#musicbrainz

how should I tag a "protools operator" credit from a japanese release?

Misty_Pond: I'd go with "miscellaeneous support" and set "protools operator" as the task

yeah, that was my first thought

*miscellaneous

I suppose it might fall under some sort of "technical" role but there isn't a credit for that

"This interface sucks": https://musicbrainz.org/edit/77644282

chaban: I mean, the workflow for adding discids to multi-disc rels when adding releases *does* suck :/

I believe there was a ticket for that.

This one has piqued my interest as to why beginners commonly pick the wrong merge strategy: https://musicbrainz.org/edit/77659485

does it have a hint that "if you're not absolutely sure you probably should be using x strategy"

Yes, I think that warning might be too coercive.

There is MBS-4396 but it's not related to this case as the artists are indeed different

MBS-4396: Distinguish between "artist" and "artist credit" in warning notice when merging recordings as part of release merge edit https://tickets.metabrainz.org/browse/MBS-4396

Yay, 2 ISEs in a row. First for Carnegie Hall: https://musicbrainz.org/edit/77663946 and another while editing the annotation of a release at the same time: https://musicbrainz.org/release/d5281611-14d1-4...

IvanDobsky gave their best trying to explain artist credits at track-level: https://musicbrainz.org/edit/77647567

chaban: interesting. were you submitting both edits at the same time in different tabs?

Yes

Not exactly the same time I suppose. The approval was still pending.

MBS-11410: Edit could not be created while at the same time approving another one had a time-out

ok, I'll look into that. we do lock the editor table for each edit as the error suggest, which was to prevent deadlocks iirc... this was even before my time

The approval time-out had happened a few times already. Always for the same kind of URL merge. This time I switched to prod to see if it makes a difference

thank god archive.org has *some* of the old review board archived, otherwise this info would be lost forever https://web.archive.org/web/20140602202250/http...

(commit message didn't say anything)

the original deadlocks in MBS-3597 were due to the edits_accepted column on the editor table, which we actually dropped in 2016. so these locks may not even be necessary anymore.

MBS-3597: Internal Server Error (deadlock detected) https://tickets.metabrainz.org/browse/MBS-3597

bitmap: drop it for the schema change, see if something happens? :p

That's how you do schema change releases, right

no need to wait for the schema change :)

oh

So drop it and see if someone screams then

well the edits_accepted column was already dropped (in 2016), removing the locks is just a matter of deleting a few lines of code

so people will scream if it causes more deadlocks somehow, but

I'm doing some tests to see if that's the case

One more vote on https://musicbrainz.org/release/47658cbb-2093-4... so it's easier to merge further if needed? :)

oi