#musicbrainz

Above all else, I think we need more foreign language native speakers

I hope we will get something like data quality but only for titles

foolip’s working on Chinese releases, you’re working on Italian, mo has Norwegian, etc - but there’s some, like Portugese, Japanese, and Russian, where I don’t think anyone’s really checking on them

brianfreud: yes, Icelandic is seriously lacking behind ;)

;)

what about German?

luks, etc

luks? german?

ruaok too

wow

luks is Chech Republic (however you spell it) I think

ruaok is originally from Germany though

brianfreud: luks is from slovakia.

not czech rep.

oops

wrong half of the former country :)

brianfreud: no worries, they had a friendly separation

ich bin Deutsch, das hast Du recht. :-)

(IIRC)

ruaok: Something we’d talked about a while back...

Given the relative lack of Japense releases, and the serious lack of, say, Russian, Korean, or Arabic ones,

Zomg Deutsch.

what about Arabian and Indian releases?

what would you think of eventually supporting localization on the site?

We counted - there were less than 1k total releases for all Arabic languages

brianfreud: its something I would love...

but the amount of effort to get this worked out is just insane.

My guess is that the site being indecipherable to people who speak Korean is the major hurdle

that's why we have warp! :-)

lol - he can fix, but alone he cannot enter them all :)

luks had shown us a site he uses that’s working on PicardQT translations

seriously, I hear ya. its an issue for sure.

I know actual implementation is a long way away, but would it be possible to extract the text strings, so people can start on translating?

I would say that once we can get NGS rolled out and smoothed out, we should consider doing just that.

Something we’d talked about while you were at )^(

I need Korean contributors for Italian prog rock of the 70s (don't ask me why they still love and re-release that music)

not feasabile until after NGS.

sooo much will change with NGS, you'd be wasting time.

hmmm

ruaok: how far (or near) is NGS?

Now that you have a gameplan for NGS, is there any kind of timetable?

LOL, jinx

nothing firm yet.

but I am personally hoping to see the first ngs related release around the end of the year.

Do you have any plans for which parts will happen first?

release changes, artist changes, track changes

year = 2007 or 2008?

2007

I know, for me personally, the two things I most am looking forward to at 1) no more & artists, and (less than #1) track masters

GSoC had made some progress towards the release portion of ngs.

When does the “summer” end for GSoC?

so we need to finish that and get it out.

its done.

What did we get from it, when all the coding was finished?

the mentor summit is next saturday where a few mentors will get together and chat about how things went.

no more artists at all would be even better (remember the ml thread?)

well, I have a number of things to catch up on with Luks, and that was one of them.

gioele: I would love that :)

in summary, I think luks and I were pretty disappointed with the results of gsoc.

ruaok: something we’ve joked about, but in semi-seriousness... The idea of a “rotating” “artist” concept, rather than trying to squeeze composer/performer/etc into “artist”

not at all what we'd expected.

yeah - seemed to start strong, but then the participants dropped off the map, and their webpages for updates stopped updating...

ding.

you know, there was this problem in the last months called "Summer", I think that distracted many GSoC students :)

Yes, but when you make the commitment, and get the $$ from Google, you are volunteering to give up “summer”

MBChatLogger: off

is not logging

is logging

thanks gioele

rant was clearly over.

np

just enough to make the few plugins I’ve made for picard

One problem I’ve found is, when last I was seriously coding, the choices were easy - C, Eiffel, or Assembly

language doesn't matter. its merely a dialect.

Now you have python, php, perl, c++, java, js, etc - almost overwhelming

the key is to get comfortable with *programming*.

then moving a dialect is merely lots of flipping in reference books.

yeah, if you know C well, you can learn anything you mentioned in a week

at least well enough to get stuff done

figuring out the code in any is easy - assembly+c makes for a good basis to figure code out in any language. But figuring which languages to become able to write, and write well, that’s different

pretty much.

brianfreud: bs.

you think?

learn python. get comfortable. learn C. then the road will be clear.

clear because as ojnkpjg says, going to something else is little work.

and the language depends on the work you want to do.

See, I still think in Stroustep’s type of thinking - any language you can pick up in a week. But being *good*, really *good* in any language takes 1 to 2 years of serious coding in that language...

for das blinkenwheel, the python global interpeter lock fucked me.

I had to port my code to C++.

takes a year to learn the idioms, maybe

that sucked, but took only a few hours.

i think with something like python, you could learn enough to write good code in a week or two

not the idioms, but more learning where the traps are, and how to not just write code, but efficient code

but it was interesting to visually see the effects of the global interpreter lock.

i'm having to maintain some C++ code now :/

brianfreud: easy with python. there is the "python way"

that uses "smart" pointers

actually, from the little python I’ve seen/worked on, it strikes me as very much like, say, js

if you stick with that you can write some pretty efficient python.

or, and I have to admit, the only language I’ve really been paid to write in in the past 10 years - VBA :(

js is not nearly as elegant.

yes - and js has the problem of having to deal with multiple interpreters (I want to say “compilers”, but...) for the same code all at the same time... leads to nasty workaround-code styles

javascript is really hackish

well, it has some serious benefits

yeah, i just mean in terms of being a clean language

it's useful for a lot of stuff

true

I remember when java first came out, the entire CS department at hopkins - grad and undergrad alike - signed up to take it, as it was “the next big thing”

when i was in college, everything had already gone to java

the *idea* was - code you write once, that can run on any machine/os/etc

so i resolved to make it through without learning java

and i succeeded

i still have not written a line of java in my life

but the implementation of java was bs then (beta 0.7), and still is

\ö/

ugh - last thing I wrote in college, I had to write photoshop in java

damn thing took 4 hours to initialize on a Solaris box

haha

back to reality: installing mb does not seem that hard. http://bugs.musicbrainz.org/browser/mb_server/t...

its not.

most people's pussies just hurt when it comes to installing shit. :)

I've got new installs down pat.

ruaok: For whenever my drives come back from the mfr, I plan to switch to Linux, or at least try to, and install the server

I'd be glad to help.

I don’t want to use FAT - is there any fs that Windows can write to, but still be linux compatable, so I can migrate 1 tb or so of data over?

brianfreud: a dedicated partition with debian and mb should be quite easy to setup (it seems)

I've figured out the crititcal path of installing mb_server on a new machine.

1. download data 2. install perl modules 3. install postgres

start those three in parallel

brianfreud, you can mount ntfs drives in linux

brianfreud: you just need a drive?

what kind?

there's an ext3 driver for windows, too

brianfreud: I use ext3 on my external HDs and explore2fs (an user-space application) in windows

everything I have right now is ntfs - I keep hearing nasty/difficult things about ntfs in linux...

it's much better now with the ntfs stuff that uses FUSE, believe it or not

ruaok: IDE - Have a couple of drives on rma now, had to dump a lot of data to blank dvds

is ntfs writable though?

yep

I’d hate to move to linux, install an ntfs driver, then find I have limits on what I can even write to those partitions

it works so much better than i ever imagined it would

brianfreud: want a 120gb IDE drive?

serious?

I have no machine left that can use it.