Slashdot Mirror

← Back to Stories (view on slashdot.org)

CDDB Shutting Down Media Jukebox

Posted by CmdrTaco on from the remember-when-cddb-was-cool? dept.

shaun writes: "According to this thread on the Media Jukebox Talkback site, CDDB is refusing connections from Media Jukebox until the Media Jukebox guys sign an "exclusive agreement" to use CDDB's database. Taking a shared public resource private has destroyed their karma, but what can be done?" Are grip and xmcd next? How do you enforce exclusivity for an open source program? Everyone should use FreeDB instead anyway: It's everything that was once good about cddb, including that little free part that made cddb itself the defacto standard before it got too big for its britches.

5 of 156 comments (clear)

  1. CDDB claims a protected algorithm by Karmageddon · · Score: 5
    Many posters here are talking about having entered songs into the CDDB and they wanted the info shared.

    Just to clarify, CDDB doesn't claim to own the song titles (the record companies do?), they claim to own the encoding that turns the length of all the tracks on a disc (the "uniqueness") into a code number to be looked up in the database. You are free to take the song titles from their database. What they'll try to stop is any use of that algorithm.

    I just thought I'd clarify because it makes a difference to how to circumvent their theft of everyone's hard work. The distributed.net idea would not suffer from this problem, BTW, not to mention it would be fun. SETI@CDDB! :) I'm sure d.net would never go for it, though, they'd rather bore us all to tears with a yet longer attempt to crack some obscure n-bit variant of a public key system. (d.net: we know that things can be cracked by brute [yawn] force. do something more interesting!]

    1. Re:CDDB claims a protected algorithm by rcw-work · · Score: 5
      Just to clarify, CDDB doesn't claim to own the song titles (the record companies do?), they claim to own the encoding that turns the length of all the tracks on a disc (the "uniqueness") into a code number to be looked up in the database. You are free to take the song titles from their database. What they'll try to stop is any use of that algorithm.

      Whether this is true or not, it's pathetic. FreeDB uses the same algorithm.

      Anyway, this is a description of the algorithm just so you can see how stupid it is:

      The discid looks like a 32 bit hex number, but in reality it's not - it's an 8 bit checksum, a 16 bit number representing the total length of the disc in seconds, and an 8 bit number representing the number of tracks on the disc.

      It's mind-warpingly simple. Ignoring MSF offsets for the sake of discussion, this is how it's done:

      Get the length of all the tracks in frames, like so:
      22047 44492 69957 85152 113637 129910 148045 165852 178462 200282 215427
      Divide them all by 75 so you have track lengths in seconds:
      293 593 932 1135 1515 1732 1973 2211 2379 2670 2872
      This is the really really brilliant part - Add all the decimal digits together like so:
      2+9+3+5+9+3+9+3+2+1+1+3+5+1+5+1+5+1+7+3+2+1+9+7+3+ 2+2+1+1+2+3+7+9+2+6+7+0+2+8+7+2
      It's 161 or 0xa1. Convert the length of the disc and number of tracks to hex too (0xc87 and 0xc) and put them together - 0xa10c870c. This is only a little bit off from the real discid (a30c850c), and only because of the MSF offsets I skipped over.

      Read the cd-discid source code for the full algorithm.

      I would be utterly amazed if they could protect this algorithm in court - it's literally just addition.

  2. FreeDB needs help by Jamie+Zawinski · · Score: 5

    I agree with others who have said that all this whining about CDDB is pointless, since FreeDB exists, and has a huge amount of data in it already.

    However, FreeDB very much needs help. There are a number of bugs and horrible misfeatures in it, and the main developer seems to have close to zero time to work on it. Pretty much any time I've found a problem, his response has been, ``yeah, too bad I don't have any time.''

    Not that there's anything wrong with that: he's doing a big service to all of us by keeping it running at all. But, if any of you do have the time and ability, it seems to me that FreeDB desperately needs a co-developer.

  3. Interesting quote from freedb.org's site... by grunby · · Score: 5

    From: Freecddb's Why page
    "(Funny sidenote: One programmer told me, that his cd-player will be banned if he is refusing to display the CDDB-logo. His software is a console-based program (it does not produce any graphical output) for blind people...)."

  4. Re:Rape the cddb, make it ours again. by nickol · · Score: 5

    No need to ' query every single disk-id '. Patch software to use FreeDB as primary source, CDDB as secondary and to COPY contents found in CDDB into FreeDB.
    Then just sit down and listen.