#musicbrainz

fog in your kitchen.

hmm...

Hmm, seems 100% relative humidity means it's at the dew point.

Apparently, it's also possible to get above 100%.

supersaturated?

under pressure?

Supersaturation.

This release-url bug is playing hide and seek with me ;(

I seem to recall that it only works for open edits

Have you seen partofaset ARs showing up in it?

I think it only does that when you specify an editor, but it was also showing when I did it for "Any"...except now it isn't.

not that I recall

you can also get hot ice

ruaok: Why is ISRC not an analog to PUID?

because it doesn't usually make sense to get PUIDs for an entire album.

whereas ISRCs may make sense.

ruaok: You wouldn't by any chance be tinkering behind the scenes somewhere? Now I can't get the bug to replicate at all :(

Why would ISRCs make sense?

navap: nope.

CD lookup for instance.

nvm, it's back.

fetch all ISRCs for a given CD? sounds reasonable.

there are not as many duplicates as PUIDs

isn't ISRC a full-release fingerprint?

no

ISRC is a track id.

If I have a set of PUIDs, it would be valid for me to see what releases match that too, no?

sure

but no one has asked for it and we didnt make it like that.

Not a track ID in the MB sense.

no

The "give me a release that has all of these PUIDs" would actually have been helpful to me.

its an id assigned by labels.

is it fingerprint-based?

xlotlu: no

for instance. you can't do that for PUIDs

ruaok: I think "fetch all ISRCs for given CD" is /ws/1/release/{releaseid}?inc=tracks+isrcs ?

yes

Wow, on the same page.

just like the link I just posted. :)

Exactly.

So, I agree that would be useful.

k. thanks.

so the ISRC would be the long wished-for unique id for a recording

nikki: The bug I "found" was essentially the same thing you said about it only working for open edits.

xlotlu: no, thats the track MBID.

one should not really use ISRCs.

they are somewhat unreliable, I hear.

so for a stable id, always use MBIDs.

track mbids aren't ids for *recordings* though

ISRCs should be used for looking up MBIDs.

xlotlu: I think the answer to your question is yes.

nikki: not yet. :)

I'm not sure why ruaok said no.

ruaok: the exact same recording will have different MBIDs on different releases. but up-upstream it will have one unique ISRC

OIC.

yes. however, sometimes tracks get passed from one label to another.

or released in different countries. and get different ISRCs.

see the link I just posed.

one track has US and GB ISRCs.

so, its not a unique is in they way you're hoping.

our track ids wile be post NGS.

that make sense?

Right.

Okay.

but PUID is (rather) unique in the sense i'm hoping :)

Did session get into NGS?

xlotlu: alas no.

xlotlu: definitely nt.

hmm...

PUIDs are fuzzy at best.

pbryan: session... remind me what sessions were?

hence "rather"

Recording session.

Whether in studio, or live.

better ask luks that one.

I'm still a little shaky on the track level things.

Okay.

that patch hasn't hit the ML yet.

I just remembered, amazon.cn list isrcs on their pages, but they always start with CN, so I guess they always get assigned new ones when they get released in china

yep.

Whats the deal with Amazon.cn? Everyone seems to ignore it.

I know they don't support ASINs, is there anything else?

there's no webservice for it either afaik

is it even amazon?

it's owned by them

they *do* have asins embedded in the pages, but they don't make any use of them that I can find, and they don't work on other amazon sites either

but the urls still use some other identifier thingy

that was pain enough to quit while I'm ahead.

time to see if its cool enough to go home.

whooo hoo.

1F above outside.

looks like I managed the cooling in my apt. sweet.

ruaok: What are you using for cooling?

fans.

placed in windows sucking cold air in.

one intake and one exhaust fan.

Ah, neat.

"20th Century Masters: The Millennium Collection: The Best of Various Artists: Millennium Sampler" The title has a subtitle, which then has a subtitle, which also has its own subtitle :/

ruaok: pong :)

Hey brianfreud.

navap: Amazon.cn was a different company; Amazon bought them a few years ago, but when they brought all the other AZN domains into one standard ASIN structure, for whatever reasons, they didn't do the same with .cn. I think, but am not sure, that it really is only AZN in name, but unlike the other AZNs, does not actually share the same distributors, etc.

hey pbryan :)

hmm I see

hey. of course it escapes me why I pinged you. :(

Thanks for pulling that RFC; I was just about to actually suggest it to you in pm for exactly those reasons, when I saw your note :)

brianfreud: well, the other sites were already using asins...

I'm wondering if I can convince you to revert the edit to the example for the time being.

pbryan: problem is, noone at least seems to disagree with my statement that that no space structure breaks on some other numbering schemas

brianfreud: I see that. I just don't think it's going to help. There are two issues in my mind...

Chris wrote that guideline to be massively expansive in its scope, but wrote the examples with solely the [0-9] scheme in mind

1. The edit being made to the example without consensus.

2. The issue of what makes sense for part numbers.

although I actually suspect the chinese site is kept separate because of chinese stuff being so controlled

I think discussion of 2 is worthwhile, but I'm concerned that 1 will impede discussion.

pbryan: The edit to the example was made 4 or 5 months ago. The only person who even said anything about it was Chris B, who reverted it (and everything else I'd written there, including the table of more complex examples based on the guidelines) and then emailed me (pretty nastily, tbh)... but like 3 *months* after my edit.

Hmm.

But nothing in anything I wrote there had anything to do with changing anything at all in the guideline itself.

Now, I agreed with him, in that my wording in describing that table of complex case examples was too restrictive, and I edited the language when I undid his revert.

I think if your change might not have been accepted if it had been published to the list for review prior to changing the wiki page.

But I think he's being far too posessive over the guideline; an example is an example, and is not a guideline in of itself.

On that point, I think at best, it's a split view.

If he wanted it to specifically be "1-3", that should be in the guideline's text, not left to be assumed from a single example.

4 people expressed opinions on whether example constitutes part of the guideline. 50-50 split.

My change to the example was perfectly allowed by the guideline's text

So would a reversion be allowed by that logic.

The "Guideline" was never changed, but all the same, it is a big change.

Yet, everyone who's commented on it also agrees that yes, that particular no space structure indeed does break on other numbering schemes.

So as I see it, we have 3 choices:

1) Let them break. Force "Parts 1-1-1-3", just to get "Parts 1-3"

brianfreud: where do we use "parts 1-1-1-3"?

2) Make the spaces universal, so "Parts 1-1 - 1-3" and "Parts 1 - 3"