#musicbrainz

wheels: no worries. Yes, the people who license this have gotten licenses for it from the labels.

otherwise amazon and co would've been in court with the labels a long time ago.

eaon: heh, its sooo much better than the first two iterations of the RDF.

this version actually does make sense. I think its RDF that sucks my will to live.

well thats an opinion

querying with RDF/XML is fundamentally wrong

also stating the status of the returned data

<mq:status>OK</mq:status>

^^^ a literal? totally wrong too.

but nevermind

thanks for volunteering to fix it up.

It would seem that all cover art being copyrighted that it would require licenses for redistribution, scanned or otherwise. Then this gets tricky when it goes up on the web which implicitly means lots of copies of the stuff flying around (various caches -- search engine, web browsers, proxies, etc.) and of course that's apparently acceptable. It's tricky figuring out where the line is on what the limits of such are.

send patches to patches [at musicbrainz.org]

and thanks for speaking so kindly about my work!

eaon: are you involved in the scrobbler stuff generally?

RDF is my work ;)

wheels: well i'm working with the guy that brings more rdf to them

did you actually write part of the RDF spec?

no

i didn't say that :)

most people misunderstood RDF, thats the biggest problem the community has.

wheels: you're right on all that stuff. its is a mess and quite frankly I am surprised that amazon is allowed to get away with the stuff they are doing.

eaon: yup. with suck a CRAPPY specification, its hard to understand it. Period.

amazon is "big", everyone likes them

ruaok: i disagree. it was very important that not too many people get onto it yet

er?

ruaok: otherwise it'd have ended up as something very wrong - there are already enough compromises (RDF/XML - the incarnation of the devil)

-d

using RDF as part of the web service is probably one of the biggest wastes of time for MusicBrainz.

fine with me :) but then don't critisize me for critisizing musicbrainz metadata ;)

yeah, understandable

at the current state, i mean

ruaok: For the stuff that I'm working on these days people keep bringing RDF back up...

RDF is not there yet, application-wise

it's good to experiment, but nothing to rely on yet

holy shit: http://www.musicbrainz.org/mrtg/zim.cgi?log=zim...

that's busy

has no-one got anything better to do on a sunday? :-)

sundays suck donkey balls :)

eaon: Honestly I don't know it all that well, but people kept suggesting it as a metadata transformation layer for some of the stuff that I'm working on and with a quick glance it seemed (a) overly complicated and (b) a bit rigid structurally.

wheels: (a) it's not overly complicated at all *if* you understand it (b) it's not rigid at all, rdf was built to be more flexible than most people need it :)

Also note the visibly increased rate of TRM growth: http://www.musicbrainz.org/mrtg/bender.cgi?log=...

ok that's enough graphs for now.

djce: the referrer log shows that someone in .fr pointed to MB.

eaon: Well, it seems like it's "explicitly flexible" rather than just being somewhat structurally dynamic.

wheels: structurally dynamic?

We've had an influx of donations in the last 24 hours and a request to translate the service/tagger to french.

eaon: Again, I don't understand it, so I'm mostly just going on first impressions.

wheels: your first impressions are right on, IMHO.

ruaok: I saw that request. Well spotted. /me goes to check the access logs

wheels: i had the same impressions, hehe. i disliked and disagreed with the whole thing at first :)

eaon: i.e. the stuff that I'm working on we moved away from explicit typing of metadata and such and also categories in favor of a value / property system where everything was a value or property and you built whatever structures you needed dynamically from those.

wheels: i'm not sure i understand, but is it similiar to apples .plist?

No idea. :-)

And I'm doing a bad job explaining out of context anyway. ;-)

hehe

but we're getting way offtopic ;)

eaon: Well, I'm not against having a Mac; like I'd use one if someone bought me one, but... ;-)

hehe

it's okay, i rant a bit too

46 minutes for a full mb database import. kick ass!

a lot of stuff is too extreme

you know, i'm not for a new mac os <X ;)

eaon: I call it "usability in a box", because it's mostly usability for usability experts. They wrote a HIG that's so long that no developer will ever read it and sit around being impressed with themselves. ;-)

wheels: oh, i'm one of the guys that calls these developers fools :)

:-)

eaon: See, to me a HIG is only useful in as far as it affects real software. Like a 15 page HIG is probably more effective than a 400 page one.

At least in OSS -- because that way the developers are likely to actually read it.

wheels: thats where i think FOSS development is flawed

you have to do it in a company, if the company is serious about their stuff

why not do it as an OSS developer?

if there are mistakes... thats no problem, but if someone completly ignores it, why use his app?

eaon: I don't think that's really true in most companies either. There are a few who do so. Most don't.

in a package like KDE

wheels: as i said, a company who is serious about their software

;-)

and i think FOSS should get more serious, at least with their "in house, for daily and heavy use" software

large companies aren't serious, they're bureaucratic :)

even apple, see the lastest iTunes ... photo syncing with a music app! yay! ;)

eaon: I think regardless of the environment -- if you want to teach developers how to work with "micro-usability" (nuts and bolts issues) you have to write a guide for them. Most don't care about the "why" -- just tell them what they need to know and they'll probably do things that way.

yep

thats where the HIG is flawed :)

eaon: It's kind of like if I wanted to learn about the history of Kenya and someone gave me an 80 volume set of books and a 200 page reader I'd probably get a lot further in the latter. :-)

sorry i'm just seeing flaws everywhere ;)

anyway, i got the stuff that i need

eaon: http://mayhem-chaos.net/photoblog/images/perfec...

thanks and bye

ruaok: and i'll work on an mb ontology for you :)

to interrupt - as things currently are, you have to assign an album to various artists even if every track but 1 or 2 is by the same artist?

:-)

DJKC: yes

hmm. lot's of little annoying bits like that

ruaok: nice image :D

see you guys

pretty much every album on the sharpnel sounds record label is by DJ Sharpnel but they all have a track or three by another artist on them. makes entering them annoying

ruaok: My favorite one lately: http://nat.org/choto.php3?p=3639 (from one of the GNOME guys, naturally)

LOL.

you guys have a friendly relationship?

Actually I don't have any contact with him, so I presume that wasn't actually related to me.

oh. I assumed it was a friendly jab.

And I know most of the GNOME guys at Redhat...

But one of the GNOME guys did point me to that, so who gnows. ;-)

are you planning to make picard translatable?

cikkolata: Right after German becomes the default. :-)

you're all determined to make me get around to learning german again, I swear.

cikkolata: no plans yet.

I'm hoping to get picard to a usable state and then get others exicited about working on it.

ahh

but once it becomes cool it will excite me again too. :)

Ah, how knows.