I’ve been watching with some interest the new Artist 2 Fan 2 Artist project, recently started up by Jon Newton and Billy Bragg. The intent of the site is to bring artists and fans together and encourage these parties to speak directly with one another, without needing to pass through intermediaries such as producers, labels, public relations groups, managers, and so on. It will be interesting to see how the dialogue develops.

One of the key elements of the site that interest me the discussion of paying artists (and other content creators); how can we avoid demonizing P2P users while at the same time allocating funds to artists/copyright owners in a responsible manner. On October 5th, this topic was broached under the posting ‘In Favour of a Music Tax‘, and I wanted to bring some of my own comments surrounding the idea of a music tax to the forefront of my own writing space, and the audience here.

I think that an ISP-focused levy system is inappropriate for several reasons: it puts too much authority and control over content analysis than carriers need, puts carriers at risk when they misidentify content, and would make carriers (for-profit content delivery corporations) in charge of monitoring content without demanding consumers that pay ‘full value’ for content moving through their networks. This last point indicates that an ISP-based levy puts ISPs in a conflict of interest (at least in the case of the dominant ISPs in Canada). Another solution is required.

I’m a Canadian, and I don’t have an issue with a levy-system that works. In Canada, we are charged a levy on all blank media that is sold. The levy originated several years ago, and was meant to recoup losses from the copying of mp3s (and related audio files) onto disks. The levy is very small:

• $0.24 per unit for Audio Cassette tape (40min or longer);
• $0.21 per unit for CD-R Audio, CD-RW-Audio & MiniDisc;
• $0.21 per unit for CD-R, CD-RW (non audio).
• In 2009 the levy on CDs and MiniDiscs will rise to $0.29 (Source)

Arguably, fewer people burn mp3s onto disks than in the past – with the advent of cheap and portable storage media, media tends to find its way to mp3 players and similar portable (and slightly less portable) media environments. Rather than targeting ISPs (who, really, should function like semi-intelligent data piping networks, rather than smart content monitors) why not impose a small levy on mp3 players/consumer electronic storage equipment? When I see an iPod or something like it, and am told that ‘160GB of storage just isn’t enough’, I have a hard time believing that someone has paid the not-so-small fortune to fill it with music, TV shows, and other content. I argue that we should toss a levy on the device at the point of sale, and then have a mechanism where the money is blackboxed and distributed back to content owners.

Is blackboxing ideal? No, but distributing the funds collected can be improved by developing either a more broadbased sampling of what is being listened to (i.e. one that accounts for the long tail of P2P downloads, YouTube watches, etc) or ‘smarter’ devices (i.e. ones that can read what is being played and then report this back to a mothership) to adjust the levy distribution in near-real time. As a consumer, the former is an acceptable shift in distributing the levy, whereas the latter strikes me as a gross violation of my reasonable expectations of privacy (regardless of whether it would cross an actual legal boundary). Let’s think through what this ‘tracking’ system might look like:

First, it won’t involve any deep packet inspection appliances, so we can dispel any worries associated with that particular technology. Instead, we get a few clever programmers together to develop an aggregation system that identifies the number of downloads of songs/albums/artists across a wide set of media environments (e.g. P2P sites, YouTube, MySpace, etc). Either the sites themselves or the aggregation system can include anonymization protocols that scrub data sets after initial collation, ensuring that the data can’t be traced to individuals/sites. Scrubbing protects consumers and has the benefit of mitigating some efforts to ‘game’ the system by demanding particular sites are included in the aggregation step at the expense of less favourable ones. It should be noted that I’m not seeing this system as one that tracks particular users but instead the number of downloads particular sites have for files. While this doesn’t generate 100% accurate picture of content transfers, it will capture the long tail better than the present Canadian levy’s evaluation metrics. Taking account of the fluidity of the P2P environment, it is important that the aggregation system is designed so that adding new sites to the scrape is relatively simple, and the system generally should be modular so that it can scrape new P2P platforms as they come available.The nice thing is that, once this system is paid for and set up, it’s largely self-running AND can be exported to different jurisdictions as needed. This system should be open source, so that anyone who is interested can take a look at how the algorithms calculate levy distribution in an effort to facilitate transparency and rigorous code-checking.

While there are certainly valid concerns around the organizational collection and distribution of these levy fees, I see a distinction between the value of a levy and organizational logics. In Canada, as an example, a key issue that I keep hearing between major players in the content industry is that the metrics used to identify popular artists is defunct – it needs to be broadened to capture the long-tail of media downloads. The Canadian situation sees (relatively) little overhead, and has distributed over $100 million dollars since the levy’s inception. This isn’t a small amount of money.

This model does recognize that labels will get some of the money out of the levy system should they own the copyright of the content being downloaded. There isn’t any reason why they shouldn’t receive some of this money from a (layman’s) legal perspective. If this is perceived as a problem, because it ignores the artist, then that’s a larger issue of copyright ownership and not a reason why the levy itself is unworkable. If artists retain their copyright, they get the cut of the levy fund instead of a label. If there is a strong desire to change how present copyright systems work (and I think that there should be) that doesn’t undermine the levy-system; we should reform rather than scrap the levy, and not let copyright/label issues motivate a refutation of the levy model.

Perhaps most crucially, the levy is not designed to capture the ‘full value’ of media – it doesn’t correlate one download to one sale. Instead, it recognizes that there is partial value derived from a download that is unlikely to parallel the same value that is demonstrated when someone purchases a CD, pays to download music through a content delivery system like iTunes, or attends a concert. If we take the stance that P2P is a means of facilitating interest (and partial value) for content then it is obvious that the levy-system is not enough, on its own, to create a full-value stream between artists and fans. Additional monetization schemes must be adopted so that, after deriving partial value from a download, fans (as distinguished from downloaders) can provide what they perceive as full value to the artist(s) in question. How this is done extends beyond a levy system – while important, I don’t want to work through my thoughts on this extension right now.

I want to quickly turn to some very broad, general, and tentative ideas of how the collection and distribution of a levy might work. This is most definitely a work in progress.

While there is a worry that the levy could increase as various parties who are not currently supported by the levy system try to muscle their way in, I think that the solution is just to set a particular (fixed) rate for the levy, and have it increased by X (where X=rate of inflation, or some other clearly identified, relatively autonomous metric that is outside of the music market), and then an organizational metric for distributing funds. Figure that every Y years the distribution metric could come up for organizational review, with the review board being composed of a balanced group of people (e.g. members of the non-profit’s board, artist and label copyright holders, hardware vendors, fans, etc). Where it appears as though the non-profit levy-distribution group is overstepping their bounds, a judicial review could occur (as happened in Canada, when there have been attempts to extend our levy to portable music devices).

In terms of actually establishing how much money the levy should collect per device, I’m inclined to set a very low levy that is based on a per GB/TB calculation of the storage media purchased. These rates should be automatically adjusted over time per a formula that takes into account drive size increases by reducing the cost-per-GB/TB so that the levy doesn’t get prohibitively high. The algorithm for this can be developed prior to the levy being restarted/instated to avoid hidden surprises and put in the public domain. Participation from a wide community should be invited, with a recognition that any formula set is a compromise position; there is no way that everyone is going to be happy. Consensus is highly to emerge between all the interested parties.

On the whole, what I’m trying to express is that levies are workable, partial, solutions. They do facilitate artists getting paid for their contributions to culture. They do benefit the consumer because it is clear why the levy exists and how funds are distributed without adding complexity (for the consumer) at the point of sale. Levy systems do not try to equate one download to one sale, and instead recognize the partial value of downloaded media. The proposed hardware-centric levy does not require ISPs to spy on content as it moves across their networks. Further, as I’ve proposed the aggregation of data points, this levy system does preserve anonymity and privacy.

Does this approach ultimately result in consumers paying a slightly higher price for blank media/devices? Sure, but I don’t think that’s anymore unreasonable than paying the trivially small environmental taxes already associated with the purchase of new computer equipment.