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."

Comments

Re:Nothing to see?

[haroldag]

Actually, there is some code out there: http://asmith.id.au/mod_libpq.html

Re:Damn.

[Otter]

Freedb is a knockoff of cddb, so I'd imagine that the grandknockoff is going to continue with the same protocol.

Re:Freedb sucks anyway

[Anonymous Coward]

So which one am I supposed to choose?

The one whose cd hash matches your cd? [insert picture of guy attempting to slit his wrist with an electric shaver, caption: "You're doing it wrong"] Each of those hashes are (supposed to be) a completely different disc, and in the case of all these different hashes, I suspect that they're from people who got a copy "ripped" from their friend, except instead of an actual copy, the guy tooks some mp3s from kazaa and burnt a cd from them. Recipient discovered that freedb didn't have an entry for this bogus disc and made one.

Personally, I've been wishing for a long time for cddb/freedb to just die already so we can re-standardize on a system that doesn't use a collision-prone hash with absolutely no way to deal with collisions (and no, marking it as a blues/ genre because some other CD was already posted in rock/ is not "dealing with it"). Maybe freedb2 can fix this.

freedb2.org compatibility

[Horar]

I am the author of freedb2.org. It currently supports a subset of the old freedb protocol, enough to rip your CD's. Just point your software at http://freedb2.org/~cddb/cddb.cgi. There are also some new features which I will be documenting shortly. For some source code and a development history, please see http://asmith.id.au/freedb.html and also http://asmith.id.au/mod_libpq.html.

Re:freedb2.org compatibility

[William Marcy Tweed]

Well, I've got the freedb from last year in MySQL format if you want it. It's available at http://www.indycomics.org/FreeDB (please be kind to my T1) Granted it is from November (20051104) from last year. I will be releasing the current update in MySQL format soon. All of the source code used to generate the MySQL version is available at the same place. Hope someone finds it useful.

Re:freedb2.org compatibility

[valshaq]

You can still download the tar.gzipped database from 2006-07-01 from the mirrors.

Re:freedb2.org compatibility

[llefler]

You can also pull the July 1 2006 version directly from freedb. Along with the source for the current release of the server, a link to sourceforge for the unstable code, and some sample clients in various languages. I'm pulling a copy from one of the mirrors right now, and I'm sure others are doing the same.

Re:no honor amongst theives

[rehabdoll]

Obviously you have no idea what freedb does nor it's purpose.
Without the CD, the service is/was completely useless.

Re:no honor amongst theives

[MichaelSmith]

Just stop being a bunch of cheap a-holes and BUY music that you like.

I rip my own CDs. Mostly because I like the convenience of listening to them on my laptop. Even here in Australia that is now legal, though it has always been tolerated.

Freedb just gives me track, artist and album names.

Re:Damn.

[RackinFrackin]

I suspect that you are right, but I wouldn't use the word "knockoff". Freedb was more like a replacement of cddb, for when cddb was sold out from under the community that built it.

Re:no honor amongst theives

[swimin]

freedb had absolutly nothing to do with filesharing. If you had a cd, and wanted to rip it to mp3, ogg, flac, etc, you would want that file to be properly tagged. Everyone wants information like title, artist, name of album, order of songs on album, and year released, in every song they have stored digitally. Freedb only stored this information, to be used as you were ripping the cd, to automaticlly fill in all applicable information.

Re:no honor amongst theives

[fuzzybunny]

Yes, you are going to get flamed for this, because it's an idiotic and irrelevant statement.

Freedb.org was invaluable to me when I was ripping the 700-odd CDs THAT I OWN.

Muppet. Accurate track listing database != music piracy. Get over it.

A quick fix

[houghi]

Not sure if the following will work, but if the directory structure is the same and you only need to change the URL, you could just edit your hosts file (or do it on router level or where ever). A lot easier then to update all the different programs for the different users:
Add to your hostfile:

203.58.241.10 freedb.org

Ugly, but it might work.

Re:Freedb sucks anyway

[Schraegstrichpunkt]

Each of those hashes are (supposed to be) a completely different disc

Not quite. Apparently the hashes are an ad-hoc mechanism created specifically for cddb, and there *are* collisions.

Re:So....

[bcat24]

The database is still there and lookups still work. For now at least.

Time to replicate the database!

[hernick]

By the Power of BitTorrent, the freedb.org database is made available to all.

Today, you can get the .torrent file on http://tracker.freedb.org/ - but if it ever becomes unavailable there, you can use a DHT-aware Bittorrent client such as Azureus and get it by using this info hash: 21AF020252FD2E556B683CEB123689733E0BC063

I, for one, have allocated a total of 16mbps of bandwith on four hosts to help seed this database. I'm seeing a total swarm performance of around 25mbps, so this should be a fast download for anybody who wants it.

Go ahead: feel the Power of BitTorrent and share this free database!

Share, my friends, share!

Re:I'd just like to say,

[Anonymous Coward]

Am I the only person who took the effort to do a bit of investigation? For instance, 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.

Grandparent's urls are Australian, he calls the project freedb2, and there's very little source code to be found: I'm guessing that he's this Australian.

Now, don't get me wrong -- I have the utmost respect for people who donate their free time to making software for gratis, but when that developer pledges (of sorts) to make a replacement to an OSS product, gets support from the developers of the product being replaced (was that support monetary?), and refuses to free that code, which in turn contributes to toppling another (well known and widely depended on -- yes, I know freedb still works, but still) project, I am slightly angered, to say the least.

And what stops him from now never opening that code? Replacing a FOSS product with simply a gratis product is a net loss, from where I'm standing.

Re:So what does this mean right now?

[EvilIdler]

I'm getting CDDB info from freedb.freedb.org right now. They're CDs in my
"to rip" pile, so I shouldn't have that part of the DB stored locally yet.

Re:I'd just like to say,

[binkzz]

He is the Aussie; from one of the original developers:

"freedb2 is the development project that played a big role in the demise of freedb. That the developer is advertising it here now, apparently trying to profit from what he caused is immoral in my opinion.
Additionally, using the name freedb2.org is stealing freedb's name. Furthermore horar has not yet released source code or a database dump, so as of this moment, freedb2 is a closed source project, which violates the GPL under which the database archives are released. Even if the GPL may not be enforceable in this case, not releasing a database dump is certainly morally wrong."

Re:Gullible?

[Horar]

I am both "the Australian guy" and the author and host of http://freedb2.org./ I sincerely hope that in future you will spend a bit more time joining the dots before jumping to such sensational conclusions.

In the meantime, I invite you to enjoy http://freedb2.org/ and browse as much of the source code as I have had time to document and post on http://asmith.id.au/freedb.html and http://asmith.id.au/mod_libpq.html

Re:no honor amongst theives

[Anonymous Coward]

what, foobar2000 can magically recreate the cd key from a handful of truncated mp3s you downloaded off of kazaa? I know foobar2000 is good, but not that good.

Re:no honor amongst theives

[Temsi]

Oh well - I'm probably going to get flamed for this....

but there's no honor among theives.

Just stop being a bunch of cheap a-holes and BUY music that you like.

Wow...
It's posts like these that make me wish Slashdot had a moderation option for "-1, Stupid".

Freedb, like its proprietary and commercial counterpart, cddb, is a perfectly valid and legal service which recognizes the CD in your drive and downloads information about the artist, the album, the songs, cover art and sometimes even lyrics for display within your CD player software.

It has absolutely nothing whatsoever to do with theft or being cheap.

Re:I'd just like to say,

[Breakfast Pants]

Not provided? The thing is online; you can query by query get the whole thing. You can't download it as a 'dump', but he doesn't have to provide it as one.

Re:I'd just like to say,

[gnosygnus]

someone mod parent up. horar/freedb2's motives are not in full disclosure. hopefully not bad etiquette to post links to digg, but the comments are worth reading.

source of parent post: http://digg.com/software/freedb_s_future_uncertain
more comments on freedb: http://www.digg.com/linux_unix/freedb_is_closing_d own

at any rate, there seems to be more to horar's involvement than originally stated.

Re:Good riddance

[Anonymous Coward]

Can't. CD-TEXT does not support double byte character. IOW, no UTF-*.

freedb has sucked for ages, though...

[yoasif]

freedb has sucked almost since it's inception. Multiple entries for the same album, hard to do Various Artist albums, lots of misspellings and mistakes, and no way to ""fix" the problems.

I really hope people take this opportunity to check out Musicbrainz, a MUCH nicer alternative. It's (mostly) open source, runs on Linux, Mac and Windows.

Also, it's community moderated like Wikipedia, and it has loads of information about releases, something which was nonexistent on freedb.

Re:Ahhh, those were the times

[andersa]

Are you sure you were not really using musicbrainz? I don't think freedb is able to do this.

Re:I'd just like to say,

[Anonumous Coward]

Please do your homework. The freedb database dump is released under the GPL with the following addendum:

For purposes of interpreting the GPL in connection with this work: The
database is distributed in the form of plain text files. These
will generally be processed into to another form. The text form should
be considered "source code" and the other form should be considered a
"compiled program".

This means, the moment you publish the database in any other format than a dump (e.g. through another front end), you must publish a dump of your own. If freedb2.org is using any part of freedb.org's database, it is currently infringing freedb.org's copyrights.

Re:Maybe move it to Wikipedia

[joe_plastic]

"I agree that Wikipedia has trouble dealing with locations"
Try wikimapia -- it's like a wiki mixed with google maps.

Re:I'd just like to say,

[Raphael]

You can't download it as a 'dump', but he doesn't have to provide it as one.

Wrong. He does have to provide a dump, according to the GPL. The GPL requires you to provide the sources in the "prefered form" for making modifications to it. In this case, requiring to fetch the whole database query by query and having to convert the result back to text files would certainly not qualify as the prefered form.

For more information, see this section of the GPL FAQ: Can I use the GPL for something other than software?.

And here is an excerpt from paragraph 3 of the GPL:

The source code for a work means the preferred form of the work for making modifications to it. For an executable work, complete source code means all the source code for all modules it contains, plus any associated interface definition files, plus the scripts used to control compilation and installation of the executable. However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable.

If distribution of executable or object code is made by offering access to copy from a designated place, then offering equivalent access to copy the source code from the same place counts as distribution of the source code, even though third parties are not compelled to copy the source along with the object code.

Note that the GPL requires distribution from the "same place", so pointing to the original freedb mirrors would not be sufficient (and would not ensure that the data remains the same anyway). This is clarified in this section of the GPL FAQ: Can I put the binaries on my Internet server and put the source on a different Internet site?

Re:Get what you paid for?

[megari]

It seems like there were three people on the project, and two of them wanted to take it non-free, one didn't

This is not true. He may have misunderstood us despite our repeated assertions that we have no intention of endangering freedb's freeness. We simply wanted to get things worked out so that all the requirements for freeness and other issues would be fulfilled so that everyone would be happy. He did something unexpected and unilateral while there was an effort to fix things which made us feel he didn't feel like discussing his actions with everyone anymore. This combined with all of the difficulties and the situation being effectively deadlocked eventually led into the decision of both of us leaving. Now having slept on it, I am not sure about whether it was right for me to feel that I couldn't continue with the person left. It is strange how one finds oneself blaming oneself over hastiness even though the decision took three days to make when faced with the need to make one. I'll see how things turn out. Things may or may not get well again. In any case, see this. It may help clear things out a little bit as it contains input from everyone involved. Also, the full front page of freedb.org contains some of our reasoning. The person left removed our response to his allegation that we wanted to make the project less free giving a somewhat distorted image of us to the general public, which has prompted me to make this response to make sure that misinformation doesn't turn into something everyone regards as the undisputed truth.

Re:Gullible?

[megari]

1) Ari and Joerg support some australian guy developing the "next-gen" freedb for two years

Accurate, though I must point out that the support was not financial (as some have concluded) rather than providing information and database updates directly to him, keeping him company on IRC, things like that. He also helped us with the power of his improved backend. One could say it was mutual technical support.

2) Australian guy doesn't want to release it as open/free for freedb (or all three?)

Only time will tell whether he will release it under a free license. That was what we hoped for, anyway.

3) Ari and Joerg have either been suckers or part of an attempt at pulling another Gracenote

We definitely did not try pulling another Gracenote, that's for sure. Not keeping the project completely free (as in beer and as in speech) and replicable (that is, anyone can get the database and all the software needed to run a server) would have been pointless and not living up to the name and purpose of the service.

4) Kaiser won't play ball, it's freedb or no db at all. He finally tires and goes to the source.

Yes, he eventually forcibly asked the developer to release the source to him immediately. For some reason or another he wouldn't do that yet. The argument caused a deadlock which me and Jörg tried to solve unsuccessfully.

5) The play is called, Ari and Joerg leave because the gig is up.

I don't know if that would characterize it correctly. I'd rather say that because of the way things developed - and, in retrospect, the way people seemed to misunderstand each other - mutual trust between everyone eroded.

Only time will tell whether we were gullible to work with the developer of the next-generation server software or not.

Re:Gullible?

[Bob9113]

If he releases the code in the next several weeks, and as he says, he's adding documentation, cleaning it up. etc., where is the problem?

The problem is, right now, he's asking us to contribute data to him while trusting to his good will. That is exactly what CDDB did. Do you remember CDDB? They were the FreeDB before FreeDB. They took our data, then told us to fuck off while they sold it.

If he just dumped the code, there are likely to be just as many complaints.

True, but making the code better will not change that. No matter how good the code is, some jackass is going to find fault with it. The trick to Free Software is not making perfect software, but realizing that there is no such thing.

If Andrew (or whatever his name is) hasn't attempted to distribute binaries that contain GPL code (and I'm not sure we know that he has for a fact), then we need to back the fuck up.

Sure, that's fine, he can do whatever the hell he pleases. But we should no more go use FreeDB2 than go back to CDDB. As it stands today, FreeDB2 is proprietary. We all know what happens with proprietary versions of CDDB, because it happened. CDDB said they would be freee. They asked us to trust them. Then they took our data and told us to fuck off.

Moreover, he broke his obligation to FreeDB. FreeDB has our support because it is Free. At least one person, the person who started FreeDB, the person who grasped why we chose FreeDB over CDDB, expected him to maintain the agreement that is the core value proposition of FreeDB. If Andrew broke that agreement (and maybe he hasn't yet), then he has stolen from a Free project. Should FreeDB have insisted on the copyright like Apache does? Perhaps, then this wouldn't be a problem. Does that mean it's FreeDB's fault? Perhaps, at least in part. Does that mean we should stand up for Andrew (or whatever his name is)? Absolutely not. At this point, he is a maybe crook.