#musicbrainz

They are all owned by a company, RCA(or whatever)

RCA? RCA Victor?

Whatever it’s called

…anyway, that is itself *also* a label

So yeah, those logos printed on the release *are* the labels, and are representative of the company who owns them.

And for better or worse, we (and the companies too) tend to conflate the logo with the company.

maybe.

Not sure what we would gain by having a separate entity for corporations vs. labels

The corporations themselves conflate the two in casual conversation.

But in your RCA <color> example, if the release had both logos, I wouldn't intepret that to mean that both were the "label" of the release, unless there were two cat#s.

Right, but what does “the label of the release” mean?

“the label of the release” appears to be bigtime

but the cat no. is RCA

I admit, I can't get reality to fit my model sometimes.

Similarly https://beta.musicbrainz.org/release/75541e3c-d...

reality is woefully inconvenient like that

yep

ianmcorvidae gets an "amen" whether he asks for one or not

On the back cover of the latter, I see BMG and RCA, and the “nipper” trademark — I can’t see the beggar’s banquet label that appears on the center label…but it’s too small to be sure

So I mean…wtf is “the label for the release”?

I see it as almost the same question as "what series is this release part of".

I can get behind that…but I think in this case it is part of several series

There are two concurrent US 1" vinyl releases in the RG. I don't get it.

One is bigtime, the other is beggar’s banquet.

As an extra bonus, both have the same (RCA) cat no.

Well, yeah, but concurrently? Does that happen? I guess it could.

The US might be wrong; beggar’s banquet is a british label.

Anyway, I totally agree that the stuff printed on releases sometimes defies simple categories. But does that make every logo equal?

but then, bigtime is australian, so…

CallerNo6: No, probably not. As in the above (bigtime), I even added RCA purely because of printed text and the cat no.

hawke_1: I might have done the same thing, in a weak attempt to impose meaning on cat#s.

It’s quite likely that the RCA logo appears on the back cover too (thus truly justifying it), but we don’t have any artwork for that…

CallerNo6: And you think it impossible for it to belong to several catalogs?

Not impossible, but not common.

ATMO.

?

http://www.urbandictionary.com/define.php?term=... , def #1

ah

(I think only Richard Sachs says it)

I think I agree with that definition, but I think the 'several catalogs' thing is more common than you do…especially when you consider the RG as a whole.

Could be.

what’s the usual process…something like: released on small indie label, released on major label, sometimes with local imprint…

…reissued on major label… sometimes reissued back on small label…

And with each one they grow some additional logos. :-)

The one that throws me is when a faithful reissue duplicates the original cover exactly, so it has the original imprint, but that imprint/label has been defunct for decades.

hehe, yeah…

The new label might even use the old cat#, even though it doesn't fit in their numbering scheme.

If they aren't sure what to do, they should ask teleguise.

(the labels, that is)

You see that a lot in the CD reissue period as well.

leaving aside defunct labels, the original label reusing its vinyl cat no. for the CD version.

When they do that (re-use the cat#), is it generally an identical tracklist? I'd be put off if the same cat# might/might-not have bonus tracks.

Haven’t noticed on that.

To re-cap, I admit that the most pragmatic solution is to enter whatever is on the back cover that looks vaguely label-ish. But it feels wrong to me, then, to call that a "label". It's just logo soup.

The cat no. is paramount for you, then?

I'm not sure. I think the catalog itself is.

I ran into a case this week where the vinyl and the digital version shared cat #s but not tracklist

http://www.beatport.com/release/the-way-we-used... vs. http://www.discogs.com/Cevin-Fisher-Check-This-...

labels being paired with catnos seems to imply that was what was intended too. if we just want to link to a load of arbitrary labels, we already have label-release relationships, if someone would request some :P

How do you identify a release as being part of a catalog though, if not via the cat no.?

maybe sometimes you can't? Not a satisfying answer.

I'm not really the right person to ask there. I like editing japanese stuff :P

I mean, like that love and rockets release I see it as “in the RCA records catalog [as determined by cat no], available in your area via local labels X, Y, Z (bigtime, beggers banquet)”

but maybe not; http://www.discogs.com/viewimages?release=1246521 is vertigo/beggar’s banquet.

So BB is common across most of the releases …that suggests that it is the label behind the creation of the release

I don't know how to fit that into my conceptual model, so reality must be wrong.

:-D

that might just be crazy enough to work :P

I *think* the solution is to somehow store what each label’s role in the production of the release is, though I’m not sure what roles there might be.

heh, I was saying that to nikki the other week

not that I want to make our interface any more complex :/

published, manufactured, …I have no idea what the role of BB is on that set of releases.

So, we don't choose one "label" out of the logo soup because it's impossible to know what each logo did, but then we introduce roles for each logo?

well, the roles that I was talking about were more mundane

like "spine" and "back cover" more than published/manufactured :)

CallerNo6: It’s not impossible to know what each label did; it’s difficult to say which one is somehow the most important.

So a distributor is only a "label" if they have a vanity logo?

That seems as arbitrary as any other way we might do it.

Seems pretty reasonable to me — to the extent that we are tracking labels and not companies.

We could certainly *also* have an AR to a possible new corporation entity, e.g. “[corporation] distributed [release]”

are there doujin CDs? i guess so, huh?

"we are tracking labels"? Or are we tracking logos?

CallerNo6: I thought I had made my perspective clear: the logo *is* the label, like with those RCA [color] seal things.

…because there is nothing else that defines a lable.

*label

It’s not a division of the company, it’s not a company itself, it’s … well, the brand (or logo)

Labels are logos, but that doesn't mean (to me) that every logo is a label.

No, but most of them are.

There probably is no answer to this that would satisfy me. I'll sit back down.

I have another bad request from picard (including yesterday's fix)

http://www.doc.gold.ac.uk/~mas01cr/tmp/mbrachur... contains the url that picard tries to get, receiving Bad Request (QT code 299, HTTP code 400)

help?

if I copy&paste that url it works

hm, yes. Picard's attempt really doesn't though

hm

I'm confused

ok, well, I'm going to ignore that for now; it's not on the critical path

I think you would have to resubmit it.

(How do you know the discID for sure?)

because it's in the ripped files I have on my hard disk

it is?

it is!

I didn’t know anything stored that…

sound-juicer stores musicbrainz_discid in rips

ahhh

(and I've found a userscript that allows relooking up, which is good because it means I don't have to find the physical CD yet)