Appealing To Greed In Digital Music Distribution
A Proposed Model
May 27, 2002

By Darryl Ballantyne

Introduction

First of all, this is a concept document, not a technical document. I don't have nearly enough technical knowledge to spec a system such as this. Unless there's a showstopper of a technical issue with the implementation, please don't nitpick.

Secondly, this isn't the be-all and the end-all of digital music distribution. I'm not trying to say that this is the only model that would work; I'm not even saying (at this point) that this model will work. There are a lot of other good ideas out there, and I believe that a number of them have the potential to succeed, if those holding the content would (or could) only play along. To that end, don't email me with "Wouldn't you rather have an all-you-can-eat subscription?" - yes, sometimes I would. Sometimes other motivators, discussed below, take over which steer me away from subscriptions. Call me single-minded if you will, but I'd rather drive discussion about this model in particular.

Lastly, please let me know what your impressions of this model are! Would you use it? Would it be a primary source for music, secondary, tertiary? Does it appeal to you? Any and all comments are welcome.

Anyway, on with the show.

Digital music and its distribution has become somewhat of a touchy subject in recent years. The RIAA and its members are up in arms over security concerns, regardless of the fact that they are destined to fail. Third party companies try to push forward their own visions of distribution - without being able to secure licenses from the so-called "Big 5" (Sony, Warner, Universal, BMG, and EMI). Governments threaten to intervene, causing the majors to release unusable services (pressplay and MusicNet). Consumers stay away in droves.

In the meantime, questionable file-sharing services, following in the footsteps of the now-defunct (or effectively defunct) Napster, flourish. Hundreds of millions of songs are downloaded each day, and the record industry - including the artists - doesn't see a cent from it. Meanwhile, record sales fall (or rise, or plummet, or skyrocket) as a result.

It's hard to compete with free. And, like it or not, the free alternatives will not go away. Even if the FastTrack services (Kazaa, Grokster) are somehow shut down (an unlikely event), truly decentralized systems (such as Gnutella and all its derivatives) will survive, and the technology and usability will improve.

So how does a service sell something that is otherwise available for free? What kind of service will convince an ordinary music consumer to leave piracy behind, and open up their wallets? What could possibly motivate them to do so?

Greed.

Yes, greed. Not on the part of the record companies (well, somewhat; that is, of course, going to be their motivation as well), but on the part of the consumer. Greed, one of the basest of human emotions, could save online music distribution.

The Keys To A Good Service

There are a number of basic elements that must be present in any distribution system for it to be successful. They are:

• Availability Of Content.
  The service must contain as much content as possible. Without content from the major labels (who control approximately 85% of all music currently in release), it's significantly more difficult to attract the critical mass of users necessary to make a distribution system successful. Just ask eMusic. 
• High Sound Quality.
  A critical aspect of being able to charge for digital music is the sound quality of the files. Files must be CD quality, or as close to it as possible - ideally, I'd be looking for the equivalent of a 192kbps MP3. 
• Reliable Service. 
  Available 24/7, all over the globe.
• High Speed Service.
  Your best customers will be those with high speed access; serve them well. 
• Ease Of Use.
  The service has to be so easy that my grandmother could use it.
  
  She used Napster. 
• Portability.
  Users must be able to transfer songs to portable players, burn them to CDs, etc. - they've bought and paid for them. They must be able to do with them as they please (without breaking the law).

How It Works

The basic premise for the system is as follows:

1. Peer-to-peer distribution system (for high speed, multi-homed downloads, and maximum content availability)
2. Per-download payment - prepaid or otherwise. The debate on how much to charge per track can follow at a later date. 
3. Each track is encoded with a unique ID, linked to a central database that contains all the necessary information for each track. The system then records the track downloaded, the account downloading it, the account serving it, timestamp, etc.
4. Revenue from the fees collected for each download are distributed as follows:
   
   Copyright holder(s): 50%
   Service: 25%
   Person(s) who served the song: 25%

The key of this system is that the user hosting the song receives a "distribution fee" for serving the song for the recipient.

This distribution fee provides a major incentive to users to connect to the network. With this distribution of funds, a user need only serve four songs in order to download one for free. High-speed power users will likely get a large portion of their music for free, as they fuel the distribution for the rest of the users.

In theory, this incentive should bring the broadband users with large digital music collections over to the network, providing the initial influx of content. By shifting their collections from the free P2P networks to this network, second-tier users (dial-up users or broadband users with small collections) will be faced with less available content, slower downloads, and more frequently interrupted downloads - prompting them to shift networks as well.

This second tier (and a potential third tier of infrequent users) will be the primary set of users fueling the revenue, as they are unlikely to serve enough files to supply themselves with free downloads.

Overall, the network brings in 25% of the purchase price for each song downloaded - potentially resulting in substantial revenue both for the company and copyright holders.

Populating The Network With Content

One of the major issues with this type of system is how to populate the network with the necessary content, in the necessary format - a classic chicken and egg problem.

However, there does appear to be an easy solution to this problem.

The software should include the ability to encode digital files in the necessary format from CDs that the user already owns. This ability could draw on a database system such as Gracenote's CDDB system to identify the disc, and verify that it is an original copy. It would then digitally encode the files in the necessary format with the necessary DRM (discussed later). This way, people with large CD collections could quickly amass a huge, legal collection of digital music, ready to be distributed - and ready to be profited from.

Record labels could also encode all of their content, and connect to the network to make it available - thus increasing the copyright holder cut to 75% for each track served through the label's connection.

DRM

Leaving aside that it would inevitably be cracked, DRM (Digital Rights Management) can easily be a component of this system. However, the DRM must allow for peer-to-peer distribution, and all the features outlined in "The Keys To A Good Service" above. DRM could even be incorporated into the tracking and payment functionalities of the system.

One of the drawbacks of using DRM in any music distribution system is the proprietary nature of the format. This can (and should) be overcome in two ways: 1) licensing (at zero or next to zero cost) as many players as possible to support the format, and 2) not going overboard with the restrictions embedded in the DRM system (related to the features discussed above). If DRM does not inhibit the user's regular usage and enjoyment of the music file, there should be minimal backlash resulting from its use.

A DRM solution for this system can incorporate the necessary ID numbers for each track, as well as ensure that only (1) full tracks, and (2) high quality tracks, are available. These two elements are also important to ensure a high level of quality and service for the users.

Food For Thought

Any system such as this has the potential to succeed gloriously or fail miserably. Feel free to crunch numbers as you wish; I will leave that as an exercise to the reader, aside from one quick example:

If the price per track is $0.50, and 1 billion songs are downloaded per month (a very high estimate, yes, but only half of Napster's traffic in its heyday), then the company's net revenue would be $125M/month, or 1.5B/year. Copyright holders would stand to gain twice that.

The service can also expand beyond just serving audio tracks. The data mining possibilities are endless. Targeted marketing initiatives could peddle t-shirts, CDs, concert tickets, and other artist paraphernalia.

User incentives could be expanded to include a second-tier referral fee, introducing a viral marketing aspect to the system.

Other forms of media could be made available at different rates, such as books, movies, and software.

A dynamic pricing ability could be incorporated, allowing users to adjust their distribution fee - effectively creating a digital marketplace where providers compete based on price and bandwidth.

The possibilities go on and on…

Conclusion

There are obviously many unanswered questions, and unresearched aspects to this proposal. However, I believe the underlying model could bring legal digital music distribution to consumers in an effective and satisfying manner.

The single largest obstacle is getting the major labels to agree to this type of system. Without major label content, any digital music distribution system is effectively stepping to the plate with two strikes already.

Obviously, this isn't a system that can be built in a basement. Major backing is needed for the system to be built, and for it to be launched.

While I'd love to build something like this to see if it would work, I don't have the necessary resources to get very far with it. If anyone does, I encourage you to research the concept, and run with it.

This article may be distributed freely in whole or in part, so long as it is not modified and credit is given to the author (Darryl Ballantyne). Also, it would be nice if you linked to this page!