Proof of editing sources

MusicBrainz
41

Absolutely! I do have some questions I do not expect answers to, and some comments… Note that none are intended to be argumentative, I am simply stating my thoughts for discussion if anyone might find them worthy of such.

  1. We given this have far too many duplicate recordings. How often is it that a musician records different recordings for different releases? I assume not often at all, if ever.
  2. This sort of makes a remix NOT a different recording, as the “remix”, by definition, is not a different recording, but a different mix of the recording.
  3. Anything “live” is a different recording, as it was not derived from an existing studio recording.
  4. Let us consider something… what exactly is a “recording”? Say I am in studio. I record the horns, I record the percussion, I record lead vocals and then I record the backing vocals. I might call that 4 different recordings. The final product, the combination of the 4, is not really a recording, but a mix.

In general terms…

I do, well used to, a lot of live recordings taking my source from the board. I would consider each one of those its own recording. My logic is that I have taken something that was performed (live, in person, etc) and recorded it. Now if I prepare a CD from them, the process becomes a mastering of the recording, as I have altered … dithered, resampled, etc… to meet the criteria for a CD. Now when I make this a digital copy, I do no such things for the raw “release”. If it is to be an MP3 for example, I do need to do some of this. The MP3 needs 44.1kHz, which makes it a final product of the recording as I need to truncate all of the headroom causing any future editing far more destructive. Not to mention that the MP3 is compressed audio.

Not to ramble on further, this brings up what was mentioned prior by another contributor. MB does not currently give much attention or importance to the mastering process. In the current state of MB, it seems a need that a new recording be made when a new “master” is made as there is really no other way to clearly differentiate. Using extreme case, a MP3 release that uses a cutoff at 15kHz is not the same as a true say 24 bit 96 or even 192kHz copy. Personally I consider them different enough to remove the prior when I get the later… Like deleting artwork that is 500x500 and replacing it with 1200x1200. The same, but not the same.

1 Like
42

^^ My point of that rambling is that there are many aspects that we as end consumers of a product simply do not know. I think sometimes we become a bit pretentious. This includes me as well, I am not pointing fingers. I think we need to look at the facts and try our best to remove assumptions. The only way I know to do this is to use provable fact as source for edits. So if Joe at Studio A sends me an email that this is that, that email is my source and should be uploaded as such. No email = no valid edit as such.

1 Like
43

That’s not true for MB recordings! The original mix is included and a remix is a new recording. That’s the definition.

And this mix is the MB recording ← studio or live recording + mix

That’s true. There are many cases where mastering on release basis is not sufficient. The quality only occurs in comments.

Vinyls and CDs would be strictly different recordings then. This would be a major change affecting most of the data on MB. And it would multiply the number of recordings. I don’t think that would be a good idea.

3 Likes
44

It would be nice to implement an upload option for related documents that don’t fit into the images category. But you cannot force anyone to upload any documents.

There is already an edit note. You are asked to provide supporting evidence for the edit. It will be difficult to find suitable criteria for the quality of an edit note and you cannot reject all edits without attached documents.

1 Like
45

The problem is that this is just not possible for a large section of the data in Musicbrainz. How do you contact someone who worked on the production team of some 1970s vinyl album, or some obscure 2000s netlabel who didn’t have any internet presence besides a long-defunct website?

3 Likes
46

Sorry for the long post, addressing a lot at once. :slight_smile:

I agree. Part of my stated issue/concern addresses this in a way. If I can contact such person today, documenting it would prevent this issue from continuing into the future. So much data is lost relating to older releases, mostly because no one documented it. A great example of this is where an artist releases a “special”, like a post on Twitter where fans can download the release. That download is not there forever, and the Twitter post, maybe, maybe not. I have a few of those and I regret not saving more information on the origin. I would have appreciated myself taking a screenshot of the twitter post, the download page, etc. and saving it in the album folder as cover art.

In reply to this and most of your comment… yes, this is a problem and it seems that without major change, just not possible. There is a need, in my opinion, for a secondary level of a recording… the “master”. Let me speak in loose terms a bit to explain.

First I would say that a CD and a vinyl are in fact different recordings (using some of the logic explained in the next paragraph). Just like a remix being a different recording, the content placed on a CD is not supposed to be the same as what is placed on vinyl. This makes it a different master (also explained more next). As we all basically agreed on though, MB does not really focus on the mastering of a recording. This same logic applied to the criteria for iTunes vs Mastered for iTunes. Whether it is liked or disliked, the effort is there. In my opinion, the mastered for iTunes is a different master and should not be the same “recording” as say an iTunes Plus release. Currently, MB does not really support such things.

Say I record in studio. This is a recording. Now I take that recording and remix it, this is sort of a different mastering of the recording, yet derived from the same recording. This is where the “proof” comes in for me. I like to keep track of things like software used, who did it and at which studio, etc. I must admit that I am not aware how one did a remix of sorts prior to the digital age, so I cannot comment on that.

I completely understand that MB by its own purpose and definition has limitations making a lot of this impossible on many levels. I see the changes in how digital releases are being added and an shocked at the level of detail, yet the “real” detail is still left out. While again this is not being debated here… take a digital release where MB populates its barcode, catalog number and whatever else… none of that really addresses what I have and a great deal of the data is assumed or derived vs shown in solid proof. While things have changed a bit, you compare a reverse lookup of a barcode on iTunes to a CD where the barcode is plain and clearly assigned to my release, there is a gap here. I prove my CD barcode with an image. How am I proving my digital release barcode? Does it even help me identify my release?

This comment is great, and directly in line with my intent. I never knew this was done and appreciate the sharing of this. While it is not perfect, it is at least something to show the proof of edit.

The reason I had stopped editing for a while is MB became sort of useless. I could identify a digital release far faster elsewhere. This was not the case for CDs, but I really do not have a need to identify them. Once I got past my self centered thoughts, I came to realize that I should still be providing data and knowledge, regardless of the rest.

Many (but not all) of the scans of cover art I upload are not done by me. I use various methods to pull these from the internet. I do this because such content will not exist forever, even though people state that once posted to the internet it never goes away… slight myth there. I only use TOC data from those I have generated myself as I cannot really prove that what you get online is true, whereas the images can be proven by comparing. While that means something to me, it does not to others. I am just a person that most all of you do not know. What does my word mean? Well, nothing. As in science, you cannot publish results without some sort of documentation to support your claims. I believe MB should be based on provable fact, and assumptions left out. OR, there is a way to mark individual aspects with data quality.

47

This is indeed a problem, although I think this is best addressed by automatic archiving of URLs in edit notes/relationships.

2 Likes
48

Yes, the wayback machine is a great product for sure. I must admit here, while I have used it I do not know the workings of it in totality. I understand that it takes a snapshot of a web page (or site) at a given point in time and archives it as static.

1 Like
49

I think @ernstlx’s explanation might have been a bit confusing. A MusicBrainz “Recording” corresponds to a mix, ie., a stage that happens after something has been recorded to tape/disc/… but before being mastered and put out to consumers. (Of course, sometimes (like bootleg concert recordings), recording, mix, and master may all be the exact same.)

  1. Not sure what you’re asking here.
  2. A remix is a different mix, so it’s a different MusicBrainz Recording.
  3. Yes.
  4. A MusicBrainz Recording is the mix, the “final product, the combination of the 4”.
2 Likes
50

Now, that I read it again… :laughing:
(I thought on bootleg live recordings (live mix) vs official mix and added “originally mixed for release”)

MB recording = live/studio recording + mix
I think, that’s clear now…

1 Like