Tom Coates suggests an app to add to LazyWebCorp’s Q1 2003 milestones – an "ID3alizer" that would enable you to batch process a lot of that pesky tag-adding we’ve been discussing here recently. Sounds like it might work nicely as a brute-force, lo-fi (cheap!) alternative (or precursor) to the collaborative-expert-based service I outlined previously. Would do some of the job – if it can automatically identify the track accurately (which seems to be getting easier these days, plus as Tom says, you tend to have some of the data but not a coherent, complete, usable set). Then we need something to infer the more complex metadata around music … perhaps an extension of CDDB/FreeDB to include more information, which in turn could pull in metadata from the likes of The Covers Project, for example (which I note is going from strength to strength).
On this subject, Ben Magnetbox commented on my previous post, pointing out that the Mixmatcher project he’d worked on is very much open to ideas as to future development. He said:
"Mixmatcher is being worked on, but slightly dormant as well … I would like it to become something more (or less?); something extremely useful behind the scenes, with just enough use and entertainment to keep people using it. I am extremely open to ideas as to how to accomplish that task, so please e-mail me if you have any ideas. Should it become like a CDDB of metadata? How do you make people want to enter in all that information?"
As the man says, email him – also feel free to post comments here. Perhaps we can come up with something.
Screw the Lazyweb, let’s do it. Who wants in?
I have set up a discussion area on this topic here.