Skip to main content

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: freedb-project is dead (Read 81268 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

freedb-project is dead

Reply #25
Well, if foosion's "prediction" on his own plans (lol) is to come true, we'll actually have another option
Ofcourse for that to happen, foosic will need more universal software-support.


foosic has currently 97.000 albums in the database, musicbrainz something over 400.000, both way below freedb's size.
I think the the freedb successor will need to import freedb's database into it's own db

freedb-project is dead

Reply #26
foosic has currently 97.000 albums in the database, musicbrainz something over 400.000, both way below freedb's size.
I think the the freedb successor will need to import freedb's database into it's own db


Actually, musicbrainz has an interface to freedb, but you can only use it to manually import albums. It's a funny solution; they want to ensure that imported entries are checked to adhere to their extensive guidelines - or at least that's what I think is the rationale behind it. I imported two or three albums that way, and even though it was a smooth experience, it's probably too much effort to make it effective.

The alternative is indeed to just import all content that is in freedb. It means you're bringing in a lot of rubbish, too. I guess the musicbrainz people have though about that, and probably they had good reasons not to go that way. Once you have a lot of double entries in your db, it's harder to get them out than not allowing that in the first place, for example.

The strict tagging and naming guidelines that musicbrainz imposes are a natural reaction to the mess that freedb is (or was?), but it's not an ideal solution. I don't like to have to tag my albums according to their guidelines if I want to be allowed to add new data. It's all too rigid. And then consider that my music is mostly easy to tag... try tagging some classical music

OK, that was sort of a braindump, probably you're thinking: "how is this a reaction to my post?"
I think in the end, I agree with your idea that a successor will have to build upon freedb's content. But it will take a really smart database engineer to make it an improvement over freedb...

Maybe this could work for foosic, if they specify that every implementation that uses their db also does the fingerprinting stuff. Then they could just import all the freedb stuff, and some of it will just never get linked to real songs, and can eventually be washed out of the db... Probably this is oversimplified and naive thinking.

freedb-project is dead

Reply #27
I've been playing around with the MusicBrainz Picard tagger. While I admire the consistency and organization of their DB, the tagging program itself is rather clunky and not very intuitive. The MusicBrainz DB is quite complete, there's some stuff I couldn't find, but everything else I searched for was spot on. And remember that freedb has a lot of info because theres a lot of repeated entries per album, this is something MB are avoiding with their own DB I think.
we was young an' full of beans

freedb-project is dead

Reply #28

Well, if foosion's "prediction" on his own plans (lol) is to come true, we'll actually have another option 
Ofcourse for that to happen, foosic will need more universal software-support.


foosic has currently 97.000 albums in the database, musicbrainz something over 400.000, both way below freedb's size.
I think the the freedb successor will need to import freedb's database into it's own db


I think if more people use the foosic component knowing that the freedb project is dead, the foosic library will expand quite rapidly!

Although I do agree to an extent

freedb-project is dead

Reply #29
Bummer 

I'm downloading (maybe the last chance) the latest 7/1/06 db.

It may be full of errors but I can clean them up in EAC before I rip.

Sure beats manually keying in all that info for each rip.

freedb-project is dead

Reply #30
The CDDB protocol's 32 bit CD-TOC checksum was crap anyway.
I'm glad to see MusicBrainz support in CDEx

freedb-project is dead

Reply #31
REAL SHAME - that's a big one leaving the scene...

Thank you guy's at freedb for all your work!!! It's alway been so seamless,
as soon as I put in my CD, EAC had all the album information ready to use
(with some corrections to do sometimes  )...

Guess there's not much one can do. There do seem to be times when you
have to let go?!

As for the audio community, let's hope for an alternative to become available
soon. MB looks quite good to me!

Best regards!

freedb-project is dead

Reply #32
That's really sad news.  Freedb is brilliant.  But I'm sure something will emerge in the not too distant future... it always does right?

---

I've always loved the Discogs system/db.  If there was a way for EAC to use that or indeed Music Brainz, then I'd be a happy man.  MP3Tag can use both.

Foosic looks cool too.  Not sure if that is allowed to be used in EAC though??

I keep mentioning EAC cos as we know that's seemingly still the ripper of choice.

freedb-project is dead

Reply #33
Maybe this could work for foosic, if they specify that every implementation that uses their db also does the fingerprinting stuff. Then they could just import all the freedb stuff, and some of it will just never get linked to real songs, and can eventually be washed out of the db... Probably this is oversimplified and naive thinking.
The foosic database grows by collecting playback events, i.e. the client software submits information about the currently playing song - with or without the fingerprint - and this gets added to the database. The lookup protocol to be used with tagging software always uses fingerprints, so just adding data from freedb without fingerprints does not really help to increase the usefulness of foosic as a tagging service.


The CDDB protocol's 32 bit CD-TOC checksum was crap anyway.
I'm glad to see MusicBrainz support in CDEx
I fully agree about the disc ID. The "ISO 8859-1" (or rather whatever client software used instead for submitting) legacy entries were pretty annoying as well. At least that's something the freedb2.org developer was working on. According to what he announced on the freedb forums, he had converted all but 30000 entries from local encodings to UTF-8.


Foosic looks cool too.  Not sure if that is allowed to be used in EAC though??
Garf was annoyed by the fingerprint used in MusicBrainz being proprietary and deliberately made his fingerprint open-source:
Quote
libFooID library, including full source code, a static library and a DLL version. Freely available under either GPL or a modified BSD license.

freedb-project is dead

Reply #34
Maybe this could work for foosic, if they specify that every implementation that uses their db also does the fingerprinting stuff. Then they could just import all the freedb stuff, and some of it will just never get linked to real songs, and can eventually be washed out of the db... Probably this is oversimplified and naive thinking.
The foosic database grows by collecting playback events, i.e. the client software submits information about the currently playing song - with or without the fingerprint - and this gets added to the database. The lookup protocol to be used with tagging software always uses fingerprints, so just adding data from freedb without fingerprints does not really help to increase the usefulness of foosic as a tagging service.
How long does a song need to play for enough info to be gathered and sent to the foosic db?

Would creating a fb2k plugin that 'plays' (for the required minimum time) or indeed simply scans someone's music library to then upload to the foosic db be possible? 
Like many people I have a large and well tagged audio collection, and I'd be more than happy to get it into foosic.


Foosic looks cool too.  Not sure if that is allowed to be used in EAC though??
Garf was annoyed by the fingerprint used in MusicBrainz being proprietary and deliberately made his fingerprint open-source:
Quote
libFooID library, including full source code, a static library and a DLL version. Freely available under either GPL or a modified BSD license.

Thanks for the info, sounds good.

freedb-project is dead

Reply #35
The alternative is indeed to just import all content that is in freedb. It means you're bringing in a lot of rubbish, too. I guess the musicbrainz people have though about that, and probably they had good reasons not to go that way. Once you have a lot of double entries in your db, it's harder to get them out than not allowing that in the first place, for example.

Musicbrainz used to have a script running to import FreeDB entries automatically, but it was dropped for the reason you give here. 

Quote
The strict tagging and naming guidelines that musicbrainz imposes are a natural reaction to the mess that freedb is (or was?), but it's not an ideal solution. I don't like to have to tag my albums according to their guidelines if I want to be allowed to add new data. It's all too rigid. And then consider that my music is mostly easy to tag... try tagging some classical music

Upcoming versions of the Picard tagger will allow you to be more flexible in the way your music is tagged with the information from the database. It's still a work in progress though.

freedb-project is dead

Reply #36
How long does a song need to play for enough info to be gathered and sent to the foosic db?

Would creating a fb2k plugin that 'plays' (for the required minimum time) or indeed simply scans someone's music library to then upload to the foosic db be possible? 
Like many people I have a large and well tagged audio collection, and I'd be more than happy to get it into foosic.
The threshold for a track to be submitted to foosic.org by foobar2000 (rather the foo_sic component) is 60 seconds, unless the track is shorter. The server has a spam protection, so it will drop submissions that come in too frequent, i.e. less than 25 seconds after the last if I remember correctly. This is because the submission protocol is intended to track playback events, not to submit track metadata en masse. Please see the foosic.org discussion thread in the foobar2000 forum for details.

freedb-project is dead

Reply #37
I too have many thousands of high quality, well tagged audio files, and I could probably contribute substantially to a project like this, if there's a fairly easy, more-or-less automated way to do it.

Obviously something needs to take freedb's place.  In a way, it could be a good thing though, because there were a lot of technical limitations to freedb.  I like how new systems maybe can support things like cover-art, audio fingerprinting, and libraries of all artists, etc... and would potentially have ways of ensuring better accuracy and correctness of spelling and consistency of formatting, which were big issues with freedb.

Freedb also had way many false positives (and sometimes false negatives).

I guess it's just time to move on.  It was nice while it lasted!

freedb-project is dead

Reply #38
Now clearly is a time for a new database to step up. Hopefully this will be a chance for MB to get better support. It would be nice to see MB add the foosic open acoustic fingerprint, for future proofing their database.

freedb-project is dead

Reply #39
Why doesn't this community start up some sort of freedb type project.

freedb-project is dead

Reply #40
Why doesn't this community start up some sort of freedb type project.

Time, money, complex technical issues...

It'd take some pretty fancy programming and such...  There are probably already alternatives in the works that could take freedb's place.  And there's always gracenote, if needs be, I suppose.

freedb-project is dead

Reply #41
Quote
update:
currently, there are talks with organisations regarding freedb's future.
I am trying to find an organisation that takes the domain and ensures
the following:

- they have to provide the freedb service, not only take the domain
- the service must remain free of charge for users and developers, regardless of the application
- database updates will be available on a regular schedule and will be free of charge
- the licence of the data and the software will remain under the GPL

freedb-project is dead

Reply #42
Dam this is sad news, I always prefered freedb over CDDB and AMG.
"I never thought I'd see this much candy in one mission!"

freedb-project is dead

Reply #43
> am trying to find an organisation that takes the domain and ensures

freedb weekly top 20 clients
1. Nero 1792188
2. CDex 676765
3. Tag&Rename 524235
4. FreeRIP MP3 462734
5. Easy CD-DA Extractor 442345
6. Exact Audio Copy 328667
7. Audiograbber 324872
8. libkcddb 301165
9. dBpowerAMP Music Converter 244666
10. freedbdemo 231942
...

Anyone else think Nero should take it over?

freedb-project is dead

Reply #44
Anyone else think Nero should take it over?

Sure if they promise to remove DAE and audio burning from their software.



(...may I have two seconds of silence by default between my lack of error detection please?)

freedb-project is dead

Reply #45
Man, I never would have thought that so many people use Nero for tagging (and ripping?) their audiofiles. Must be all those OEM versions and some piracy.

 

freedb-project is dead

Reply #46
Man, I never would have thought that so many people use Nero for tagging (and ripping?) their audiofiles. Must be all those OEM versions and some piracy.

Let's hope the pirates aren't using Nero.

freedb-project is dead

Reply #47
Anyone else think Nero should take it over?


Think about the bloatwared computers 
"I never thought I'd see this much candy in one mission!"

freedb-project is dead

Reply #48
I don't think Nero would be the best choice, but they may be interested in freedb. Probably they will be not to eager to implement several solutions into their program without knowing wich one will become dominant.
The numbers inducate that the Nero customers are using this feature, so Ahead needs to provide it to them.
The server infrastructure is already available, so taking over freedb seems to be the easiest solution.

But there are some people involved in Nero around here on HA, maybe they could add something to the topic...

freedb-project is dead

Reply #49
Any comments by André regarding possible alternatives for EAC he might be already considering?
WavPack 5.6.0 -b384hx6cmv / qaac64 2.80 -V 100