Continuing the discussion from Digital releases and barcodes:
Perhaps store links should be displayed more prominently on MBz. How about adding a new column to the release table, and when an online store is linked to a release, displaying that store’s icon in that release row.
That is an interesting idea. Something I think should be done as the release table right portion is not so valid for digital. The label (imprint really), barcode and catalog number are all not key like they are on physical. Well the label is, but in a different way, but that is a separate discussion.
I think label and catalogue number are just as important for digital as physical.
Can you explain? My statement is based on the fact that I can look at a CD and see the imprints and catalog number… vs … if I look at a digital release, the imprint and catalog number are not there. So I would ask if they are truly important, where do I find those pieces of information on my digital releases?
EDIT: Clarification… imprints (label as MB calls it), no. The record label (not necessarily the imprint MB wants), yes.
Because, just like with a physical release, I want to know what label it was released on and when.
A lot of labels have strong ties to specific genres, scenes or places which is important information when I’m browsing music. Whether or not there’s a physical object to write on is secondary.
When you say label, do mean the imprint or the record label… as there is a difference. MB currently asks for the imprint, which can be different than the reported record label / record company, since an imprint is a type of record label. The problem is that this information that you want to see is not available on digital releases as it is on CDs. Digital releases, if at all, are showing the record label / record company. Entering this information is a violation of MB guidelines, as it is not the imprint, but the record label(s).
When I enter a CD, you ask me for the label, really meaning imprint, that is printed on the back label, the catalog number listed on the artwork and/or CD and the barcode printed on the back cover art. So for digital releases, please advise where you would like this information obtained from? What I am trying to say is that it is not there, so although you or someone else might think it is important, if it is not there, it is not serving its purpose. CDs can be differentiated by their markings, like barcode, catalog number, etc. Digital releases cannot, thus such things do not serve the same purpose. I have not at all seen all digital releases, but I have a digital library of 40k+ songs just on Google Play Music, and another terabyte or so local. So I speak from my own experiences on what is and is not there. Most stores have required data that needs to be there, and the imprint, barcode and catalog number are not in that list and most times not there.
Here is an example of a barcode conflict:
This barcode cannot be used with the iTunes and most likely other stores as the iTunes API does not recognize the barcode. So this is a separate release, which is fine as long as it warrants a separate release.
When I search for ‘Flying Nun’ in my tags in my files, or go to their Bandcamp page to browse their digital catalogue, or search it on MusicBrainz, it is serving its purpose even better than it would be if it was printed on the case.
What I mean is that the barcode for CDs help you match the actual release to its entry in MB. You can look at the CD, see the markings (barcode, catalog, etc) and match it up to MB data (or Discogs or other) and match the data to your release. At least that is my understanding of the logic of the release data.
For a digital release, this does not work like that. If I have a digital release, where do I get the barcode, or the imprint, or the catalog number? For CDs, I get it from the release itself, which is very logical and makes complete sense. You identify a release by looking at the release.
If I look at my digital release, how do I know what barcode is assigned to it? If I look at MB and see multiple barcodes, how do I know which barcode I have for my digital release? The barcode really serves no other purpose other than to identify its matching physical product, or digital product or whatever it may be.
EDIT: This post however was intended to question the idea of listing the store on the list of releases for better identification, not really to debate if a barcode or label, etc is useful or not. The idea behind this is that the information that is there is not on digital release files in most cases, so something is needed to help identify which you have.
I’m just disagreeing with your statement that label and cat. no aren’t important for digital releases. They are important for me.
If you have a lot of releases that you have incomplete information for (eg digital releases that you don’t know where they’re from) that’s not a good reason not to have that specific information in MB.
All good, I am just trying to understand why, and how to get you information that is not there to get.
I know where they are from, the thing is, the barcode, label and catalog number are not part of metadata. I mean nothing negative by this, but I want to ask… do you have digital releases? When you get releases from iTunes and Amazon for example, that information is not there, not provided to you, etc. It is not a matter of not having it, it was not and is not there to begin with.
My digital files, whenever possible, contain the original supplied metadata, as it was. This is why I do not use Picard to tag, well one of the reasons. I want the metadata to be original and correct to original. That is the only identification you have on digital releases, so it is important to keep all of it, as it was provided, in order to retain as much information as you can. But as I mentioned, the imprint, the barcode, the catalog number, etc are NOT in there. So my point is… if it is not there, where would you like the data to come from?
Barcodes and catalog numbers I cannot see how they are important to a digital release. The importance to CDs is to identify the specific CD you have. So you can scan a barcode, for example, and a database can tell you what you have in your hands. This is just not the case for digital releases. MB asks for the imprint (but calls it label). The imprint can be seen on the CD itself and back cover most commonly. On digital releases, the imprint is not anywhere at all. The Record Label is, which is not always the same as the imprint. I mentioned this in another post and waiting to see how that data can be applied. But if you want the barcode, catalog and imprint for digital releases, you are asking for data that is just not there… not on the files, metadata, retail website, etc… in most cases. Yes, there are exceptions to this, but I speak on the majority of all collective digital releases as we need to look broad and not narrow.
If we look at digital music, we need to look at the big players and the product they sell. I exclude streaming because there is nothing to have or own, that is a different sort of release.
For providers, we have Amazon, iTunes, Deezer, Tidal, HDTracks and I am sure more that I am missing. But with Amazon and iTunes, we have the majority of the share anyway. The product they provide is MP4 and MP3, store dependent. So this means we deal with the MP4 meta structure (or lack thereof) and the MP3 structure. There is also FLAC, which cannot be ignored, which uses a fairly loose system. Vorbis comments system being a primary meta structure used.
So this now gives us the spec possibilities. Yes, there is no limit to spec only, but this is a base to use. Next, we need to look at files provided by such services and see what is actually used and how they use it. This will provide a very clear picture of what is and what is not there. Simply speaking, if the data is not in the metadata, said data cannot identify the release, technically speaking. The concept of identifying is to look at something, collect the facts and provide a determination of what it is. For CDs, MB has this covered well, the data shown on the case, covers and medium are accounted for ans work well. Same goes for vinyl and so on. For digital, do the same. Lets look at digital release examples, identify what data is there and find a way to use that data to identify a digital release.
This is important in this effort because you cannot identify a release using data that is not on the release, that just makes sense I think.
I can help most on the atom structure of MP4 (M4A) files. As you may or may not know, the MP4 meta structure is poor, and most look to Apple/iTunes as the “go-to” for definition. When I say most, I mean like the devs for things like mutagen for example, not people like you and I. Thus, I do the same, whether I like it or not. As presented on this topic, one proposal is to use the store as a factor. This would work well, at least in my experience. But again, I have not used each and every store, but I have used the major ones. This works because the stores use IDs that are kept in the metadata that can identify a release very well, but only is you have the stores database to use, or copy the needed references for archival purpose. After the store, you have the other metadata. Now for iTunes, my opinion this is the atoms for copyright and xid. This tells us (in addition to having already the artist and release name) the record label (NOT imprint, but the label), the ISRC and the copyright holder(s), most commonly the p holder. Like it or not, that is what is there to work with.
I’ve already said why it’s important to me. And it’s not just to identify the release ‘in hand’.
I have, and my label puts out, a lot of digital releases.
And this data is placed in the meta data of the files it/they/you release? I have no idea who you are or your label, but I do not doubt that there are some who do embed such data. I also stated that I am thinking of the majority. One cannot build a system on my needs, your needs or the needs of any one thing, company or person.
I hope you are seeing what I mean. At least my intent here is to look at digital releases as a whole and determine what data is available in the majority that help identify the release. There is nothing saying that data cannot be added in, or that it should not be, but that when you look at a list of releases, the right side is usually all blank or has assumed data in it. If your label sells on iTunes, Amazon, or other, what does the metadata look like? If you place such data in there, does it stay? This is what will matter vs data that is hidden from the end customer/user.
I am also not sure what you have or know, so I do not at all mean to imply that you do or do not know something. I can only speak to what I know, which says that the good majority of digital releases do not have such data. So as a database, and the need to look at the whole, we need to look at what is common on the majority.
If someone should disagree with my logic on looking at the whole and looking for the most common ground for identification, let me know. Otherwise, that just seems logical to look broad vs narrow. There is really nothing more to say I guess, as the facts are quite solid on what data is and is not in metadata, big picture, and barcode, imprint and catalog are not in that list. Please note this did not or ever say that data should not be there, but that is is not so useful, if at all, in identifying digital vs physical, considering the product in the hands of the customer.
Whether or not it’s in the metadata of a file from a store, or helps you identify an ‘in hand’ digital file, has nothing to do with why I find it useful information for display.
I simply find it extremely informative when browsing releases in my collection or on the MB site.
It’s a different use case to what you’re trying to establish, but I do believe a lot of people browse databases by label.
Well, again, there is a difference between the “label” used in the meta data and the imprint (called label) MB wants. MB wants IMPRINTS, not labels. The metadata does not specify imprints. I hope that makes sense. I fully agree, the record label is important, very important. But the issue is that when I enter a digital release, I cannot say for sure the imprint, just the label, so I cannot enter as I am unsure. There is really no way to PROVE that a digital release is under a certain imprint, in most cases, easily.
I understand that data is useful, for sure. But I am looking at it from the side of the editing user. When I enter a release, how and where do I get the data asked for? Sure, the additional data can be entered, like that discussed here, but the main elements are not main for digital as they are for physical. Maybe that explains better?
I did a bit more looking, because I am not disagreeing with you on a fundamental level at all. I mentioned above what iTunes uses, and although I personally do not like Apple, I do not rule over such things. that is all confirmed to be true. I found that Amazon also SHOULD be including the ISRC and UPC as it crosses to their also included ASIN. But, this UPS as used there might be a UPC for the physical or other as they do that at times. Unfortunately, there are some issues here.
iTunes is reliable in the sense that they are consistent. One issue is that the collection ID (the release) can change and they do not do ID linking like MB like is done in a merge… one will redirect to the other. This is an issue for past releases that have changed somehow. That sucks. It is also bad (to me) that they are proprietary. That means they do as they please and no one else can do anything but go along with it. But, Apple (iTunes) is looked at as the “MP4 spec” for metadata. MP4 being the same as M4A in this sense.
Amazon is a mess, even worse than iTunes. Personally I have found no consistency in their files. I start with that as I also read that, but need the third party a bit as I do not tend to keep Amazon files. The metadata is very random it seems.
FLAC is easier, although it is also unstructured mostly. But it is fairly common what you find in there which is the Vorbis comments tags.
As somewhat of a digital guru, that is the fact as I know it I would like to help the best I can, but my thinking is to mimmick that of CD data entry, which is where you look at your release for the data you enter. That is a tough task, no question. My input comes from a very large collection of original releases, physical and digital, and not the corrupt metadata pirate stuff. That metadata is unreliable and problem causing. I have also spent a good deal of time with taggers for audio files and after finding the right one for me, picking apart the support and functionality for M4A packaged file metadata to ensure its correctness. It is worth noting that MP3 files have a bit more flexibility, since MP4 files are more strict on structure… said without getting too technical. But since the MP3 is obsoleted, meaning that Fraunhofer did not renew its claim to it, designing around the MP3 is not logical anymore. It is replaced, but not fully yet in practice, with the MP4 (M4A/AAC) and OPUS formats. OPUS using OGG/Vorbis.
I can help with the facts and what is and what is not there. That makes most sense to me, since it is how CDs are done. But asking editors to enter releases using primary data that is not present on their release seems counter productive. I hope you can see my side and point like I see yours. I just don’t know how to get the data you look for easily for releases, unless the record label can substitute the imprint… then that one I can do.
EDIT: I wanted to clarify a portion… I do at times modify metadata. For example, I just now did, I have an iTunes release that does not list the disc number in metadata. So, I have added “1/1”. But the point of my statement remains, that I do not alter that they place there. Even on things like “bonus track” in titles. If that is how it is titled, that is how I leave it. I am unsure if that is overall best or not, it is just a personal choice and preference. Just like a CD cover, if they print something on it, it is what it is. I treat metadata in the same.
I know a lot was said and it was likely missed, but could you share where these releases are sold/released so I can see the metadata and how it is presented to the buyer / user / possessor of the music? I am trying to see how it is that you provide the metadata and how the vendor(s) keep it in place so it is on the release.
Since you referenced it as your music and your label, I understand if you would opt to not say. I am just trying to see your example and how this information is provided with the release for its users/customers to see.
You can see a number of labels using Bandcamp here:
https://bandcamp.com/labels (links at the bottom)
Hmm, I will have to find a release from Bandcamp that I have to look at the metadata. I know I have them, just cannot recall which ones are from there.
Let’s assume though that the metadata includes both a barcode and a catalog number along with the name of the label (imprint) as MB wants. Are there others like this? To consider such data primary type data (that listed on the release group page), it should apply to most releases out there.
My point is not that this information should not be entered, but that this information should not be conidered important that it be used as primary information. For CDs (as an example), it is very primary and important, so important it is printed right on the cover. But for digital, it is looking like it is only provided on releases from smaller labels and independent labels. (I mean no disrespect to independant or small labels, it is the artist and music that matters). But as we determine which pieces of data are shown to differentiate releases, if 95% of the releases do not have or disclose to people the data, I question if that makes sense.
In my opinion, it is more like supporting information as it relates to digital, especially given how MB currently treats digital. If I have Paul Simon’s CD Graceland, I will need something more than that to identify which CD I have. Starting with the barcode and catalog number will get me most of the way there, if not all the way there. But for digital, it is generally not that way. If you were to call the label behind that release and ask them how many versions of the CD they have, they (as a possibility) might give you a list of the countries and barcodes and catalog numbers, and maybe others. Now if you ask them which digital releases of that release are out there, what would they say? Would they likely give you a list of the same, barcodes and catalogs, or a list of stores like iTunes, Amazon, etc. Or maybe a listing of file formats, like we released this as FLAC, MP3 and M4A? I have not done that, so I don’t know the answer, but I do wonder since many of the album releases sold on iTunes need to get barcodes assigned in order to sell because they did not already have one. That tells me that the barcode is not so important.
I found an example that I/we can dig into and see what the deal is. Respecting your anonymity, I selected a release
This release I have, I am in need to locate my original files though. But looking at the web page portion, there is no barcode or catalog number listed there. This might be a bad example though, but I am trying to use what I have and can see vs something I am guessing with. Could someone provide a better example to look at?
Apart from a URL to the bandcamp page I don’t believe label or catalogue numbers are included in bandcamp track metadata.
However I still find it important primary information, and am happy with how we do it now (entering the label that released it into the label field). Whether or not you can use it to identify a rogue file is only one function of mb data - I find label information useful and interesting in itself.
For instance most of the releases on the grindcore karaoke web label (https://musicbrainz.org/label/4e38f3fb-d4e8-430a-964c-e9d88b589b66) are not of much significance individually, but are significant as a group/curation.