- The direction of digital music changed last week when Bertelsmann CEO Thomas Middlehoff stood on a podium with Napster founder Shawn Fanning and CEO Hank Barry and declared: "I'm happy I met these guys."
He was talking after Bertelsmann's decision, announced last Tuesday, to invest in Napster and steer it towards a membership-based model which would pay royalties to record labels, artists and producers.
However, the deal is no guarantee of a solution. Seventeen record labels are suing Napster, and if they are successful that could put the network out of operation.
And alongside the strategy announced last week, Bertelsmann – with 200 record labels, technology companies and electronics manufacturers – has been pursuing another that is designed to scupper Napster's system.
The Secure Digital Music Initiative (SDMI), was created to develop an international standard for protecting music copyrights on the Internet. It has been quietly plodding along for almost two years and, so far, its main achievement has been the invention of an encrypted "watermark" that would be placed within the code of a digital song.
The watermark would mean fans could not listen to music unless they had purchased the rights to do so.
In September, SDMI – based in Turin – announced it had chosen six technologies as candidates for the international standard. It decided to test them on their harshest critics: the hackers themselves. On 10 November, Leonardo Chiarglione, executive director of SDMI, will announce the results of "Hack SDMI". The group invited hackers around the world to try their hand at "cracking" the six technologies being considered for the international standard. Four of the technologies were digital watermarks and two were technologies that would allow users access to the music if they own the CD. The reward for cracking each technology was $10,000 (11,900 euros).
But just weeks after the challenge was issued, a group of computer scientists at Princeton University in New Jersey, Rice University in Texas and the Xerox research lab in Palo Alto, California claimed to have "successfully defeated all four watermarking technologies". They said they planned to post their successful hacks on a Web site.
SDMI officials argued that there was no way of knowing how successful the scientists had been in removing the watermarks without damaging the music until the fidelity of the tracks had been tested.
Edward Felten, professor of computer science at Princeton and leader of the attack, maintains his group succeeded.
"We know the sound quality is adequate; we can tell by listening to the music." he says. "We also know that our attacks can be improved, leading to even better sound quality."
According to SDMI management, there are two criteria for success. First: the removal of the watermark must not significantly degrade audio quality. Second: the removal must be easy to execute – easy enough that anyone, not just a group of university professors, can use it to beat the system. That second criterion may save SDMI from having to go back to the drawing board.
"Just because a near-genius can hack the watermark, does that mean that the rest of the world can use the hack software?" asked a sceptical Jonathan Potter, executive director of the Digital Media Association in Washington, DC. "The typical consumer is not interested in hacking or stealing," he added.
But as long as Napster remains in its present form, it only takes one rebel genius to hack a song, upload a track, and instantly render a technology two years in the making obsolete. This is the challenge SDMI faces – and unless Bertelsmann succeeds in changing the way Napster operates, it may be swappers, not hackers, who are the undoing of SDMI.