[mac_the_knife]

To be accurate... it's a prof of economics at Harvard and another at UNC-Chapel Hill who have collaborated on this study.
The study is here: http://www.unc.edu/~...ers/FileSharingMarch2004.pdf

A NY Times article breaking down the study and some arguments, pro- and con-, can be found here: http://www2.mrbrklyn...ces/filesharingproventonothurt[u]recordsales2.html
I've read the study. I've sat and heard a debate between Prof. Strumpf (who co-authored the study) and Stan Leibowitz, another respected economist out of UT-Dallas (I think?). I've emailed Prof. Strumpf and asked him questions about the paper. IMHO, their methodology is ridiculously suspect. They state that a school holiday in Germany resulted in an increase in file-sharing activity on the OpenNap servers they monitored, but yet they couldn't see a resulting decrease of sales over the same time period, so therefore, file-sharing had no effect. I asked Mr. Strumpf what if sales preceding the holiday were affected, such that there would be no effect on post-holiday sales because it was already built into the data. His answer was to the effect that, "it's very complex economic stuff, just trust us." Well, I don't trust them. And neither do Mr. Leibowitz and a few other economists.
Then again, I don't trust the RIAA either. Where does this leave me? I don't think there's truly a way to prove with any amount of precision the effect file-sharing has on record sales. Furthermore, it's even less possible to determine whether a specific act of file sharing (such as an individual P2P user offering specific mp3 files) affects sales of those same records. Fortunately (or unfortunately), the copyright law provides for the award of statutory damages for copyright infringement. So for the purposes of the lawsuits, in many cases, it doesn't matter how much damage filesharing does. The file-sharers have reproduced and/or distributed copyrighted content without the permission of the copyright owner. That by definition is infringement, and statutory damages would apply no matter the extent of the "actual" damage to the copyright owner.
Let's get to the article: anonymity over the internet. While I don't agree with the RIAA's Gestapo-esque handling of the entire situation, I would like to pose a thought: why should IP addresses NOT be used to find out the identity of someone committing unlawful conduct? If a car with tinted windows hit-and-runs another car or pedestrian, why can't I use the license plate number of the car to figure out who did it? If someone's prank phone calling my house, and I have caller ID, why can't I use the phone number on my caller ID to figure out the identity of the caller? What about threatening mail where the sender leaves his/her home address? Or use of an employee ID number to figure out who stole company property after hours? Why should an IP address, used at a specific date and time, be any different from those forms of identification?
Now, one could argue that the piece of identification inaccurately identifies the actual culprit (perhaps the car was borrowed, perhaps the prank call was from a payphone, etc.) But that's a debate on the facts. According to this ruling, the 8th circuit won't let the RIAA even get to that point. Now, if I'm missing something, by all means, tell me.
Jeff, I've debated you on copyright/DRM issues many times in the "old forums." I think you make logical, compelling arguments for your stance on copyright and general tech policy, which I think lies a little to the more "chaotic" side than mine. However, I think we would agree on many specific issues. You and I probably agree that copyright is too dagg namm long. We probably also agree that Grokster should not be held liable for the infringing uses of its users, and that copyright laws today are set up more to benefit the "banks" or financiers of artistic creation than the actual creators, as it should be set up.
That said, your approach to dealing with copyright infringement seems to be that the RIAA and member companies should do nothing on the "enforcement" front, drop prices, come out with better product, and hope and pray that people don't infringe copyrights as an alternative to purchasing, and if it happens anyway, just look the other way, because it's going to happen whether you police it or not. Is that accurate? If so, I wholeheartedly disagree. I think that record companies should make more compelling product and price it more competitively. But you can't compete with free, and the copyright owners should enforce and try to protect their copyrights.