#musicbrainz

rdf ftw

?

MB is also the first listed community relationship. yey.

it's rather messy. as far as i can tell you can't describe e.g. mozart's requiem with it

gafuuu: helloo :)

why not?

hi yvesr, lol

xlotlu: it indeed needs a bit of spring-cleaning

but i don't see why you couldn't describe mozart's requiem with it?

(if you want to get into much details, you may have to fall back to a more specific vocabulary though)

get the may queen

yvesr: see http://wiki.musicbrainz.org/CSGStandard/Mozart#... (warning, huge page)

xlotlu: hmm, i don't see what could cause troubles, here - you'd have a mo:MusicalWork for the whole (with two composers), and a number of movements - the different completion would be captrued by a set of events (in a way similar to the Arrangement concept)

and symphonick. and cooperaa.

because of the per-track compositions

xlotlu: i started importing terms from BrianFreud schema for musical work (he did an amazing work for that)

in MO, a track and a work are rather different - you have separate entities for works, arrangements, performances, scores, tracks etc.

the idea being that you can always decompose any of these entities

so you can perfectly state that a part of the requiem was composed (or completed) by sußmayr

as you can state that a particular performance used two different completions (eg. Beyer and sußmayr)

good afternoon

good morning :)

is it by any chance a known issue that disc id calculation in picard can result in an incorrect track length for the last track of a disc?

Some CDs have copy protection for a couple of seconds at the end, but, iirc, it shows as a seperate track.

two discs both have a about 3min shorter track at the end

I'm looking through the bug list but can't really find anything similar yet

Interesting that EAC gets it right.

indeed. At first I was thinking about programs like anydvd but that would have caused the same result in other programs

funny you should say that Prodoc, I'm seeing three megadeth disc ids just added with incorrect last tracks

I just asked for some additional info in the edit above

xlotlu: http://pastebin.com/m3b4d2065

just a bit of RDF about the requiem, just to show how that'd be described in mo

yveser adopting something like musicontology seems like it woudl go hand in hand with introducing new entities to mb

like works/revisions etc.

and linking tracks as instances of works

*sorry probably not usting mo's actual terminology

yep, from what i heard abotu the new schema, it looks like the mapping mbz/mo will be much easier then

is that due to come in any time soon?

hmm, i really don't know

last i heard, it was a 'who knows when thing'

but that was probably two years ago

but i guess the people who were at the summit in london should know more about that?

and i havent' been arounds ince :p

*since

Well, it won't happen overnight in one big go :P

lol :)

But afaik, the BBC and luks are making a start on it

oh right, i forgot BBC got involved

The BBC with release related schema changes, and luks with track merging, but I may be wrong

apparently, the two big new things are "work" and "package"

and there's some last.fm involvement with mb i heard about

gafuuu, we only import their data, that's it for now

(We = Last.fm)

their = musicbrainz's :p

you're with last.fm?

Yup, sitting in the office right now

heh

office == ballpit

Muz: say hello to elias and mark :)

I can't honestly say whether I know or not

heh

gafuuu: i wrote a rdf wrapper around last.fm api, some time ago

what I do know is that we are working towards improving our catalogue structure and it appears to be going in the same vague direction as Music-ontology describes

yvesr what would that do?

but i'd so much like last.fm to provide rdf (or RDFa) at the source :-)

gafuuu: dbtune.org/last-fm/

Muz: as you already have quite a few microformats on last.fm, are you planning on putting some hAudio also?

At some point or another, yeah :)

It's not particularly high on the long "to do" list though :(

MB has dropped use of RDF, that preceded the Xml webservice

but you can extend their relaxng structure

ijabz: indeed

ijabz: i think that was quite a good decision

outsidecontext:Agreed

ijabz: i maintain a more "scalable" rdf mapping of musicbrainz at dbtune.org/musicbrainz

ijabz: the good thing is the links you can draw to equivalent resources on further datasets

eg. http://dbtune.org/musicbrainz/page/artist/c9673...

or http://dbtune.org/musicbrainz/page/artist/c3aeb...

yvesr: the mo:musicbrainz id seems to be wrong for albums

e.g. on http://dbtune.org/musicbrainz/page/record/0719d...

oh!

artist instead of album

How are you creating these pages, are you build them from a copy of the database once a day, or are these queries built on the fly

i like the fact you are exposing the data quality :)

outsidecontext: should be fixed now - sorry about that

outsidecontext: still in the middle of the mapping... it's kind of a tough job, so it takes some time :)

that was quick :)

ijabz: i am working on a db dump atm

So out of interest could you do the same thing my extending the relaxng schema, or do you need rdf to do this kind of thing

i do need rdf, of course :)

rdf as in the data-model, not as in a particular syntax for it

i think it should be possible with an extended schema, in the end the rdf does nothing more than consolidate different data sources#

Oh , so rdf just links things togther

ijabz: yes, exactly

can you *please* stop writing "mo"

:(

lol, sorry mo

thanks

name conflict

:-D

novice question:Are 12" singles always marked as E.Ps in Musicbrainz or not

I mean type=E.P

A single is a single.

%)

The release media can indicate what format is was released on

I mean the release type field lets you specify Single, E.P album ... If I have a 7" and a 12" version of a release do I have to put them bothas release type Single if the 12" has nothing on it stating that its actually an E.P

muz:by release media do you mean format it just lets you specify vinyl, CD extera

Yeah

adamgolding has joined #Musicbrainz

gafuuu has joined #Musicbrainz

hey all, i wanted to know if anyone had a working libofa on a PPC mac and how they went about it ... is there a guide somewhere? i'd like to generate PUIDs for my music

er, to be more specific, PPC OS X 10.4.9

adamgolding has joined #Musicbrainz