Slashdot Mirror

If you want credit, you could claim it in a public metadata catalog keyed by file hash. That way, if someone obtains a file you created through a source that has no connection to/information about you (P2P trading obviously, but really any source other than your website), they can still find you based on the contents of the file itself.

I usually download stuff from the web as well, but if I could trust the files, P2P searching and downloading is awfully convenient.

You compile a list of top gnutella queries (no visualization required), but that would be more of a buzz index, wouldn't tell you what people were actually downloading or listening to.

You could track (voluntarily I'd hope) users' media player actions and compile top lists. I'd guess that playlist sharing companies like Uplister have or will do this, though I haven't seen it. Companies like Kick and MoodLogic may also be positioned to do something like this.

Finally, Bitzi (disclaimer: I'm involved) has an ugly most reported/highest ranked type page, though it bears no resemblance to the real world yet due to low traffic.

I can't wait for record labels to start promoting their wares on p2p networks and helpers like Bitzi -- hopefully that'd mean they'd have at least partially accepted the value of decentralized/uncontrolled distribution.

You compile a list of top gnutella queries (no visualization required), but that would be more of a buzz index, wouldn't tell you what people were actually downloading or listening to.

You could track (voluntarily I'd hope) users' media player actions and compile top lists. I'd guess that playlist sharing companies like Uplister have or will do this, though I haven't seen it. Companies like Kick and MoodLogic may also be positioned to do something like this.

Finally, Bitzi (disclaimer: I'm involved) has an ugly most reported/highest ranked type page, though it bears no resemblance to the real world yet due to low traffic.

I can't wait for record labels to start promoting their wares on p2p networks and helpers like Bitzi -- hopefully that'd mean they'd have at least partially accepted the value of decentralized/uncontrolled distribution.

In any case, I almost never download more than two songs of an artist through mp3.com or similar sites. The click-click-click-click for each song interfaces are a pain. If I find I like a band I'll do quite a bit of searching on gnutella et al before I tediously download from mp3.com.

The search/select/download interface that all P2P filesharing clients more or less have is just much more convenient. KaZaa/MusicCity and various gnutella clients are a pleasure to use.

The pain of downloading from the web might be worth it if the download web site actually contained a wealth of ancilliary information that placed the tracks in context, but it almost never does. Certainly not mp3.com artist sites. A low quality image and occasionally a silly description doesn't do it.

I do have a solution, which means I'm horribly biased. :-) P2P distribution is the way to go: low cost, high usability. An external metadata catalog is the way to get context: find out more about the files you've downloaded or are considering downloading, whether there are better versions available, the artist's web site, etc.

An external metadata catalog can also provide a near universal point of reintermediation, connecting the artist/publisher with the listener regardless of where the latter obtained the file -- a file metadata catalog can be keyed by a deterministic calculated file identifier, so that if two people have the same file, they get the same catalog record, and can interact with each other, and with the file's creator.

To get a hint of what I'm talking about, check out Bitzi. It's targeted at developers right now, but if you're a super smart creator/publisher you should be able to figure out where we're going. If you're aren't that smart (just kidding, if you can't figure it out it's our fault), I'd be happy to explain.

I'm biased though, as I'm working on a global metadata repository.