Is it possible to uniquely identify an audio CD-ROM? I’ve “stolen” the disc ID code from the various CDDB clients I’ve seen, but I wonder how unique those values are.
I’m looking at setting up a simple CDDB/FreeDB-esque server to store audio information which can be queried in a sane manner. The results will be XML data.
Using XML data I can have individual details for each track. Ideal for soundtrack albums. Plus I can require clients to ignore attributes they don’t understand to allow for future growth.
I could use the ID + sha1 hashes of the song lengths I guess …. or even a SHA1 hash of the track-data (but I’m worried that read-errors could cause problems with data summing, plus reading each track will be slowish).
I see comments like this a lot:
Conceptually very similar to the ISBN number used for books, the ISRC (acronym for International Standard Recording Code) is a 12-character code defined to uniquely identify a music track song). Unfortunately, over the years very few authors have adopted this standard and currently less than one percent audio CDs use the ISRC.
The UPC (Universal Product Code) is a 13-digit number that uses the UPC/EAN bar coding standard to uniquely identify the whole CD. The same ISRC considerations apply to the UPC code: very few audio CDs currently use the UPC.
In related news two tickets for Radiohead + Beck are mine :)