Slashdot Mirror
Freedb.org Ending
haroldag writes "Freedb, the free music database used by tons of CD ripping software, has been shut down due to a disagreement among its developers. One of its developers used a data dump from the original freedb.org and is providing the service at freedb2.org, though, and will be adding features and posting them at his site as they become available. Unfortunately, a database dump or source code for freedb2.org is yet nowhere to be found."
That sucks. I hope that freedb2 will be compatable with the old freedb protocol. Pretty much every open source tagger/ripper/whatever I know of uses freedb.
:)
Then again, maybe it's time for MusicBrainz to take over.
The quality of submissions were total, utter shit. You'd be better off typing in the info yourself.
Not so, well, not really. If you used it as a starting point, and checked the entries against the CD you were ripping then by and large the entries were really good. (Some freaky choices in categories sometimes)
Where there were issues, it was far easier to quickly edit one or two entries or the artist name etc. rather than type the whole thing in.
It is/was a great service.
I just now remembered what I did once, quite a while ago:
.wav, burned them to CD using gcdmaster, and ripped them to OGG, only to find that the ripping program actually guessed *correctly* the album and the titles of all the included songs.
I recorded some of my (difficult-to-find) LPs to
Note that even though I marked the beginning and end of each song manually, it still found the right titles. freedb really rocks!
On one hand, I guess I can't criticize them too harshly, because it's not as though many of us (myself included) who are using the service were paying the developers any for their time, on the other hand, it seems a little unfortunate that the one developer decided to just abruptly pull the plug on the service when it was in use by so many people, without trying to see if there were others in the community of users willing to take over the project, if he no longer wanted to manage it.
It seems like there were three people on the project, and two of them wanted to take it non-free, one didn't; although I'm glad the remaining developer didn't go along with the other guys if they really wanted to make it non-free, I can't really understand why he would choose to just kill it outright rather than find people who were willing to maintain it, if nothing else.
I'm not sure whether this shows a shortcoming of the collaborative development model or not. It seems like it might be -- although I suppose projects managed by a "benevolent dictator" are also prone to shutting down if the person moves on / dies / whatever; however it seems like the a not insignificant number of projects that are run by teams without a clear leader close due to 'personality conflicts' over time.
On the other hand -- what is it with CD meta-databases and going non-free? Is it just that they seem like tempting revenue sources or what?
"Ladies and gentlemen, my killbot features Lotus Notes and a machine gun. It is the finest available."
Joerg on freedb:
For almost two years now Ari and I have supported a developer from Australia, who was working on the next generation of the freedb server, which would have overcome most of our current technological problems and offered text searching. This was the biggest chance for freedb in years. Unfortunately there have been rising tensions in our team about the question, how long we should support a development project, which has not yet been made open source by the developer and which is not yet running on freedb servers. Last weekend the line was crossed by the founder of freedb, who owns the domain, when he took action against that developer without talking to the rest of the team first, while we were still trying to find a solution in everyone's interest.
Well, if I'm reading between the lines correctly:
1) Ari and Joerg support some australian guy developing the "next-gen" freedb for two years
2) Australian guy doesn't want to release it as open/free for freedb (or all three?)
3) Ari and Joerg have either been suckers or part of an attempt at pulling another Gracenote
4) Kaiser won't play ball, it's freedb or no db at all. He finally tires and goes to the source.
5) The play is called, Ari and Joerg leave because the gig is up.
To put it this way, I would not be surprised to see another CD database show up soon, lead by an australian and maybe with a few more anonymous employees. Either that, or they're been really gullible. Never ever trust someone who says they'll open source it "soon". If that is their true intention, they would have no problem being open about it all the way. The only reason not to is when you're pulling a bait-n-switch like here. It seems clear to me that they expected it to be open source ("not yet open source", Joerg), it wasn't ("did not seem to be kept free", Kaiser) and that tore them apart.
Live today, because you never know what tomorrow brings
Wikipedia is busily replicating GraceNote and IMDB, by hand, and not too well. They're using a wiki to do the job of a database. Some music types from Wikipedia should take this database and the data in Wikipedia and make something useful out of it.
Personally, I think that Wikipedia needs something like "Wikipedia Music and Movies", to which all content associated with music, movies, TV, and the people involved in the industry would be moved. More structured than Wikipedia Encyclopedia, Music and Movies would have standard database formats and slots for music and movies, indexed so that you could see all movies by some director or all songs by some musician. Wikipedia can't do that, but IMDB can.
Then Wikipedia needs "Wikipedia Atlas", a map-based system, for all those "State Route 93" entries. Wikipedia isn't spatial, and space is what keeps everything from being in the same place. An atlas system would be able to handle an endless number of "my favorite restaurant" articles. Wikipedia Travel already has something like this.
With that out of the way, Wikipedia would become more like an encyclopedia. Right now, it's drowning under the incoming cruft.
For almost two years now Ari and I have supported a developer from Australia, who was working on the next generation of the freedb server, which would have overcome most of our current technological problems and offered text searching. This was the biggest chance for freedb in years. Unfortunately there have been rising tensions in our team about the question, how long we should support a development project, which has not yet been made open source by the developer and which is not yet running on freedb servers.
Is it true that you accepted money (which is how I interpret "support") to do open source development and then did not release the code? I'd like to hear your side of the story.
Accepted money? Hell no. I've spent 6 months full-time out of the last 19 months, and most of what little money I had, to develop freedb2.org. The fact is I've had very little useful support from anyone and one hell of a lot of obstruction.
The so-called freedb team has always been incompetent and dysfunctional and I have made it my mission to fix that. Kaiser and Hevers have behaved like a pair of spoiled brats from beginning to end and now you see the result of that splashed all over the internet.
Whatever freedb2 ends up becoming I can guarantee you that it will be even more free than freedb.org. Please see http://freedb2.org/news.html for the specifics on that.