very hard to set standards that enshrine technology choices (mandates) into law without running into antitrust problems.
Fingerprinting is used because it doesn’t require that data be inserted into files. And it needn’t be done in real time: the leading fingerprinting vendors (e.g., Gracenote and Audible Magic) have databases of millions of tracks’ worth of fingerprints. You can use fingerprinting on tracks in the wild. But fingerprinting doesn’t help you track usage unless there is a secure way to apply it within the player. This in turn is impossible unless you use DRM to create a closed environment… and DRM has its own set of issues.
Thanks Bill. I’m learning a lot, thanks very much. Another angle…
When one looks at the problem of who profits from file-sharing, the debate often revolves around on the one side: copyright holders who lose licensed revenue and, on the other, ISPs gaining revenue from bandwidth usage with their customers enjoying tracks for free. But isn’t this framing of the ‘Copyright War’ disingenuous?
As I understand it, in the UK broadband is provided by BT and what is now Virgin. Two providers only. All ISPs sublet space on these networks. Then there’s the Mobile network and Satellite to add to the mix…It’s not a wide open market with multiple players. If the flat tax were aimed at the providers of the infrastructure of the communication network rather than the Service (ISPs) would that make my argument simpler?
Isn’t there some kind of follow the dollar principle in law that shows these services benefit from copyright consumption to a degree that suggests they should contribute a flat tax?
Why is “this framing of the copyright war disingenuous?” I don’t necessarily disagree, I just wonder why you say that. Because that’s pretty much the standard set of arguments nowadays.
Now, if you change a blanket license (agreed by the industry, e.g. to stave off decades of litigation with uncertain outcomes) to the notion of a statutory license, one required by law like the ones for broadcast radio in the US, then the game changes. In that case, as you correctly imply, the challenge is to identify who has to take a license — the actual “pipes” providers or some larger set of service providers?
The answer is not necessarily obvious, and content owners probably don’t particularly care one way or another, as long as they get paid. The problem from a practical perspective is that there is a long-held principle in telecoms called common carrier exemption, which says “telecoms are just the wires,” meaning that telecom providers should not be held responsible for anything having to do with what goes over their networks. This means that telecoms are not held responsible for things like libel, fraud, and obscenity as well as copyright liability. So what you’re talking about is reversing that longstanding principle.
Interestingly enough, there is a recent recognition in US law that the major telecoms’ ISP businesses are to be treated separately than their basic networks. This was embodied in a 2005 Supreme Court decision called FCC vs. Brand X, which I have also written about (in my previous newsletter, DRM Watch, still findable on http://www.drmwatch.com). It seems that the ISPs want some room to maneuver regarding net neutrality, so that they can get away with charging tolls to content providers on their networks without jeopardizing the common carrier exemptions they enjoy for their infrastructures.
Anyway, assuming that you could charge infrastructure providers “content levies,” then the question for them is how they would pass those costs on to value-added service providers and/or customers. That’s a market question.
But it seems to me that opening common carrier liability to question leads to a huge Pandora’s box of ripple effects (to mix a metaphor) that could have a whole host of only tangentially related and possibly unintended consequences. So while it’s a road that telecoms will resist going down with their last ounce of strength, it’s also a road that regulators will need to be very cautious about following too.
The sense in which the question is disingenuous is this:
As we move forward (and to some extent we are here now) the infrastructure is the broadcast. The BBC does not, to my knowledge, use an ISP for its iPlayer and web services, for example. So, if I’m right about this is when the BBC operates on the web is it an ISP or a broadcaster or what? I would argue it is an aggregate channel operating on the new broadcast network. And if I’m right about that then it makes no sense to levy multiple blanket licenses to all services such as these, it would make more sense to levy a single, catch all charge at the level of the network.
What are the ripple effects you mention, or is that just too big a question- could you outline something?
I understand the good sense of common carrier exemption, it’s a sound principle but perhaps my remark here questions whether it still applies to the Internet as we move forward.
Thanks for the note on the SC ruling, I’ll give it a read.
I’m not a telecoms law expert, but I’ve talked to a couple on the subject of common carrier exemption. So I am not the one to do justice to the topic of what happens when the copyright industries try to chip away at the common carrier exemption.
But let’s leave it at this: if you get rid of common carrier exemption for copyright, then it will be argued why you should keep it for other crimes. The most obvious next choice would be obscenity, as opposed, say, to fraud — because obscenity seems easier to detect automatically the way the content owners would like copyrighted works to be detected automatically (through technologies like fingerprinting and watermarking). Then you would have all sorts of first amendment (free speech) issues at stake, and there are many policymakers would would not want to touch that one with the proverbial ten foot pole.
In the end it would come down to lobby efforts, which only coincidentally bear any relationship to moral imperatives…
These services got into trouble for 2 reasons.
1. They allowed users to search content
2. They never removed illegal content after abuse report
Mike