#musicbrainz-devel

aCiD2: artist i18n

Right, so I didn't get time to sit down and code - but I'm know you're eager to update replication

we have 2 options - wait until I can do it all, or just get the schema change in now and start replication

it's up to you how soon you need this though I think

I'm for getting the schema change in so we can continue to test replication.

I have enough other things on my plate that I'm not graving for that replication just yet.

we originally said that the functionality doesn't need to be exposed yet.

s/graving/craving/

I just wait to avoid doing a schema cahnge for a while.

well I think it would be done this week

but a schema change I might even be able to get done tonight

aCiD2: ok, then bust out the schema change stuff when you can and we'll see about getting replication going next week.

aCiD2: even better. :)

so you want that schema change asap then?

don't bust ass for that.

ok

its not that important. but in the next few days would be awesome. :)

warp: last item for today: script detection

yay

alright, if I haven't done the whole patch by thursday, i'll just up the schema change

ruaok has changed the topic to: agenda: script detecting

aCiD2: thx

when trying to detect the script, most common scripts can be matched easily. with Han characters we will need to offer a choice between a few options.

what is the context for script detection?

ruaok: MBS-416, people want a better way to select scripts compared to the current musicbrainz server.

got it.

ok, what is the question at hand then?

ruaok: but, when using that matching... there are about 40 scripts left which can currently be selected as a script on the server, but which will never be matched automatically

I think some of them may not even have any character in unicode (at this time)

(we currently offer the full set of ISO_15924 codes)

I'm wondering what to do with those

but we have releases in those scripts?

warp: is this on the main server or on test?

pronik_tp: I haven't looked, that's something I probably should have done before the meeting :)

can we make it so that your guess script offers the guessed choices and "None of these, show full list" ?

warp: test isn't hiding the scripts which aren't even in unicode (iirc)

nikki: I haven't look at what http://test.musicbrainz.org currently has.

nikki: when you say "not even in Unicode", are you including ConScript scripts in that?

ruaok: we can make it anything you want :)

even it what I want doesn't make sense? :)

brianfreud_: klingon isn't in unicode

lol

smack down!

exactly, but it is in ConScript

and in our current mason scripts list :P

some people agreed on some private use characters so they could share some fonts, but there's nothing which says those private use characters are klingon and only klingon

warp: what exactly is the issue we're trying to decide then?

ruaok: well, if they can't even be represented in unicode. I was thinking, perhaps we could just leave them out. and only put them back in when people start complaining.

ruaok: for most, it's unlikely that we currently have releases in them... and that there will ever be releases in them.

(this is ofcourse the IX answer... prune the stuff ppl aren't using)

warp: http://www.evertype.com/standards/csur/ is what I'm talking about, and iirc, we do have at least one release in the db which uses Klingon

if it can't be represented in Unicode, is it fair to say that it cant be in MB?

ruaok: except it is in MB :)

ruaok: no, we have some releases in the DB with tracklistings which cannot be represented.

luks: HTH does that work?

Unicode Private Use Area (E000-F8FF and 000F0000-0010FFFF)

is it "romanized"?

ie, ConScript ;)

brianfreud_: the current list of scripts I cannot detect is http://git.musicbrainz.org/gitweb/?p=musicbrain...

it's still a private use area, I don't see how it's any different from having a font which uses klingon glyphs for a-z and saying a latin release is in klingon script

feh.

because any release in Klingon, using a-z, is transliterated; a-z excludes Klingon CH, GH, NG, QH, TLH, and punctuation

ruaok: anyway, so, from an interaction design perspective, I would prefer to just not include any scripts which will likely go unused. but if that's not an option we can just make them fairly well-hidden :)

brianfreud_: you can still pick characters to assign the glyph to

why not do the same as on the live server?

a 'Show all' button

warp: make it so that people *can* get to them, but hide them well to not confuse noobs.

or even a tiny link

ruaok: good enough for me. next topic :)

How about "Show all", and when you click it, a "Show Klingon" second link pops up :)

next topic? LUNCH!

</BANG>

MBChatLogger has changed the topic to: http://musicbrainz.org/#devel

thanks for your time everyone!

thank you!

honot and nonoes, warp!

you guys going to be apply for SoC next week?

hellz yes.

what, nonoes too? awww maaaan

cool

when are the ideas pages due?

when we apply, right?

yeah. the deadline to apply is the 12th

registration opens on the 8th

easy. :)

cool

what are you bribing them to do? accept you as a mentoring org again this year?

I always have choco for the team at the summit

i see :-)

I'm heading out in a sec, but before I go, quick Q - anyone able to point me to, or pastebin for me, the current static list of all ARs that are live on the mason server?

haha

brianfreud_: send me a mail about that. I can do that later this afternoon

can you see http://musicbrainz.org/edit/relationships/link_... ?

ruaok: if you have a bit of your time, look at the MBH-6

ruaok: you needed to talk to me and aCiD2 a bit?

nikki: Yeah I think that's open for anyone to view.

thanks; doing an audit of everything AR-related in the wiki, and I want to check for any ARs I've missed which never actually get a Relationship Type page

pronik_tp: that ticket is more djce, not I.

brianfreud_: your ibdb/iobdb ones are still waiting for pages :P

I wasn't invoved in the setup of the git stuffs.

nope, IBDb exists, just IOBDb needs one :)

warp, aCiD2: I can make this short.

we need to make sure all the tickets are estimated tomorrow.

I would very much like to make the schedule and announce some dates.

all, 100% all?

ok, some new ones have popped which I haven't estimated.

all NGS.

ok

and there may be a couple more that I have on my list too.

and, I have received no timesheets from either of you.

Cover Art Relationship Type is needed too

thats all.

ruaok: how detailed do you want that

?

try me something that you think is reasonable.

I'll give it a look see and tell if you if I want more/less info for next month.

that reasonable?

the data I have is per (ticket + description), or just the description for everything which isn't a jira ticket.

if you want I can give you just that, or aggregrate on jira tickets.

the former should be fine.

I'll get that to you tomorrow.

ok

aCiD2: do you have time to continue helping me with git?

sure

hm, now I get

6 files changed, 32 insertions(+), 2 deletions(-)

isn't that what you want?

oh, had three dots instead of two.

7 files changed, 33 insertions(+), 5 deletions(-)

I still think it's easier to just do the merge, check the results in a visualizer, and if it's wrong then just "git reset --hard HEAD^" and revert the commit :)

aCiD2: that's the different with master

merge commit that is

aCiD2: if I try to merge into next, a whole bunch of other files change

have you looked at the visualization of the merge yet? gitg, gitk and giggle will all give you a clear idea of exactly what got merged where and why certain things happened

I started gitk now.

I have no experience with no-commit either, I'd just go balls out and commit it. if you break it, git makes it trivial to rollback

17 files changed, 126 insertions(+), 319 deletions(-)

that's what git merge did.

doing "git checkout next; git merge --no-ff --log my-branch" right?

yes

the --log will at least let you inspect the commit and see what commits it merged in too