I’ve been busy adding series that were published in printed productions, one of these is SPIN magazine, and some of their lists give a cataloguer/list builder/list writer/compiler credit, and there are some which have interesting relationships that could be useful, namely the series where artists chose releases they feel defined / influenced their sound:
As some examples.
@chaban removed the cataloguer relationship on these, Edit #87755506 - MusicBrainz - providing reason that these were “not a catalogue series” and linking to the style guides:
I assume this is because of the language of catalogue is being referred to for classical music; if that is the case then can we have a relationship for “popular” music?
Although this doesn’t make perfect sense as the URL provided for the relationship does not mention it for being classical and instead reads:
This relationship is used to link a catalogue work series to a person who was involved in compiling it.
Which seems to fit perfectly for the relationship link I am trying to build?
I’m also going to tag the other editors who voted Yes to this change as I’d like their input too as how to overcome this obstacle.
@DenizC , @salo.rock and @Deleted_Editor_1440677
The emphasis is on work series, i.e. this relationship is intended for series of work entities only, more specifically a catalogue series (which is a subtype of a work series). See https://musicbrainz.org/doc/Series:
A series of works which form a catalogue of classical compositions.
Should the ability to use it for non-work entities be removed?
And we make a ticket for a new ‘compiler’ relationship. @reosarevok?
(all I could find is this label > release compiler ticket)
My understanding from the guidelines and from existing usages is that the cataloger relationship is specifically for catalogs of classical works. Classical has it’s own set of guidelines and certain relationships on MusicBrainz.
Separately, I have concerns about eligibility of series based on charts. Corporations like Billboard or Rolling Stone assert copyrights on their charts. While many jurisdictions don’t allow simple lists of facts to be copyrighted, these charts are created with editorial overview and subjectivity. MB does not allow digital bootlegs based on charts, with the guidelines noting that “these playlists are often copyrighted by their issuers”. The same copyright concerns would apply also apply to a series.
Wikipedia disallows full copies of lists of articles on like Rolling Stone’s 500 Greatest Albums of All Time due to copyright issues.
It also seems that a series supposed be part of the e.g. release itself or from an official source like the artist or label and not something created by a third party. (Again, classical has it’s own guidelines).
From the series guidelines:
A series is a sequence of separate release groups, releases, recordings, works, artists or events with a common theme. The theme is usually prominent in the branding of the entities in the series and the individual entities will often have been given a number indicating the position in the series.
The problem is that we currently don’t have any way to restrict a series<->whatever relationship by the type of the entity the series is about We can say “this is an artist to series rel”, but not “this is an artist to work series rel” (in the same way we can’t say that “parent” is a “person artist to person artist rel”). It would be sweet but would involve a huge redesign of the relationship system, AFAICT.
Maybe so but there are already a considerable amount of series (not created by me) that exist in the database; for example:
Which fall folly to the same logic. Namely with this, people re-create these lists in many forms anyway; playlists exist on most streaming platforms or other databases (e.g. Discogs).
I understand that copyright understanding is different between territories and even between individuals, but I always think that simply providing a list of items in a sequence that they were printed in isn’t really breaking copyright; copying the included text with each item is.
Also another argument I have is that these lists are “locked” away in defunct publications (such as various music magazines) and it’s nice to make them available again, even for research to see how many times a particular album/artist/song was featured in such a list.
There’s sadly very little in this world that doesn’t have potential copyright concerns.
Unless we’re lawyers being retained by the relevant publisher or whatever, or we know it’s illegal, let’s let people make lists from old magazines
Copyright only applies in this way when asserted. Because a copyright holder could stop you from doing something doesn’t mean they want to, and we shouldn’t assume.
Unless you’re a lawyer who’s willing to pro bono defend MetaBrainz in court, what you think is or isn’t copyright infringement doesn’t matter all that much.
X number of wrongs do not make a right. “Sweat of the brow” copyright is a thing, and while it isn’t (generally) recognised under US immaterial rights laws, it is recognised under EU (and most European countries’) immaterial rights laws and probably other jurisdictions too. The official stance of MusicBrainz is that Series entities are not (currently) for “top X” lists, and I can’t imagine that we wouldn’t remove any such lists in case we ever receive a copyright complaint about one. Edit: We also won’t (currently) remove existing ones if the users/contributors are happy with them.
(This is also why MusicBrainz’s core data is released under a CC0 license since MB’s data itself is mostly “sweat of the brow” content, so could be “protected” under copyright in some jurisdictions. The CC0 license is used to try ensuring that it remains free.)
(Also also note that there was a case against Spotify, but it was settled out of court, so no ruling was made. It did, however, result in Spotify saying they would decrease the visibility of the playlists in question.)