"Ownership" Is Why Divx and MP3 Matter.

They were small news items in June: Divx proved to be a failure, while MP3 proved to be legal. But behind the news was the fact that two technologies have forever changed users' conceptions of file and data "ownership." Underlying Divx is the core technology of video-on-demand, albeit in this case, it involved an asymmetrical network whose uplink was a phone line and whose downlink was a disk.

The Divx player is a DVD-Video player with a phone jack. You cable the player to your TV, plug in a phone cord, and the player dials up a Divx database, registering your name and credit-card number. When you buy a Divx movie disk--for about one-fourth the price of an ordinary DVD movie--the player lets you watch it for 48 hours. To see it afterward, you must key your player to phone the database and authorize more plays: a pre-authorized credit card transaction. After a few viewings, though, you'll have paid the equivalent of the retail DVD's price, and incur no more charges: the disk is yours to watch forever or throw away.

The core technology behind MP3 is file-compression. A Compact Disc needs all of its 650MB to store one hour of audio data, such as music; but if that data is compressed, a CD can hold almost four times as much programming. More important: a file of digital music, compressed with the MP3 algorithm or something similar, can be downloaded from computer to computer in about one-fourth the time it would take to transmit an uncompressed file. And from there it can be burned into a CD-R disk, or stored in a handheld "player" with semiconductor storage onboard. Why did the fortunes of Divx and MP3 diverge?

In the case of DVD, the problem lay with the proponents--chiefly the Circuit City consumer-electronics retail chain and a couple of Hollywood studios--who under-stood video-on-demand technically, but not emotionally. Twenty years of experience with videocassettes should have proved, by now, that people either buy a movie or they rent it--there's no model for a transaction that splits the difference. The Divx approach was especially impractical: unless, for example, prospective customers already had a satellite-dish or one of the new digital cable-TV boxes, it was unlikely that they had (or wanted to install at their own expense) a telephone jack convenient enough to their TV set to handle the uplink.

In the case of MP3, it was the opponents--chiefly the Recording Industry Association of America (RIAA)--which understood the technology, yet utterly misread the movement behind it. To them, file compression isn't a convenience; it's an aid to piracy. Anyone could already copy a CD onto a CD-R disk. With MP3 software (itself freeware or shareware) they can do that four times faster or stuff four disks' worth of programming into one.

The RIAA felt the law was on their side: a 1992 statute that they'd lobbied for, enabling them to collect royalties on the blank DAT tapes and CD-R disks that are sold (at royalty-inflated prices) with the stereo-component versions of those recorders. So they asked the court for a preliminary injunction against sales of MP3 players. What they really wanted, I believe, was enough time to invent a way to charge and collect royalties from MP3 downloads. But in June, the U.S. Court of Appeals for the Ninth Circuit ruled unanimously that the first commercial handheld MP3 device (the Rio MP300 from Diamond Multimedia Systems) is not primarily a recording device: it's a player, they said, and therefore not covered by the 1992 law.

Like the Software Publishers' Association, the RIAA has never acknowledged that ordinary people making home recordings are the smallest fish in the sea. The pirates they should be going after--and whom they never seem to catch--are those who build disk-replication factories and stamp out CDs (or CD-ROMs) by the thousand. Moreover, at least three other audio compression-and-down-loading schemes (including one from Microsoft) have emerged since the Rio suit was filed. And many small-time musicians who can't get big-time recording contracts are giving away--as loss-leaders--MP3 versions of what they previously would have put on "demo" tapes, in hopes of getting gigs and/or selling their own CDs. So, what lessons should the storage industry draw from Divx and MP3?

First, the good news: networks of the future will carry video and other large files, so networks will never again have "too much" capacity and administrators will have to add more capacity than they thought they needed, and probably more often than they'd planned.

The bad news is that the line between "infringement" and "fair use" of copyrighted material is fading; so the concept of data "ownership" is evolving. Who can say or predict, any more, what is the organization's data and what is one's own? Users will create and send "their" files around even the most secure networks, whether the administrators officially permit it or not.

That "bad" news won't be too bad, though, if network administrators, developers, and storage vendors listen to users' suggestions for improvements, and empower those users to manage "their" files themselves, on some safe level. Why risk disrupting the entire enterprise to stop something that can't be stopped anyway?