Kazaa, Brilliant move will spanner MPAA, RIAA

THIS LAST WEEKEND, Brilliant Digital Entertainment, a company closely associated with Kazaa, the file swapping network, announced a network within a network. This new entity is designed to offer paid content, as opposed to the usual free file trading that has got Hollywood, the record companies, software companies, and just about everyone else in such a tizzy. So far so what? This has been tried time and time again, and most of those companies couldn't make a go of it, or were mercilessly crushed by the record companies. The lucky ones were cold-shouldered and left to flounder in a limbo of subsistence living.

The latest announcement has been covered by almost everyone as just another in a long line of companies destined to be killed for trying to legitimize an industry that the collective forces of the MPAA and RIAA have sworn to crush. It however is different, very different, and could change the way we see content distribution on the net, if they live long enough. How are they different? Content locality, pure and simple.

Almost every system of paid content on the net has taken the model of a central server, or servers where the content is located. When you pay, you are authorized to download or stream something from the server. Most of the companies have a server bank in several geographically diverse areas to speed up transmissions to the purchaser. Some of the more adventurous ones use boxes placed at ISPs, a more sophisticated version of Akamai. Either way, you go to the source to get the files, and these sources have to have one hell of a big pipe to support the timely download of full length CDs or movies. It is hard, or at least expensive to scale this architecture up to a level where it would support millions of people.

This is where Brilliant is, well, brilliant. Their plan is to send out content to users of Kazaa, in an encrypted form, and ‘hide' it on their machines. The users don't have any idea what content they have on the computers, and while technical details are still scarce, any single user will probably not have anything close to the full file on their system, this would only encourage hacking. When someone buys a file from the network, it does a simple search across the hidden, encrypted files, and downloads them just like any other Kazaa file. The users never realize that they are being used in this fashion, but do get a small fee for doing it. If Brilliant is not overtly stupid, and all signs point to them not being so, they will keep the bandwidth used by any single individual low, and give them less than the bandwidth would cost them on the open market. Since uses of the network tend to buy net access on a monthly, all you can eat plan, they don't pay the additional uploading. It is a win/win situation for both the users and the network.

Brilliant gets a free, secure, worldwide network for content distribution that is already in place. If they use it intelligently, and load the users with the content slowly over time, say two weeks before it is first sold, they can avoid the spike in first day downloads, and users of the network won't be crushed by the uploads. Again, win/win.

This network would have enormous advantages over a traditional site. Lets use the sale of a streamed, first run movie for example. If the viewing of the movie needs 1Mbps to be viewed, you can either do it through a single fast server that can support the upload, or 1000 servers that can support a lowly 1kps stream. If you have 1000 people watching this movie at once, you need a gigabit pipe to support them. Last time I checked, that wasn't cheap. The Brilliant network needs a million users, all of which send at 1kbps. Last time I checked, the user base of Kazaa was far above that.

As things scale up, the cost to the traditional networks gets higher and higher and higher. Brilliant, on the other hand, just adds more users to the pool that they are already have in place. When the rush of users dies down, the old school networks have a very expensive pipe that is going mostly unutilized. Brilliant has nothing to pay for, no overhead.

In network provisioning, you have to plan for peak usage, not average. If you don't have the capacity to support your peaks, service suffers, and you annoy more people at once than any other time of the day. Not good. Annoyed people don't buy from you again. Brilliant's advantage here is so huge it is hard to put into words. They can change the way things are done, if they don't screw it up initially.

The biggest challenge they have to face is not technical, but marketing. They are the sworn enemy of the very people they want to serve. This is not a good starting point. Add in the fact that the consortia that are targeting them use every legal and quasi-legal tactic in the book to spend perceived enemies into the ground, and you have a real uphill battle. If Brilliant Digital can overcome these substantial hurdles, what they will have is simply brilliant. µ