Digital releases: Merging? / Long country list? / Just [Worldwide]?

That would be a nice solution for browsing.

Updating every digital release available on one of many streaming services whenever the regional availability of one of these is updated (often for completely inscrutable business reasons) does not seem feasible.

When data through sheer scale appears to imply a level of accuracy which is not actually there, there very much is.

9 Likes

Neither it is to record every piece of music ever written, recorded or released, but here we are.

And what are the rules for this implication? In releases where there is only one country registered, is it implied that it was released only in that country, or that it was released including that country? What if it had three? six? ten? I don’t think a list of release events in MB could ever be considered complete, so why would this accuracy should be demanded just for releases listing dozens?

Puting it in another way: If what you’re saying is that listing that many countries implies that any country not in that list doesn’t have a release event, then that is an erroneous implication which fault lies on the interpreter as MB desn’t have a way to asses such claim about any release.

2 Likes

Are you going to enter + vote on those edits? Because I definitely won’t. Store availability data would need to be centralized (i.e. managed in one place in the database) if this should become the official strategy.

I don’t mind many release events in general, but in the specific case discussed above, the release events the API claims are so obviously untrustworthy that they should be disregarded completely, IMO.

4 Likes

Spotify is not the only digital vendor/distributor. The fact, e.g., that India is in the API tell us that the release is licensed there; if Spotify became available in India only in 2019, that doesn’t mean no other store had it before then.

I must clarify that those edits are the result of gathering as much info as possible, not just random data. I even, when possible, query every iTunes store (arguable, the most widespread vendor in the world). The fact that most vendors don’t have an open API or even show the complete data in their pages doesn’t make it easy, but one have to try al least.

4 Likes

@marlonob so you stand by these mass release events then ? I don’t understand your point?

Yes, I do. They are useful and better than any generic moniker.

4 Likes

useful how the information is based off of one API adding unverifiable dates of no use to anyone ? it also breaks layout so how is it by any means helpful? it is destructive and misleading

Missing territory information. Normally that means it has been taken down. But not that one, I can stream it just fine.

BTW, distributor is Ditto. They distribute globally by default.

@chaban so that means what in regard to the data is it worth keeping or not even im getting confused now lol

I don’t think that word means what you think it means.

I’m tired of getting over the same argument again and again. You guys are splitting hairs over the imperfection of a method of entering data (using the official Spotify and iTunes API) as if this were introducing inaccuracies otherwise not present in the DB.

Any digital release have a problem with dates; the bast majority of them lists only the first release date (not the digital one), even claiming sometimes that it was released in the 60s, 70s, now, those are easy to spot as they’re impossible, but many albums released, say, in the first half of the 2000s were not released digitally then, but now they are, and the date they use is the date of the CD release. Are we going to ban all such entries because we don’t have any evidence of the album being available digitally in that date, but only proof that it is available now?

If someone have evidence of a case where the Spotify or iTunes API is inaccurate, they can explain their reasoning and change the data: that’s how a collaborative database works. I think the policy for entries in the DB is that an official source is to be believed until proven otherwise. It is simply absurd to pretend discrediting a whole source of data just because there’s a risk of inaccuracies.

Mention any way of entering a release and I will have a list of problematic issues with it, just as long as you have with this.

Now, if you allow me, I will voice my personal opinion:


You (a general you) are searching for any and all flaws in this method, and then treating them as absolutely unacceptable, to justify that the reality is that you think is ugly and unnecessary.

Furthermore, you may think, “Andorra?, Liechtenstein? Why is this editor throwing this unheard-of countries at my face? What’s wrong with my normal, North-American, Europeans and other selected countries that dominate the internet? Or, if you wanna be super multi-cultural, why not just say worldwide and be done with it; just as any music exotic enough is ‘world music’, period?”.

You are projecting a cultural upset as a concern-for-the-integrity-of-the-data.

So, suppose an Andorran is the first to enter an album by a British band, and they use an importer for the iTunes Andorra store, so Andorra is the only country listed: Are you going to remove that country and replace it with the UK. Are you going to school him for not doing “enough research” and add the UK themselves (a thing that has happened to me in the case of Mexico)? Probably you are just going to add UK and move on, right? Suppose now that a Liechtensteinian add their country with their local iTunes store as reference; and then a Paraguayan, and then a Malaysian, and so on until the release looks exactly as it does using the APIs. At which point are you going to override those entries and just use worldwide or any other label that isn’t as confusing to you?

This posture has no solid stand.


If someone wants to really discuss this seriously (i.e., defining basic concepts, breaking the issue in its main points, keep focused in the topic and not tip-toeing towards the detail that are more favorable to them, etc.) I’m open to it. Otherwise, life is too short.

I’m just gonna keep using official sources to add entries to the DB, if someone have evidence of some of it being erroneous, I would gladly change it or vote for their edit.

7 Likes
1 Like

@marlonob: Discussion about Release countries is one thing; but your adding every existing iTunes store Edit #63621568 on this release https://musicbrainz.org/release/128fea20-85f5-4604-9c07-6792bf5a369f (and maybe more) is absurd and clearly against our guidelines: https://musicbrainz.org/doc/Style/Relationships/URLs (see multiple relationships).

4 Likes
1 Like

I think we need to assume that world just means as many counties as humanly possible without overpopulating the list at this point and i dont think dates should accompany digital releases

1 Like

Typically albums in Australia are released on a Friday.
I believe in the UK and USA albums are typically released on the Sunday.
So what date would you record?

1 Like

There is also the global release day:

2 Likes

This is a really bad idea, there are plenty of digital releases with perfectly valid and concrete release dates. For instance Bandcamp or self hosted releases.

Do you use digital releases at all? Do you add them or tag with them?

8 Likes

Thanks. I will not do that anymore.

2 Likes

@ aerozol
also a good point but im at a loss then for how to handle digital because to me a release date is not the same thing as a license date which is always in flux depending on content

1 Like