Hi, sorry for the possibly stupid question, but this is about two releases by the Belarusian musician Ivan Kirchuk/Іван Кірчук, which are largely but not completely identical:
“Heritage of the Lost Villages” released 1999 by PAN in the Netherlands and Спадчына Загінуўшых Вёсак (Spadčyna Zahinuŭšych Vësak, same meaning) which was published 2000 in Belarus. Discogs treats them both as a single album, even though their contents are not completely identical. For most tracks, only the script is different (Belarusian Łancinka vs Belarusian cyrillic), but some tracks also seem to be different, notably the very first track (which is incredibly beautiful appears absent on the Belarusian release. But also the order of the tracks isn’t completely identical.
How does Musicbrainz deal with this? What I did so far is 1) add the diacritics which were missing from the track list of the Dutch release and changed the artist so that it is clear that Ivan Kirchuk and Іван Кірчук are identical, and it seems I will have to wait 7 days for these changes to take effect (why this crazy long wait?). But again, is there any relationship type that is appropriate? Also between the individual tracks on both releases as the order of the tracks isn’t identical
What you need is not a relationship between the 2 releases or release groups.
If I follow all that you say, you just need to merge the two release groups, so that the 2 releases will be grouped together.
Exactly!
See that this is a master release at Discogs, not a release:
Discogs master release = MusicBrainz release group
Discogs release or version = MusicBrainz release
With all that you say, I don’t see a problem in merging release groups:
Difference in naming the tracks is absolutely OK, as long as the track audio content is the same.
It’s a frequent situations with albums issued in other countries, with different writing scripts.
When some tracks are missing/added, it’s also absolutely OK.
Think of the album deluxe editions with bonus trucks.
If they are the same track audio content but they are simply in a different order, it’s also once again OK to keep in the same release group.
And if the track audio contents are the same, recordings should be merged as well.
Thanks, I guess for that to be possible, the two artists have to be merged first. Right now, musicbrainz still thinks that Ivan Kirchuk and Іван Кірчук are two separate artists. I have already made an edit to merge them but it takes 7 days to complete, and then, merging the release groups takes another 7 days. This is crazy. I have been a contributor to Musicbrainz since 2003, and still these over-the-top restrictions apply? Around the same time I joined Wikipedia and there, after a certain number of approved edits, your new edits get auto approved. Why doesn’t musicbrainz have something similar? As it currently stands, it really deters contributors.
For each MusicBrainz entity (release, label, recording, release group, work, etc.) there are two important docs to read before everything else, its definition and its guidelines.
But as a matter of principle, the 7 day wait is ludicrous. In Wikipedia, there is a mechanism, where edits by new and anonymous editors have to be approved, but when you have a certain number of confirmed edits, edits you make after that get auto applied. I’m sorry, but on bad days this feels like a smack in the face.
I wonder what to do with the tracks. E.g #16 on the Dutch release A Ŭ Našaha Svata is identical to #17 on the Belarusian release. А ў нашага свата. Does one also have to do merge edits for these individual tracks?
Those two artists have been in the database since 2006 and 2014. Waiting 7 days to make sure no one wants to comment is not long when compared with those time spans. There is no “undo” option once a merge has completed.
When I look at my edit history, out of 3840 edits 3 have been voted down. So, judging from my personal experience I would say this is completely unnecessary. Again, I have contributed to Musicbrainz (as well as to Wikipedia) since 2003, so I would assume that at a certain point you earn some amount of trust so that each of one’s edit doesn’t have to be quarantined for a whole week. If anything comparable had been in place on Wikipedia, I swear, most editors would have given up quickly and it would never have become the super complete free encyclopedia that it is today.
As much as I love most aspects of Musicbrainz, I just find this endlessly frustrating and bureaucratic.
In my humble opinion, I think that the voting system is ok and it’s kinda necessary, but the underlying problem here is that we don’t have enough voters on the site
They aren’t “two artists”. I don’t know if you read Cyrillic, but if you do you should understand that this is the same name, just in two different scripts.
You see, only today (after 18 years of contributing to MB, on and off) I learned about the existence of the “voting threat” where one can canvas votes. For all these years, I endured those superfluous stretched out quarantine periods for countless times. They are a superb productivity killer, I’d say. And I guess that they absolutely deter new users. Really. Why don’t you look at how Wikipedia handles similar situations?
But I guess, a few years from now, MB’s relevance will have declined quite a bit anyway, as the number of CD sales is rapidly declining and everyone uses streaming services now. I only got to use it again, because these days I had ordered two CDs that I just wanted to own physically, because of how great I found them. But since my son “forced” me to buy a family subscription to Spotify (2,5 €/month per person) I find myself using streaming most of the time and therefore I rarely have need for MB services. I resisted for as long as I could but the day came eventually…
It is frustrating, but you get used to it. I think it put me off for a while too…
However with 0 votes there is no applicable experience to apply, sorry
If you are just editing stuff when you want to then I’m sure it’s a big hassle. If you’ve been keeping an eye and voting on any amount of edits the one week wait won’t seem very long anymore! (It used to be longer, ai ai ai)
Particularly if you’ve had to untangle a disastrous merge…
I think MB could do away with the wait period if it was easier to undo some of the edits. Note that only ‘destructive’ edits, that are hard to undo or delete information, have the waiting period applied. Wikipedia has the upper hand here, where you just have to dig into the text history to undo a change, so there’s really no need to give people time to have a look.
Well, that’s the way I have been using MB for 18 years now: Whenever I bought or borrowed a CD for ripping, I added it to the MB database, because my musical taste is at times a bit exotic, so that very often, those CDs are missing from the datebase, so that I have to add them first before I can do a proper rip with proper metadata (I use a simple Linux ripper app called “sound-juicer”, which has MB integration). In essence, it is something I am doing for myself in the hope that it will sometimes also benefit others. And whenever I make a typo while entering a tracklist, I have to wait for days on end for the fix to be applied. That’s the irony. The initial tracklist becomes visible immediately, but a small correction of a typo needs either 7 days waiting period or several votes. I find this absolutely superfluous. Moreso, since such minor corrections don’t have any “destructive” potential. This is not about merging releases or artists. Simply correcting one wrong letter in a track title as I did today. It is completely Kafkaeske, that this needs such an extended waiting period, or several votes or manual approval by someone with extended privileges. Seriously, do you really think that in such a case, there is any need for discussions, votes etc? I mean, people have better things to do, haven’t they?
You don’t vote on edits so you have never seen the edits that get blocked (phew), and the discussions that are allowed to happen on controversial edits, thanks to the 7 day waiting period. Or the destruction and work that can be caused by a ‘simple’ artist merge.
Of course you are absolutely right though. It sucks. Nobody likes the wait! But MB is doing it’s best.
Over the last years:
Waiting time halved
Lots of edit types have had the wait period removed
Edits made within 24 hours (if I remember correctly) to a release you’ve added are immediate
If you have more ideas for what could be sped up safely I would make a ticket! You will get dev input into the suggestion then.
I can’t confirm that. Again, just changing “Du” to “Da” in a track name was supposed to take seven days. But, yes, I remember that it used to be even worse.
Sorry, I mistyped. I meant “two entities that have been listed in MB as two artists but are actually the same person”. I have also merged similar myself. I understand totally what you are doing there and why. And this is why 7 days really isn’t long for an irreversible change. As @aerozol points out, Wikipedia has an Undo option. Something MB is not able to do.
I deal with the 7 day delays by working on overlapping artists. One day on Artist A, next day on Artist B. Or just working on a different release in Artist A. Sometimes fixing a release will take a month or two and yes, that is silly, but it is a short time in the length of a database.
Personally I don’t use the “voting thread” as I don’t mind waiting. I also use the seven day period to review my own edits before they complete.
What I would love to see is an option in Picard to allow it to see my edits and use them. It is the tagging that can be a headache when I know I have made a typo correction on a track list. The good thing I notice is that many typos on a new track list can be corrected within 24 hours - but that has limited scope as doesn’t fix a recording.
