Freedb.org Returns to Life

Trogre writes "The recently troubled free CD database freedb has been picked up by a group called Magix. From Kaiser's blog: 'Following my announcement that I would like to let freedb go, I was approached by many interested parties ... Even if I shall no longer be actively associated with freedb, I shall continue casting a critical glance on freedb's future. The decision in favour of MAGIX has given us a new prospect of further development, offered a congenial and comfortable atmosphere during difficult negotiations, and provided the newly implemented hardware with generous capacities.' This might be good news since Grip still doesn't support MusicBrainz."

Comments

Freedb2?

[OverlordQ]

And what's going to happen to Freedb2 (site that one of original founders forked from Freedb) then?

Re:Freedb2?

[Trogre]

It seems to have stagnated, possibly due to immense suspicion from everyone except the single developer. The project had been using modified GPL'd code from freedb without disclosing any substantial source code. It was this issue that caused the head freedb dev (Kaiser) to 'take action' against this guy without the consent of the other two devs who left and AFAIK are now part of neither project.

They seem to have taken it a bit personally if you ask me, but then again there might have been more going on than publically stated.

Re:Freedb2?

[Horar]

Once again I would like to set the record straight. http://freedb2.org/ was written from scratch by myself with no reference to any of the original freedb code whatsoever, and I am the sole copyright owner on all of that code. It was derived strictly from the published cddb specifications and the contents of the database itself. You are right in thinking that there was much more going on that was publicly stated.

http://freedb2.org/ continues to thrive and grow and has been very well supported. I would like to thank everybody for that, and can guarantee that you will be able to enjoy the superior levels of service offered by freedb2.org for a long time to come. Please feel free to email me directly should you have any specific questions about it.

Re:Freedb2?

[Jay Carlson]

http://freedb2.org/ continues to thrive and grow and has been very well supported.

Cool. So where can I download your database?

I'm not joking. I can download wikipedia. I downloaded a couple versions of the original cddb back when we were all running off Sparcstations.

The way we got here was to freely exchange metadata about CDs we own. freedb2.org doesn't say anything about how to get at the data behind it. In fact, it doesn't really say anything at all about where its data came from. (Before you claim that you can't pay for the bandwidth to support downloads of your full database, trust me, I can find somewhere to host it for you.)

I typed in plenty of CD metadata. I showed you mine; so show me yours.

freedom at any price

[User 956]

Following my announcement that I would like to let freedb go, I was approached by many interested parties

Well, given the name, I hope they got it for free.

MusicBrainz

[BobNET]

Most FreeDB/CDDB clients can access MusicBrainz through a CDDB gateway: http://musicbrainz.org/doc/CddbGateway

MusicBrainz is superior to FreeDB

[pen]

CDDB and FreeDB are old news. MusicBrainz is by far superior. It accounts for different release years, different formats, multiple artists, compilation albums, etc. "Why would I need to use your site? What's wrong with FreeDB?".

I'm not affiliated, just another happy user.

Re:MusicBrainz is superior to FreeDB

[Jeffrey Baker]

Oh no! They don't support _genre_! Call the metadata police!

Genre is a broken concept and everybody knows it. Practically every CD ever released is listed in FreeDB under half a dozen different genres, all entries having slightly different errors. No FreeDB booster was ever able to sufficiently explain to me why, for example, Hotel California should be listed under New Age.

The multiple genre CDDB defect has this amusing side effect in all FreeDB-reliant CD rippers:

Multiple results found. Please choose:
1) The Same Title
2) The Same Title
3) The Same Title
4) The Same Title

The system is practically useless for anyone who actually cares about consistency in metadata and/or has a large collection to rip.

Re:MusicBrainz is superior to FreeDB

[shish]

No FreeDB booster was ever able to sufficiently explain to me why, for example, Hotel California should be listed under New Age.

IIRC, it's because the database entries are plain text files named "[genre]/[CD checksum].txt" -- if two different CDs have the same checksum, then they have to be in different genres. Also, one can't update the genre of a CD once entered, you can only create a new entry somewhere else.

Re:MusicBrainz is superior to FreeDB

[Anonymous Coward]

I agree with you on the 'genre' field. But what I'd like to see is support for the id3v2 'composer', 'conductor', and 'lead performer(s)' fields.

Man, this takes me back...

[bennomatic]

One of the first sites I was asked to design was a CD trading system which never went anywhere. The founders were too cheap to license a database or build a system to grow one organically, but they wanted names of all of the artists, albums and song titles that people owned on the CDs they wanted to trade.

So at their request, I built a system which would send a query to the CDDB.com page (back before they became Gracenote), excise out the useful data, and store it, one album at a time.

I got it through proof of concept, and then explained while it was technically possible to continue in this vein (I had probably pulled three albums correctly in testing, one more at the demo), they would be fools to continue because the page format could change at any time, and if the fine folks at CDDB figured out what we were doing, the owners would be begging for a lawsuit.

They still didn't want to do the right thing, so the project eventually got dropped (I think Napster made the CD go bye-bye), I moved on to greener pastures, and the owners went on to found a handful of failed dot-bombs, I guess.

Ah, the good old days.

Thanks, Michael

[MikeO]

Hats off to Michael Kaiser for being the caretaker of freedb for the past 7+ years and remaining true to the community spirit the service represents.

Many digital music collections, mine included, owe a lot to freedb.

Re:Thanks, Michael

[Horar]

Please put the credit where it is due.

Joerg Hevers (and more recently, Ari Sundholm) did all the work while Michael Kaiser merely took all the credit and money. Why did you think they all resigned the way they did?

Re:They should make the database public

[sdo1]

OK... public appology here. Me = moron.

http://www.freedb.org/en/download__database.10.htm l

I could have sworn that I'd loooked for that before and couldn't find it.

-S (feeling like a dipshit)

Re:Grip?

[Trongy]

Is Grip still creating invalid ogg files by default (with ID3 tags in the header)?

Re:Not clear that the GFDL/GPL can cover the datab

[indifferent children]

The legalities of applying the GFDL or GPL to a database like the one managed by freedb.org are unclear.

IANAL, however as I understand it under US law: it is not possible to apply any license to this database, because the licenses are grants of rights, based on the copyright of the owner. However, copyright does not apply to this data. No one can hold a copyright on 'facts', only 'expressions'. A clearly stated (though no more authoritative) explanation from the 'copyright' article on wikipedia:

Compilations of facts or data may also be copyrighted, but such a copyright is thin; it only applies to the particular selection and arrangement of the included items, not to the particular items themselves.