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."
Actually, there is some code out there: http://asmith.id.au/mod_libpq.html
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.
True, but all CD info databases have that problem, some more than others. But still, freedb was cool for when you just want to *play* a CD without ripping it. (Yep, people still do that.)
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'm sorry the staff fell out, costing us access to a useful resource. freedb was a useful tool but it was always in need of improvement.
It really should have had facilities for submitting an md5 hash of the CD so end-users could avoid collisions, perhaps an easy way to edit or rate database entries, so that submissions where the track titles were wrong could be corrected by the community, etc...
Hopefully whatever replaces it will be better and more robust..
--
""Freedb, the free music database used by tons of CD ripping software, has been shut down due to a disagreement among its developers."
And in other news. Slashdot has been shut down due to a disagreement between Taco and CowboyNeal. The former likes the new layout, while the latter hates it. Apparently one of the readers has mirrored a copy of the "/. database to slashdot2, which will be undergoing a year long "burning server" effect.
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.
"Another One Bites The Dust" ?
spoonerize "magic trackpad"
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.
Obviously you have no idea what freedb does nor it's purpose.
Without the CD, the service is/was completely useless.
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.
http://michaelsmith.id.au
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.
Just stop being a bunch of cheap a-holes and BUY music that you like.
Umm if I was downloading MP3s from P2P networks, why would I need a freedb tagger? Find a source with EAC verified, high bitrate, properly tagged music and forget using this, chances are if it doesn't got the tags it'll suck anyway. CDDB, FreeDB and the like are fixes for an outdated format (CD Audio) from a time when noone needed those tags. Unless you think all the people ripping their own CDs to their iPod / PCs / HTPCs / media centers are thieves. This is too braindead to even be good flamebait.
Live today, because you never know what tomorrow brings
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.
Cole's Law: Thinly sliced cabbage
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."
Add to your hostfile:Ugly, but it might work.
Don't fight for your country, if your country does not fight for you.
At the end of proprietary software development, the project ends and the free software community has to either do without or start anew from whatever they've got (which is not the proprietary program's source code and a license to run, inspect, share, and modify at any time for any reason). At the end of a free software project, others can pick up where the former free software hackers left off and continue improving the free software. If the license for the program is a copylefted free software license, the improved software continues to be free.
Let's hope source code for freedb2.org and database dumps from freedb2.org are shared under a free software license so that if freedb2.org dies we're not left with nothing but an increasingly out-of-date freedb.org database and freedb.org software.
Thanks so much for all the work, freedb.org hackers. Your efforts are greatly appreciated.
Digital Citizen
While I believe that free, open source software is very good and should be used more widely, this is an example of where corporate solutions can prevail.
I've used FreeDB for a while now with the CD ripping program I use (Goldwave, highly recommended), and it had its pros and cons.
On the plus side, I could find listings for more foreign/anime CDs than I could using CDDB (a corporate company, used by the likes of WinAmp and WMP, I believe).
On the minus side, there were a few moderately popular to very popular CDs that had no listing. Also, more than a few CDs (including the foreign CDs mentioned) had more than one listing, each with small differences (some with large differences, such as translated song titles, or even just misspelled words), so you had to go through each one to find one that suited you. (One might argue that the choice was good, but in this case it was just annoying.)
The reason that FreeDB stopped is because those in the lead couldn't come to a decision. This would almost never happen in a corporate environment. Any dispute would go up the chain until it hit the CEO or board of directors, where a firm decision one way or another would be made. In the mean time, the product would merely remain unchanged (unless company policy specifies otherwise), so there would be no interruption in service.
Had FreeDB used a similar hierarchy (which they may have had, but it just fell apart), this might have been avoided. The programmers/engineers would dispute something, and the project lead/lead engineer would hear both sides and say "This is this, and that's that."
Certainly, this will be an inconvenience to those who use programs that use FreeDB, but have no idea that the program does.
Not quite. Apparently the hashes are an ad-hoc mechanism created specifically for cddb, and there *are* collisions.
http://outcampaign.org/
Damn... burningserver.com is taken! That'd be a pretty sweet domain name...
The database is still there and lookups still work. For now at least.
By the Power of BitTorrent, the freedb.org database is made available to all.
.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
Today, you can get the
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!
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.
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.
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.
I just wish everyone would use CD-TEXT and rid us of the necessity for cd databases. I continually wish iTunes would burn CD-TEXT as well.
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."
'For we walk by faith, not by sight.' II Corinthians 5:7
Neither do I. IANAL, but I think it's just someone trying to take a personal issue and make it a legal one.
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.
-- This sig for rent.
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.
--
WHO ATE MY BREAKFAST PANTS?
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.
n d own
source of parent post: http://digg.com/software/freedb_s_future_uncertai
more comments on freedb: http://www.digg.com/linux_unix/freedb_is_closing_
at any rate, there seems to be more to horar's involvement than originally stated.
Can't. CD-TEXT does not support double byte character. IOW, no UTF-*.
And I've heard criminals can use pencils to stab people. What does your law breaking have to do with freedb?
Maybe not
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.
You could always just MD5 the CD. (Replace MD5 with your favorite crypograhic hash function.) Or, even better, MD5 the individual tracks, so it would even recognize tracks on collections. As you are already reading them to rip them, it would not be much more work. The problem is that it would not be useful for just playing CDs because it would only have the song/album metadata after reading the entire track/CD. Also, in the case that the data is not already in the database, the user would not be asked to enter it until after the CD was ripped, which would be annoying. Of course, there is the even better solution of publishers actually using CD Text in the first place so the data is already on the CD.
Centralization breaks the internet.
And who deals with the copyright issues for the artwork?
Please do your homework. The freedb database dump is released under the GPL with the following addendum:
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.
I don't know much about the format itself, but if they let you store 8 bits/character, then you could just put UTF-8 data in, and it should work fine. The unofficial CD Text FAQ does say that at least the lead in area does support storing double-byte characters (for Kanji), so it seems like that still should work).
Ewige Blumenkraft.
Not sure how much I want to get into this discussion (I AM NOT A LAWYER) but...
Copyright is for creative works, not database collections of fact. While there have been countries making changes to allow for the copyright of databases (collection of facts), it is very much untested waters.
You'd need a lawyer to make a definitive decision (OBVIOUSLY), but it is quite likely that no licensing terms can be applied to the database files, or any derivative works thereof.
Wikipedia:
The U.S. Supreme Court ruling
I would think that the case would apply here.. since both the information in FreeDB's db, and the information in the phone book are just collections of public information.
Help me take back Slashdot. When did 'News for Nerds' become 'FUD and Conspiracy Theories for Extremist Nutjobs'?
I think the point being made was that in a world where someone can be sued and forced to remove software from distribution - either from legal rulings or the threat of litigation forcing a chioce of financial priorities - saying that the source will be released "real soon now" may mean it's never released at all. Obviously, releasing well-formated source code with documentation is better than releasing poorly formated code with no documentation, but both are better than releasing nothing at all.
I mention DVD Decrypter in the subject of the post because it's a well-known (if you're into that kind of thing) DVD ripping software that was taken down from a number of (US-hosted) sites after threat of litigation and, since it was closed-source, no further developement was made. Now, I'm not trying to compare FreeDB to DVD Decrypter, either in terms of legality or morality. Just using DVD Decrypter as an example of software which, had it been open source, could still be under developement but because the source code wasn't being released is now "lost forver." (Of course, in actually, it's not that hard to find a download of the final released version of DVD Decrypter, but it would have been nice if the code was out there for other people to continue improving.)
Again, I'm not trying to compare FreeDB to DVD Decrypter. Just providing an example for the parent's point that you never know how some court ruling or sue-happy lawfirm is going to affect what's out there. I'm not even trying to say FreeDB *should* post every line of code currently written - I can understand why the author wouldn't want that. At the same time, I can understand why the GP has an attitude of "all-or-nothing" in terms of calling something 'open source.'
-Trillian
Nice plan, but it won't work because CD-Audio is too error-prone to have a high confidence of getting the same bitstream twice off different copies of the same CD, or indeed off the same CD played in a different player.
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:
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?
-Raphaël
Who says some developers can't get together and just start "www.reallyfreedb.org?" We can download July version of the dataase, host the software, and continue on as if nothing happened. Anyone interested?
The essential part about all this is the collaboratively filtered and collected data, right? And that is out in the wild and still available as a package, correct? ...Mmmmh... Coming to think of it ... does anyone know the mean load of late freedb?
Copy, Fork, Install, Build a cool website, have yourself a fresh OSS project. No big deal.
Ideal for anyone who needs to make themselves a name as DB admin / web services expert.
Anyway, a handfull of weeks and we'll have an alternative and freedb will be history (no pun intened).
My 2 cents.
We suffer more in our imagination than in reality. - Seneca