an alternative? musicbrainz.org

i realise a lot of this forum seems to be devoted to creating a new discography, which is a valiant idea, but you know...

what about http://www.musicbrainz.org ?

forget any preconceived notions you might have about this site for a moment and check this out
- it's non-profit
- dedicated ACTIVE group of opt-in 'moderators' who oversee all guideline changes. no 'community manager' here.
- anyone can contribute to the code
- tagger that takes the DB info and uses it to tag MP3s (etc). this is highly scriptable and sophisticated.
- it has a voting system. changes appear straight away, but the voting system is more logical than discogs. this is no wiki.
- advanced crediting system that is actually indexed
- advanced label inter-relationships

and in the pipeline they have:
- work lists - every seperate audio recording of a song has separate entity on the DB, so you don't have to duplicate credits over every release (matter of weeks away)
- inheritance - following on from the above, soon there will be ways of showing credits across re-releases/remasters of the same release (for example).
- 'wantlist' and 'collection' features
- NextGenerationSchema which will basically allow for all unique releases, like discogs

i've been involved in both projects equally. whilst discogs does (or at least did) fulfull all my collection needs, i've been using MBz for my tags, and where discogs has faultered, MBz has grown and grown, and ever closing the gap, and i actually get to help the project rather than just moaning with an influx of expertise like that available here i am confident it would reach it's potential.

here's an example of a musicbrainz release

Sorry, never liked musicbrainz and probably never will... I can't exactly tell why, if I can put my finger on it I'll post it here.

don't like it either. IMO, it's a bit shitty that e.g. releases can be added without even having a label, or 2 CD boxes as 2 separate CD's etc.

Further the lay-out is even crappier than on Oggs

It started as a CD database, as an alternative to CDDB, focusing on main artists and labels only, no interest in the minor players. If we could talk them into allowing us to augment their data with Discogs-like detailed artist information, maybe it'd be interesting. But until then, their focus is rather different than ours (I think).

azzurro wrote:

Further the lay-out is even crappier than on Oggs

Yeah, it looks like a mess. Definately designed by a programmer.

deejsasqui wrote:

It started as a CD database, as an alternative to CDDB, focusing on main artists and labels only, no interest in the minor players.

Found the reason why I don't like it: way too mainstream, very amateurish level.

Yeah, as above pretty much. I couldn't wrap my head around the layout for browsing, much less adding info. And it does seem less about all the artists involved (not to mention all the versions of a release) and more about the CDDB data. At this point, anyway; I see you noted some future improvements.

Modding + non-profit = good, but the focus on MP3s/CD data sort of leaves a lot to be desired (IMO).

this is a very new db
http://www.music-whiz.com

I'm not sure about it but it does seem to have potential.

The framework's definitely there.

Took out a random release: http://www.music-whiz.com/141/Operator_Please/Yes_Yes_Vindictive/album

Looks pretty worthless to me...

There's structure, but the site is a fsking eyesore, everything that our fallen site is/was not.

re: MusicBrainz - I like the ability to use CDs for some data input / verification, but I agree that's no bedrock I want to base my Ideal Future DB on.

deejsasqui wrote:

that's no bedrock I want to base my Ideal Future DB on.

Exactly, there is no bedrock out there, no site that really provides what we all seem to want. Discogs was it and now it's dead so we just need to start anew.

The great thing is we have an increasing number of very qualified, very dedicated people here right now all wanting something new and all willing to put in the work. We also all have contacts (on Discogs and elsewhere) who we can direct to our site when we get things going so this can really be a valid contender in the world of online music cataloging.

Shadowslip wrote:

this is a very new db
http://www.music-whiz.com
The framework's definitely there.

Naw... more flash than content. Check this out as their "featured album" - http://www.music-whiz.com/159/Elton_John/Elton_John_s_Greatest_Hits/album

But I do like they way they display their (flawed) info on their front page.

just an idea i wonder f he would be up for some help etc?

"Music-Whiz.com is the newest (and best) music community on the internet. It's a huge music database which YOU can edit! In other words, it's like IMDB.com, but for music, and like Wikipedia in that anyone can edit it.

Music-Whiz.com is a relatively young web site that will only flourish with help from people like yourself. If you can help us in any way, it is much appreciated."

Sounds like Discogs v4, with more flashy graphics, and obnoxious ads, but no marketplace =)

Hi All,

I created Music-Whiz.com.

There are some things that I would like to share with you:

I have been building Music-Whiz.com in my spare time as a hobby for about 6 months. I'm not a professional programmer, I'm not a professional web designer, I am possibly the worst person in the world when it comes to making graphics and I have had absolutely NO money to put into it.

I am building Music-Whiz quite simply because everywhere else I could find is really shockingly difficult to use or has a crap load of pop ads etc.

I know it's not perfect, and I do still have lot's of work to do on it.

I have added a lot of information to it, but it's quite difficult to find the time to add information to it and keep building and maintaining the site as well.

The main theory behind Music-Whiz.com is that anybody should be able to add information to it and edit information thats already in it. I have made Music-Whiz.com so that it is very easy for people like yourselves to add information about your favorite artists and bands.

Like I said, I do realise that it is seriously lacking in substance, and I ask that if any of you think Music-Whiz.com has potential, please add some information about your favourite bands. I also encourage you to add information about any bands you might be in yourself.

The more people who add information to Music-Whiz.com, the more time I will be able to spend adding new features and maintaining the site.

I like the option to click on a track title and get a view of all releases it appears on though.
For the rest it's too much bright blue and V4 imo.

Hi Richard

Richard wrote:

The main theory behind Music-Whiz.com is that anybody should be able to add information to it and edit information thats already in it. I have made Music-Whiz.com so that it is very easy for people like yourselves to add information about your favorite artists and bands.

We believe (well, at least I do) that not everybody should be able to add information, at least not without some kind of verification by other people. This is the same system as Discogs v4, which is a complete disaster. I know it would be great if everbody could submit and enter information correctly, but that will never happen.

Making a database open for everyone to submit in an easy way is good for the level in quantity but can in no way be good for the level in quality. You may want to take a deeper look at how Discogs is doing at the moment to see what opening a DB leads to.

Yea it's like anyone could help write the newspaper, lots of fake stories would be published.

I swear i didn't tell him about this site

Somebody's been surfing his referral logs...

Yeah, I was looking at my referral logs and noticed a some traffic from here so thought i'd take a look.

I've actually never heard of discogs until now. And by the looks of it, you guys aren't happy with the way that site is being run?

I see where you guys are coming from with your opinions on allowing anyone to edit the database, but it seems to work ok for Wikipedia, so why couldn't it work for a music database? The idea is that users moderate each other.

It also allows a wider variety of music to potentially be included in the database. There are some great local bands around where I live, but the chances of a moderator knowing them are very slim. You need to allow people to add their own bands no matter how well known they are if you want to eventually have a really comprehensive database.

As for the complaints about 'too much blue' on my site. I agree, but i'm really crap with graphics. If anybody is better and would want to donate their expertise to make a better layout they are more than welcome to.

In fact, I welcome any suggestions or help at all from you guys, who obviously want somewhere better to find your music info. My email address is musicplace.co.nr [at] gmail.com if anybody here wants to contact me.

Richard wrote:

I see where you guys are coming from with your opinions on allowing anyone to edit the database, but it seems to work ok for Wikipedia, so why couldn't it work for a music database? The idea is that users moderate each other.

Wikipedia cannot be compared to music DBs. Adding info to a DB is much more strict, especially cross-linked DBs. We believe that everyone should be allowed to add data, but the data should be confirmed by experienced users.

The Wikipedia system is working because there's a wide field of interest and knowledge regarding the different articles across the (much bigger) user base compared to discogs. You search wikipedia because you want to know something about a specific subject. This way you can find some interesting articles via links in other articles and can correct some errors when you see them.
But Music DBs deal with a very specific area of expertise. There it's all about collecting music and the fan base is pretty small, too. If you want to search for a release you know then there's no problem - you can spot the errors and correct it. But if you try to search some rock releases for example you get thousands of search results. Sure - you can follow some label or release links there too, but do you know the release data has been entered correctly when you don't know that special piece of vinyl? You have to trust the user who entered the data without moderation. And if nobody actually owns that release (you can manage your collection in discogs) then these errors would never be found.

Discogs would be better if there were the mechanisms to find shitty release submissions more easily but the site owner leaves us alone with our wishes.

And please remind: Some articles are also moderated in WP, too.

We try to be open minded but we found no solution to the moderating problem yet.

The most distinct difference I see between wikipedia and Discogs is the level of detail. The sheer number of obscure singles from forgotten bands is pretty impressive. And when you try to add all the details from forgotten records, you end up with a lot of similarly named artists and groups. Heck, even some groups who are well-known in one country have the same name as another big group from another country, resulting in some localized name alterations.

And once you get to assistant recording engineers and some session bassist with a generic name, links can get REALLY messy. That's why we feel that a music database site with fine-level details requires more oversight. But if there were better automated checks, and a really clear set of rules and guidelines, there could be a lot less time spent by moderators and more done by submitters. Except not everyone cares about the fine-level details. So your local rock band Wicked Wizard might get connected to a 1970s glam-rock band from Poland, and no one notices until months later, and no one who submitted data for either band is active. Maybe the band was reborn in Tulsa, Oklahoma, we don't know now.

The people on this site are something of elitists, only because we value quality well above quantity. We're OK with waiting in line for our submissions to get accepted. Not everyone feels the same, and we recognize that.

skutchy wrote:

deejsasqui wrote:

that's no bedrock I want to base my Ideal Future DB on.

Exactly, there is no bedrock out there, no site that really provides what we all seem to want. Discogs was it and now it's dead so we just need to start anew.

The great thing is we have an increasing number of very qualified, very dedicated people here right now all wanting something new and all willing to put in the work. We also all have contacts (on Discogs and elsewhere) who we can direct to our site when we get things going so this can really be a valid contender in the world of online music cataloging.

i think no-one seems to appreciate the work that needs to go into something like musicbrainz, or even something (relatively less complex) like discogs. it's 1000s of man hours. we need 1 (preferably more, else it will just go the way of discogs) paid programmer to get anything even remotely close to what we're after, and even then it will be at least 1-2 years before even the basics are up.

by this point, i will eat my hat if MBz isn't covering exactly what discogs used to cover, and a whole lot more. from where i've been sitting, MBz has moved at a breakneck pace whilst discogs has made missteps.

deejsasqui wrote:

It started as a CD database, as an alternative to CDDB, focusing on main artists and labels only, no interest in the minor players.

eh? how does "it" have "not interest in the minor players"? if you want a release on there...add it! there's no restrictions, or 'focus', beyond what the users are prepared to add.

Gecks wrote:

i think no-one seems to appreciate the work that needs to go into something like musicbrainz, or even something (relatively less complex) like discogs. it's 1000s of man hours. we need 1 (preferably more, else it will just go the way of discogs) paid programmer to get anything even remotely close to what we're after, and even then it will be at least 1-2 years before even the basics are up.

by this point, i will eat my hat if MBz isn't covering exactly what discogs used to cover, and a whole lot more. from where i've been sitting, MBz has moved at a breakneck pace whilst discogs has made missteps.

deejsasqui wrote:

It started as a CD database, as an alternative to CDDB, focusing on main artists and labels only, no interest in the minor players.

eh? how does "it" have "not interest in the minor players"? if you want a release on there...add it! there's no restrictions, or 'focus', beyond what the users are prepared to add.

Sorry for my lack of clarification, but I wasn't dismissing MB for a lack of work or complexity to data collecting, but at first look MusicBrainz didn't look like something that would could work with Discogs data details. After looking over the Nick Cave & The Bad Seeds page, there is a lot of data related to the artists and releases. And there are releases with full artist information, as seen on Where the Wild Roses Grow. To my untrained eye, it seems cluttered and layered. Maybe I am judging everything from the basis of Discogs layout, and limiting my ideas about what could work.

With that, maybe we could become a filter or fork of MusicBrainz? Maybe I'm picking at the details, but it seems like MB could use some reconfiguring to display data more intuitively. I'll try to learn more about the site later.

As for the amount of work involved to start this, there is an amusing / interesting post by 0x00 from January 7th, with three links to some basic freelance scripting requests to clone Discogs, with bids ranging from $250 to $2,500, and delivery times ranging from 4 to 45 days:

http://www.scriptlance.com/projects/1171281134.shtml
http://www.scriptlance.com/projects/1160912300.shtml
https://www.scriptlance.com/cgi-bin/freelancers/project.cgi?id=1178189584&sortby=delivery

It seems the question is then, would it be better to lend our efforts at expanding MusicBrainz, or should we still start out on our own?

Gecks wrote:

skutchy wrote:

deejsasqui wrote:

that's no bedrock I want to base my Ideal Future DB on.

Exactly, there is no bedrock out there, no site that really provides what we all seem to want. Discogs was it and now it's dead so we just need to start anew.

The great thing is we have an increasing number of very qualified, very dedicated people here right now all wanting something new and all willing to put in the work. We also all have contacts (on Discogs and elsewhere) who we can direct to our site when we get things going so this can really be a valid contender in the world of online music cataloging.

i think no-one seems to appreciate the work that needs to go into something like musicbrainz, or even something (relatively less complex) like discogs. it's 1000s of man hours. we need 1 (preferably more, else it will just go the way of discogs) paid programmer to get anything even remotely close to what we're after, and even then it will be at least 1-2 years before even the basics are up.

by this point, i will eat my hat if MBz isn't covering exactly what discogs used to cover, and a whole lot more. from where i've been sitting, MBz has moved at a breakneck pace whilst discogs has made missteps.

Gecks, believe me, we do realise how much time is spent in that project. It's just not the way most of us want it.

Gecks wrote:

deejsasqui wrote:

It started as a CD database, as an alternative to CDDB, focusing on main artists and labels only, no interest in the minor players.

eh? how does "it" have "not interest in the minor players"? if you want a release on there...add it! there's no restrictions, or 'focus', beyond what the users are prepared to add.

I still have the feeling the site is too mainstream. There aren't many obscure releases, and the focus is too much on commercial releases. I can't help it but I've got a "web 2.0"-feeling out there... even if it exists longer than that.

On the Expansion of MB, I was just browsing through their forum, and came across this post about importing RollDaBeats releases, which is not automated, but being done by some RDB folks. It seems MB is taking an interesting direction.

dmaxx wrote:

I still have the feeling the site is too mainstream. There aren't many obscure releases, and the focus is too much on commercial releases.

again, that's entirely the prerogative of the users. your music won't add itself. i really don't get your criticism, sorry

BTW, with discogs export tools (esp the recent revision that removes those old restrictions), and the fact that musicbrainz is non-profit, i believe there should be no problem importing discogs' data in its entirety to musicbrainz. for me, data isn't really the problem (though certainly MBz DB needs a few developments to finish before it can handle discogs-type 'all unique releases' content)

for the record, i to don't like MBz layout at all - i think it's too clutted. ultimately i don't have the time to design a replacement, but maybe someone here does? i know there's one in the works...

Gecks wrote:

i think no-one seems to appreciate the work that needs to go into something like musicbrainz, or even something (relatively less complex) like discogs. it's 1000s of man hours. we need 1 (preferably more, else it will just go the way of discogs) paid programmer to get anything even remotely close to what we're after, and even then it will be at least 1-2 years before even the basics are up.

I'm glad you don't size projects where I work, because nothing would ever get budgeted.

Assuming a single developer who's good with the technology you picked, I think you could put together something relatively close to discogs' core functionality (no forums, no marketplace, and none of the foo-foo AJAXy cruft that apparently went in as part of V4) in four to six months. And that's in normal-people hours -- forty-hour work weeks -- not dot.com hours.