reosarevok closed my favorite discourse thread? curses! foiled again!
2016-07-16 19843, 2016
chirlu
You have strange favourites.
2016-07-16 19854, 2016
CallerNo6
thanks!
2016-07-16 19832, 2016
krono joined the channel
2016-07-16 19851, 2016
krono has quit
2016-07-16 19802, 2016
JesseW joined the channel
2016-07-16 19818, 2016
hawke
CallerNo6: Yeah, that's pretty disappointing. ;-)
2016-07-16 19805, 2016
krono joined the channel
2016-07-16 19816, 2016
CallerNo6
well, I mean, the underlying questions are good ones. is MB data easy to use? if not, is there a way to improve?
2016-07-16 19858, 2016
krono has quit
2016-07-16 19804, 2016
hawke
Yeah. I'm actually not a huge fan of the CSG weirdness either.
2016-07-16 19814, 2016
hawke
It's just the best that seems to be possible at the moment.
2016-07-16 19824, 2016
hawke
I really don't like the idea of a CSG "flag", mostly because there will be edge cases and I don't see what possible benefit there would be out of it
2016-07-16 19838, 2016
CallerNo6
I don't understand.
2016-07-16 19836, 2016
hawke
?
2016-07-16 19802, 2016
CallerNo6
I don't understand the "edge cases" part.
2016-07-16 19827, 2016
hawke
I mean releases that are borderline between classical and pop
2016-07-16 19805, 2016
CallerNo6
But all a CSG flag says is "this releases was entered using CSG". Not "the immutable nature of this release is Classical".
2016-07-16 19822, 2016
CallerNo6
er, s/releases/release/
2016-07-16 19805, 2016
hawke
That doesn't really solve the problem though -- someone comes along and goes "oh, that's not classical" -- changes the flag and edits accordingly.
2016-07-16 19835, 2016
CallerNo6
I think we're talking about two different things.
2016-07-16 19852, 2016
CallerNo6
If they edit it to be non-CSG, of course they should change the flag.
2016-07-16 19801, 2016
yeeeargh has quit
2016-07-16 19818, 2016
krono joined the channel
2016-07-16 19801, 2016
yeeeargh joined the channel
2016-07-16 19845, 2016
CallerNo6
Or, maybe I should ask, what do you think the problem is?
2016-07-16 19853, 2016
hawke
I think the problem is that CSG overrides the standard artist credits and basically forces it to "composer"
2016-07-16 19804, 2016
hawke
When indeed not all releases actually do that (just most of them)
2016-07-16 19820, 2016
CallerNo6
Well, okay, that's /a/ problem :-)
2016-07-16 19842, 2016
hawke
I would say that arguably most (classical) releases don't really have track artists per se.
2016-07-16 19815, 2016
CallerNo6
IIUC, the problem ijabz is trying to solve is "I have no idea if this was entered using CSG, so I don't know what criteria the editor used to determine the track artist"
2016-07-16 19831, 2016
yeeeargh has quit
2016-07-16 19831, 2016
hawke
Yeah, that...but I don't really see how it helps.
2016-07-16 19837, 2016
hawke
So you know the artist is the composer...how does that help?
2016-07-16 19849, 2016
hawke
You can already find that out by comparing track artist to composer AR
2016-07-16 19805, 2016
CallerNo6
If there is one.
2016-07-16 19837, 2016
yeeeargh joined the channel
2016-07-16 19841, 2016
hawke
yes.
2016-07-16 19819, 2016
hawke
but my point is what do you as a tagger/player/whatever do differently if you know the track artist is the composer?
2016-07-16 19845, 2016
CallerNo6
presumably ijabz has some ideas about how to display them differently.
2016-07-16 19844, 2016
CallerNo6
or what if you want to override CSG? it would help if you /knew/ it was a CSG release.
2016-07-16 19801, 2016
CallerNo6
when tagging your own files, I mean.
2016-07-16 19827, 2016
hawke
I think the comparison, "does track artist equal composer" covers that though. And if it's missing then fix the data.
2016-07-16 19851, 2016
CallerNo6
I mean, if "CSG weirdness" is a problem , then doesn't flagging something as CSG at least help mitigate the weirdness?
2016-07-16 19838, 2016
hawke
That just bothers me aesthetically.
2016-07-16 19808, 2016
CallerNo6
Okay, you're right, if I'm tagging my own files then I'll fix the MB entry (if it's incomplete).
2016-07-16 19828, 2016
CallerNo6
Not every Picard user wants to be an MB editor.
2016-07-16 19804, 2016
CallerNo6
(although maybe they would be if they knew that the mission statement is #MitigateTheWeirness!
2016-07-16 19805, 2016
CallerNo6
)
2016-07-16 19803, 2016
CallerNo6
grr. Typos are ruining my bad jokes.
2016-07-16 19816, 2016
hawke
I really would like to know how one handle it differently though, if you knew it's csg
2016-07-16 19822, 2016
hawke
how one *would*
2016-07-16 19818, 2016
CallerNo6
Well, okay, so you're going to display a track, and you want to overeride CSG track artists.
2016-07-16 19810, 2016
CallerNo6
You could compare `track artist` to `composer`. but what if, I dunno, Taylor Swift is the composer and the track artist?
2016-07-16 19847, 2016
CallerNo6
So then you look at the recording rels to see if the performance credits are different.
2016-07-16 19848, 2016
CallerNo6
Taylor Swift is almost certainly credited for vocals, but is that enough to go by?
2016-07-16 19811, 2016
CallerNo6
that's a lot of hoops to jump through when all you want to do is decide "should I use the `track artist` or `recording artist` in my display?"
2016-07-16 19839, 2016
hawke
hmm.
2016-07-16 19848, 2016
hawke
Couldn't you just assume recording artist then?
2016-07-16 19858, 2016
CallerNo6
I guess you could use `recording artist` by default ... er, that ^
2016-07-16 19859, 2016
hawke
It really depends on what you're trying to do, I think.
2016-07-16 19848, 2016
CallerNo6
Me? I'm trying to avoid invoicing. It gives me panic attacks.
2016-07-16 19802, 2016
hawke
Personally I think the best solution would be allow there to be no track artist at all.
2016-07-16 19817, 2016
hawke
Then most classical could have no artist
2016-07-16 19824, 2016
hawke
so it's neither ugly nor 'wrong'.
2016-07-16 19829, 2016
CallerNo6
I'll have to think about that one.
2016-07-16 19825, 2016
CallerNo6
I'm not a die-hard composer-as-track-artist partisan either. But changing it at this point would be huge.
2016-07-16 19844, 2016
hawke
It would, yes.
2016-07-16 19857, 2016
hibiscuskazeneko joined the channel
2016-07-16 19808, 2016
hawke
I really like the recording-artist-as-performer bit, track artist is way less important to me.
2016-07-16 19805, 2016
CallerNo6 is trying to understand: <hawke> I would say that arguably most (classical) releases don't really have track artists per se.
2016-07-16 19856, 2016
kartikgupta0909 joined the channel
2016-07-16 19819, 2016
JesseW has quit
2016-07-16 19841, 2016
JesseW joined the channel
2016-07-16 19841, 2016
JesseW has quit
2016-07-16 19810, 2016
hawke
CallerNo6: I mean that for most or at least many classical releases they are closer to ARs than track artists.
2016-07-16 19831, 2016
hawke
i.e. it is better in the composer field than a track artist.
2016-07-16 19846, 2016
hibiscuskazeneko has quit
2016-07-16 19823, 2016
hibiscuskazeneko joined the channel
2016-07-16 19854, 2016
hibiscuskazeneko has quit
2016-07-16 19811, 2016
hibiscuskazeneko joined the channel
2016-07-16 19801, 2016
hibiscuskazeneko has quit
2016-07-16 19823, 2016
hawke has quit
2016-07-16 19838, 2016
hawke joined the channel
2016-07-16 19839, 2016
thebradleys joined the channel
2016-07-16 19852, 2016
kepstin has quit
2016-07-16 19848, 2016
kepstin joined the channel
2016-07-16 19846, 2016
zas has quit
2016-07-16 19801, 2016
thebradleys
This discussion always comes down to ambiguity of "artist" (composer vs performer in this case). If we were to do anything to flag the style used, I would think it would be "track artist is composer"/"track artist is performer" clarification.
2016-07-16 19804, 2016
zas joined the channel
2016-07-16 19840, 2016
reosarevok
hawke: the main utility of a flag is allowing stuff (from userscripts to eventually a full-fledged classical-oriented editor if someone wrote one) to apply to releases marked as classical
2016-07-16 19847, 2016
reosarevok
Or well, as "csg" anyway
2016-07-16 19813, 2016
reosarevok
Apparently, it's useful for some people in Picard and the like - I'm not sure why, but it doesn't really hurt and it doesn't change how we store the data, so why not
2016-07-16 19808, 2016
GuidovanPossum joined the channel
2016-07-16 19823, 2016
thebradleys
When I tag with Picard, I get most/all the relationships into the tags and they work great. That might be because the player I use handles the collections of artists, though.
2016-07-16 19855, 2016
thebradleys
It was fun to ask for all the recordings that had some person playing solo violin, though!
2016-07-16 19854, 2016
CallerNo6
thebradleys, your player let you filter for solo violin?
2016-07-16 19809, 2016
JesseW joined the channel
2016-07-16 19856, 2016
reosarevok
We all deserve a player that does!
2016-07-16 19857, 2016
thebradleys
Musicbee
2016-07-16 19842, 2016
thebradleys
When I was looking for one, it was one of the few that would split collections of artists in the tags
2016-07-16 19856, 2016
reosarevok
Aww, Windows only
2016-07-16 19805, 2016
reosarevok
Although I mostly stream stuff anyway so
2016-07-16 19826, 2016
reosarevok
(It's pretty nice how good the classical catalogue in Spotify is - also pretty sad how hard it is to find things that are indeed there...)
2016-07-16 19856, 2016
thebradleys
Streaming is a whole different issue - we don't get much control over the source tagging and search.
2016-07-16 19846, 2016
reosarevok
Yeah
2016-07-16 19852, 2016
reosarevok
Spotify pays for our data
2016-07-16 19802, 2016
CallerNo6
So what would be cool is an mb-aware playlist manager that worked w/ streaming sites
2016-07-16 19805, 2016
reosarevok
But it doesn't use it for making classical good at least :p
2016-07-16 19856, 2016
thebradleys
I was excited to see Spotify using the MB data - I'm sure it would take them a lot of effort to search all of that data. Apart from searching, if you at least had the mbids in the stream, the player could pull all kinds of good information by itself.
2016-07-16 19805, 2016
CallerNo6
totally. if spotify, if you only do one thing with MB data, please use/expose MBIDs
2016-07-16 19849, 2016
yeeeargh has quit
2016-07-16 19815, 2016
regagain has quit
2016-07-16 19804, 2016
chirlu has quit
2016-07-16 19827, 2016
hibiscuskazeneko joined the channel
2016-07-16 19830, 2016
chirlu joined the channel
2016-07-16 19842, 2016
reosarevok
oh, hawke: another possible use would be for me (or even better, a group of us) to try and review all new additions with that flag to fix them up :p
2016-07-16 19847, 2016
thebradleys
You might even be able to query against the relationships if they are available and compare since you know which ones should be involved.
2016-07-16 19833, 2016
armalcolite has left the channel
2016-07-16 19803, 2016
krono has quit
2016-07-16 19841, 2016
CallerNo6
reosarevok, doesn't that assume that somebody 1) entered a release poorly, but 2) knew enough to flag it "CSG"?
2016-07-16 19854, 2016
reosarevok
Yeah
2016-07-16 19800, 2016
reosarevok
That's why it's doable :D
2016-07-16 19819, 2016
CallerNo6
I like you, kid. You've got spunk.
2016-07-16 19840, 2016
reosarevok
(there are waaay too many classical releases being added to be able to check them all easily, but it might be possible to at least try to make sure the ones marked CSG are actually decent CSG)
