#musicbrainz

the composer :)

did the composer name it "Two Strolls Down Memory Lane", or did the composer call it another "etude"

is CSG only going to be music by "dead white guys" -- to quote yo-yo ma

what if it's named both in the score (or recording)

I think that phrases it to offend. It's intended to give titles where no official title exists for the work, not to somehow turn "4:33" into something unrecognizable.

which happens a lot, by the way

It just happens that much of the music that has no such solid title was indeed composed by dead white guys.

ha, well let me correct you there :P

yllona, I'm not looking to debate the color of the skin of the composer of this or that work.

no my point being. CSG just doesn't work well

]

I'm pointing out that using Ellington to argue CSG is broken is a false claim, as Ellington named his works, thus CSG wouldn't apply anyhow.

QED :)

would I prefer to have full work identification built in to the db? Sure. But that's 20+ months off. A fixed CSG is doable now, today.

i'm thinking of many other composers, i'll point you to the recent article from film composers and why they have to "name" works or pieces for copyright/royalty issues

(and yet, still fail to do so often...)

that's why they have music editors. those pieces have names oar no royalties can be paid. it's required....

or "composer's promos"

that's why i said working from the score will be useful

naming pieces in a film score is often left to the music editor

Oh, I agree scores are helpful. Been reading through the score to every single work by Mozart :P

hang on for a few and i'll grab the interview, it's audio... it's the guy who wrote the score for "the incredibles" and "ratatouille" -- he's up for an oscar (again)

I'm just saying that naming this or that composer as reason CSG doesn't work, or isn't comprehensive enough, most esp when we come to 20th century composers, is imho a false argument, as CSG doesn't apply where names exist, and 20th century composers tend to give "official" names to their works, much much more so than in the past.

CSG doesn't work for number of reasons, that's just the one at hand

the other main reason that CSG doesn't work is composer as artist

how does that not work?

well it doesn't te;; me *which* version of mahler's 4th for exampel.

umm, but that's not the job of the artist field...

and more importantly, as a web service, composer as artist can't link to a concert performance

that's work identification, not composer/performer/artist/etc

given that we can't link anything to a concert performance, classical or no, I still don't see the point...

really, so? look at your local philharmonic schedule.

how does my local philharmonic schedule have anything at all to do with a listing for a CD/LP/etc in the database?

you don't? then i give up. because isn't that the whole point of a web service?

no, that would be the point behind LocationProposal

I would like to see what you're talking about too. But it's not that some field isn't correctly supporting it, it's that the fields and database structure to support it plain don't exist.

and neither which version of a work, nor where it was performed (or when), has anything to do with composer as artist.

because hopefully internet radio stations will use MB as a reference, and maybe i'm an outlier, and totally old-school, but i'd like to know who and what is on a recording. also i'd like to know when a particular work is being performed in my local area

Ok. 1) how does which version of a work is on a track have anything to do with the artist field? That's work ID - currently track field, eventually hopefully work IDs.

so yes, that's working well for "pop" music... i know that aretha franklin, jill scott, and a dzen or more other performers were in LA this month.

2) ARs exist exacly to cover much much more correctly and comprehensively the performer ID you're talking about.

but the opera i saw last night and the one next week? the only way i knew that 3 of my favorite singers wee in town was to listen to terrestial radio or read the physical newspaper.

and why?

compser as artist

as for when they're local to you, considering we have no "performance" object, nor any way to link it to a release, I have honestly no idea what you're talking about. That *is* the LocationProposal.

i have verdi as the artist for "otello", well that wasn't very useful was it?

huh?

what are you talking about

composer as artist won't work in identifying a particular production

MB has no performance object. Even were the artist the BSO, you still would have nothing in MB to tell you if the BSO is in town or not.

yep.

hence why I have no idea what you are talking about.

okay, we *are* talking about the same thing

lol

LocationProposal, right?

why wouldn't MB have BSO (with seiji ozawa hopefully) in the MB record?

yes.

but the point is, we would have the BSO

cool beans, please point me there and i'll add on to it?

they'd be AR'd to the release, just like the composer, the conductor, the chorus, etc

hence why I would love to just eliminate the "means nothing" artist field

brianfreud: extremist :P

brianfreud: got it thanks!

warp, wouldn't it be better, if you even want "artist", to be able to set "artist = this AR"

but as your view, not as a universal "truth of artist fieldness"

and yllona, OT, but you definitely should give Alexandre Krein - Songs from the Ghetto a listen :)

brianfreud: yes, i noted that earlier. can you recoomend a recording?

yllona: sorry, was afk

that's the name of it

Ghetto in this case = WW2 Germany/Poland, not "the hood" :)

yllona: in the context of my problem, the tempo "singing slowly, with tenderness", is less important than the common name, "Romeo and Juliet"

they are both artist intent. but for this work you can perfectly do without the tempi. you can'd do without the names

yllona, pulling it up, 1 sec, browser freezing on me

thus naming the track Sonata .. : II. Singing slowly, with tenderness "Romeo and Juliet" would be a total mess

not quite sure if it's properly Latinized as Alexandre or Alexander

brianfreud: yes. i know what ghetto means, i did work in the jewish ghettos in venice, back when i was doing art restoration

np, didn't intend to imply you didn't, more just clarifying, since that title could go either way :)

xlotlu: the name of the "song" would be followed with [tempi / stage direction]

unless the name of the song is of course "slowly, with tenderness" ;)

so far managed to duplicate the link that never ends :D

xlotlu: like this...

like what part of that?

xlotlu: this work has both the european tempi (and english names for those tempi)

like the first and second movements "I. Slowly - II. Slowly and Deeply Expressive"

well all the movements actually

brianfreud: i more in the 'record whatever is on the cover' camp than the 'music encyclopedia with accurate data' camp. for most releases i edit/listen to, they are attributed to an artist -- but often it's not clear (without research) what _exactly_ the artist does on the record.

and tan dun's opera "marco polo" is basically a clusterf**k of tempi and poetry stanzas. i'm still working thru the 2nd CD

warp: i garee with you there.

*agree, rather especially now with the "digital download" era.

yllona: i but that's only the 'tempo', no "common name"

yllona: nevermind it. the more i talk about it the more puzzled i get by this work :)

xlotlu: you should subscribe to grove's online (or snag a connection from the library/university)

yllona: i'm not that set on encyclopedic cataloguing of music to pay for stuff that takes time to read :)

xlotlu: grove's is definitive, please see: http://www.grovemusic.com/

are they using musicbrainz? :P

xlotlu: groves is also just interesting to read.

xlotlu: grove's pre-dates MB by *decades* ;)

yllona: so is the squid definitive guide, which i've been trying to read since yesterday :)

squid as in "food" or "programming"

piece of software

yep. hat's what i thought you were talking about. these days i wait to see if somethings going to last, before i read or buy yet another o'reilly book :P

okay here's grove's list of newly added contemporary (classical) composers and/or performers

grove's also has a comprehensive publication for jazz and(more recently) women composers

oops! one sec...

here 'tis: http://www.grovemusic.com/grove-owned/music/wha...

i'd fancy direct access to the entire listing, and see what they have to say about Henze, not only the updates :)

and *if* i were really interested, i'd subscribe... or if i had way more spare time

warp: me too.

xlotlu: agreed any of the oxford press publications are fascinating, but a guaranteed time-sink.

heh

you should be able to get a free connectin if you're stll in school tho.

stupid banks and their love of phones

brb

any coders around?

What sort?

anything dealing with regexp, I '

expect, but js in particualr

What's on your mind?

ok, I take HTML, parse it to compliant moin. No problem.

Now I'm trying to take that output and convert it back to HTML.

So I try to go after the anchor tags. And this works in a sandbox: var x = annotationString.split(/\[/);

and if I write the moin output to the screen, all looks good. But if I try to split that same string in memory, rather than using copy/paste to the sandbox, I always get that x.length = 1 after the split, not the 4 it ought to be.

So it's obviously getting stuck on something in the string that's blocking it splitting...

What is the string?

the HTML--> moin of the annotation of http://test.musicbrainz.org/show/artist/?artist...

How do you get that string?

reinventing wheels?

I already wrote one function which does the HTML -> moin conversion. This function would be to reverse that.

that's not moinmoin, btw

yeah, I know, but close enough :)

http://musicbrainz.pastebin.com/m18d9dd90 < x.length in line 83 ought to be 5 not 1

Your class regex is b0rked.

It ends up being: (^|\\s)foo(\\s|$)