#musicbrainz

Rexodys[m]: You can use $num to format numbers with leading zero, e.g. $num(%tracknumber%,3) for 3 digit track numbers

Or conditionally to use 3 digits if total tracks is more than 99, but 2 otherwise:

$num(%tracknumber%,$if($gt(%totaltracks%,99),3,2))

Oh I love you! Had no idea this was possible!!

Another question. can I easily make a new release based on 2 existing releases?

This is CD 1 : https://musicbrainz.org/release/0b874f1d-b7fb-4...

This is CD 2: https://musicbrainz.org/release/078b7787-4a93-4...

Source: https://www.discogs.com/Alain-Morisod-Sweet-Peo...

finalsummer: about Picard picking the first country alphabetically: you can configure preferred countries in options, Picard will use the first one from that list if possible.

finalsummer: and there is also the %_releasecountries% variable. I just noticed that's missing in the docs

Rexodys[m]: you're welcome

For the combined release I think it's possible to reuse existing medium when adding a new release on MB

But other data you need to enter

when you add a medium you can select from an existing medium

and it pulls in the recordings too

finalsummer: https://community.metabrainz.org/t/handling-of-...

rdswift: is that something we should add to the docs? Maybe under Tutorials?

Yes, I think that's a good idea. I'll start a PR.

Top, thanks

hi here !

Here’s an interesting thing I hadn’t noticed before. Vote yes on an edit and submit. Realize you don’t actually have enough information; change your vote to none and submit. Your vote is still recorded as yes. Kind of annoying when it’s a whole page of edits…

The UI should not, IMO, permit you to change your vote back to none if it will not be recorded that way.

crism: Agreed. Please make a ticket under MBS: https://tickets.metabrainz.org/browse/MBS :)

Will do, Freso. Projects have multiplied, though… is MBS appropriate for UI?

after waking up, a few more thoughts about digital "availability" and a-tisket.

1) really, if it was per-service release events instead of per-country, I'd mostly be fine

2) if it was per-country release indicators without release *dates* then I'd be mostly fine too. so it's linked to 3 services and they all have different dates. we can't have multiple release dates per country. so we just.. pick a random one? the first one? why? "it's available from some service in that country as of date [x]" seems like a weird piece of info to be linked to the URL of the release

3) a lot of the time, the meta information differs in ways that, if it was printed on a record sleeve, would justify being a different release. for example the streaming services don't list catalog numbers, or have a different "label" listed, from beatport/juno/bandcamp. or bandcamp or beatport has a different artist listed (because of their garbage metadata models) or so on. why do we suddenly decide

that they're the same because it's digital, and collapse them, when we don't collapse because of other things like release date?

4) if bandcamp is an "XW-ifier" (the only one?) then is a corrolary to 2) that what really matters whether the bandcamp release came before the other releases? if so, it's XW alone, if not, it's... all the release countries+dates ... and XW+bandcamp date?

5) a key difference between the way that different services region lock is that some of them region lock by ip (finalsummer's example/view) and some of them region lock by region of the account ("you can use american spotify in a country spotify doesn't operate in and it'll act like you're in the US"). given these two different paradigms, what does it actually mean to be "available" in that country?

it doesn't seem to mean *available to the user* it seems to mean *legally available to the distributor*? but if it's not monitized in that country, as in bandcamp streaming everywhere for free and not paying anyone any royalties, is it actually even legally available in that country, or just no one cares?

yes, i think it would be helpful to clarify what the release countries/dates generated by a-tisket actually mean

finalsummer suggested that it's fine to just keep updating a given release with the current availability, but I don't think that is what the release relationships actually say.

to my mind, these release relationships imply that a given release was released at that date., whereas the actual countries generated by a-tisket actually imply where it is *currently* released, regardless of that date.

if this is *not* what a-tisket is saying, then I don't think we should be generating those lists. a different release relationship is necessary for the *current* availability.

Can someone please provide me with an example release that has a long list of release countries. The example that I was using for the documentation has now been changed to "XW" so it won't work for the examples. Thanks.

rdswift have a look at the release history for this one https://musicbrainz.org/release-group/53e37a58-...

there were two digital releases added, one 2012-09-04 one 2012-04-09

these had a different list of countries. i just merged them into XW as the date and thus the country list seemed supremely unreliable and problematic.

(sorry, maybe this is not that helpful)

But I need an example that has the full (lengthy) list, not "XW". Thanks.

here's one https://musicbrainz.org/release/c53c6360-9251-4...

from 1 day ago

fresh off the press!

Great. Thanks.

the editor who added that seems to add a lot of them, so there are some more examples if you look at their editing history.

Okay. Thanks again.

6) given that beatport doesn't expose what countries a release is available in, but does region lock, what "should" we set as the country for a beatport release? [unknown] ?

or... XW until we notice by trying to buy in a restricted country that it's not XW, then... <some list of countries that beatport also doesn't produce via API>?

Libera matrix bridge!

crism: Yes. Thanks. :)

Freso: do you have access to run arbitrary database queries/reports?

Well, depends on what you need. I have a somewhat out-of-date mirror db, so if it’s regular data lookup, then sure, but I don’t have access to live/prod. db.

yeah, doesn't need to be live, just needs to be from the a-tisket era

"how many releases are there where there are more than 100 countries and more than 1 release date"

my hunch is that this number is "zero"