#musicbrainz

I do :p

Pray tell :-)

Go to http://beta.musicbrainz.org/artist/0c0e9800-155...

Press Edit

????

PROFIT

rswarbrick: yes... though that's kind of the thought

brianfreud: Oh? But doesn't that mean that after the merge when someone searches for Dubin they get Jimmy McHugh as well?

(I'm coming at this after getting random conductors when searching for composers' names...)

MEH - recorded in concert hall N on 9 Jan, 5 Apr, 16 May 2006

One track?!

sure, but in a case where they're searching for the artist, and they're searching by lyricist's name, wouldn't they want the composer's name anyhow?

An album

God, they were rubbish

Ah :-)

There might be more specific dates

But in the booklet

(and Naxos are nice enough giving the back cover away :p )

9 Jan to 16 May, or just 2006?

for the relationship

brianfreud: What I mean is that sometimes you search for eg. "Mozart" in the "relate to" box and get the Vienna Phil (or whatever). This is confusing to me...

yeah, those are pretty useless ones

reosarevok: Since they're that far apart, I think I'd just use the year.

that was kind of my thought re http://tickets.musicbrainz.org/browse/MBS-4016 , though there I was specific to works

I really dislike those relations that say "this track was recorded from January to November in 1990" or whatever

brianfreud: Ah, I like that idea!

Sounds relatively sound, yeah

rswarbrick: why?

Only annoyance is the results popup being even bigger

But still

gioele: Which bit are you asking about

?

rswarbrick: tracks recorded from Jan to Nov 1990

rswarbrick: that is the result of moving many ARs from releases to tracks

reosarevok: yeah, a bit bigger... but I think far more friendly

(or at least, more comprehensible)

Well, frankly the information isn't any better than just saying 1990. And it looks bizarre.

it makes sense for a release to be recorded from Jan to Nov, much less for tracks, but still...

Precisely. What one is really saying is "I know this track was recorded sometime between the following two dates"

Clearly the answer in this case is asking Naxos to give the exact dates :p

But we don't really have a way to express that: most other timed relationships on MB show the period of time over which something happened

reosarevok: I just released a new update, can you check if it works correctly in Chrome? http://userscripts.org/scripts/show/121723

reosarevok: Well, of course :-) (They might inside the actual booklet)

Probably :(

But it's weird

Usually they give a range

But in the previous one I did they gave the exact dates for every performance

I presume it's just a different editor on the CD materials.

Or shorter titles and more space, maybe :-)

Heh

There's not too much space in these

Almost all are partially paid with Spanish public money

And that implies logos

kovacsur: re dates, in batch, maybe you're right to leave them out

Heh :-)

brianfreud: In http://musicbrainz.org/edit/16027388, should we be killing the work as well? Or is it supposed to survive?

brianfreud: yeah, I think so, too. Maybe if it also worked on search results pages it would make more sense, but not from artists' works lists

rswarbrick, orphaned works are autoremoved

Ah, I thought it still had a composed relation. Ignore me :-)

I had actually been thinking more along fields per work, rather than globally

maybe that'd be too noisy

Dunno if *too* but quite

And tiresome

I ended up adding the hide ability to the tagger for just that reason D:

:D

I'd rather have another script to update dates like mad

While I'm pestering brianfreud... (sorry!) should the recording & track list responsible for http://musicbrainz.org/edit/16027262 also get changed? Or was this genuinely misnamed on the CD or something?

reosarevok: Working on it :-) (but it won't be a userscript: I reckon hunks of CL code are more usable by me. Sorry!)

it'd be so nice to have a script to do that... date support came too late for Dylan's performance of work ARs

CL = Common Lisp?

Yep. That said, I think I might not make non-browser-based code that makes edits public: probably lots of potential for havoc...

It was convenient for tagging lots of recordings with works (more than fitted on the recordings page) though.

how many people do you think know what to do with lisp code anyway? :P

Ah, well. :-)

(and we already have public python bots)

I'll happily stick it up on github if anyone would find it useful. But I wasn't going to work on frontends. I think it's better as a "How can I solve the following problem for the recordings in front of me" sort of thing.

I S2 bots

And I'm rubbish at writing guis...

Is S2 a verb? (I'm genuinely confused!)

It's a heart

Somehow

The mix of Spanish keyboard

Ah!

*US keyboard

With Spanish layout

Means I have no less than symbol :(

how come?

Heh. Feel free to copy + paste: <>

In a Spanish keyboard, less than is at the left of Z

In this keyboard, there's nothing there

oh right

I've pressed lots of other random keys and can't find it :p

What's to the right of M?

(with shift)

; : _

Fair enough.

(as in a normal Spanish keyboard)

I can see that the US layout has it there

That's because Americans find all these buttons confusing...

(it's printed)

But...

Ah, I see.

rswarbrick: well surely if they're that confusing, just one less is still far too many :P

try the backtick

left of the 0

One solution is to switch layouts again! (Typing this on Dvorak)

�!"�$%&/()=?� is my upper line

Spanish uses more stuff than English (that's why we need more keys! :) )

I was going to say: calculations must be hard :-)

Don't you guys use AltGr heavily as well?

the germans do

I'm still shuddering after teaching a German how to write C.

and the hungarians :p

haha

The {} are on AltGr somewhere on the top right.

My qwerty touch typing is rusty as it is. On a German kezboard, it's even more fiddlz.

\|@#�}][{

Are our altGrs

So, mostly for @

And for piping sometimes

(and we don't have tilde)

you need altgr + 7 to 9 for { [ ] and }