IRC log of musicbrainz-devel on 2013-11-18

Timestamps are in UTC.

00:10:45 [JonnyJD]
JonnyJD has joined #musicbrainz-devel
01:13:15 [JesseW_not_logge]
JesseW_not_logge has joined #musicbrainz-devel
01:46:36 [jseet]
jseet has joined #musicbrainz-devel
01:47:13 [jseet]
jseet has left #musicbrainz-devel
01:47:30 [jseet]
jseet has joined #musicbrainz-devel
01:50:46 [jseet]
Hi - I'm looking to install an MB server and I'm following the instructions at https://github.com/metabrainz/musicbrainz-server/blob/master/INSTALL.md. The Perl requirement says 5.10.1 but when installing the Perl dependencies, cpanm says Sys::Hostname is out of range and needs to be version 1.16...which appears to be linked to Perl 5.18.1.
02:20:35 [DWSR2]
DWSR2 has joined #musicbrainz-devel
02:20:50 [DWSR2]
If I have the vm image, how can I dump everything and import a fresh snapshot?
02:37:55 [JonnyJD]
JonnyJD has joined #musicbrainz-devel
02:39:11 [ianmcorvidae]
DWSR2: use the postgres dropdb command, ensure you have enough disk space (roughly 2x the complete DB's size, since you need the uncompressed data plus the finished DB by the end before temp file cleanup), and otherwise follow the main installation instructions re: importing the DB.
02:41:56 [DWSR2]
ianmcorvidae: Ah, wasn't sure if there might be some additional steps due to the way it's installed on the vm.
02:42:23 [DWSR2]
ianmcorvidae: If I'm doing a full reimport, should I bother doing a schema upgrade or no?
02:42:55 [ianmcorvidae]
well, you'll want a codebase that's matched to whatever you're importing, but I wouldn't bother running the upgrade on the DB you're going to drop, no :)
02:43:01 [DWSR2]
Yeah, figured as much.
02:43:17 [DWSR2]
ianmcorvidae: Can I just keep checking out the latest branch?
02:44:20 [ianmcorvidae]
assuming you mean latest tag, but yes
02:45:53 [DWSR2]
sorry, yes.
07:00:52 [andreypopp]
andreypopp has joined #musicbrainz-devel
08:05:17 [zas]
Mineo: ping
08:17:38 [kepstin-laptop]
kepstin-laptop has joined #musicbrainz-devel
08:59:31 [andreypopp]
andreypopp has joined #musicbrainz-devel
09:27:06 [ruaok]
ruaok has joined #musicbrainz-devel
09:27:06 [ruaok]
ruaok has joined #musicbrainz-devel
09:43:49 [andreypopp]
andreypopp has joined #musicbrainz-devel
09:52:35 [marcooliveira]
marcooliveira has joined #musicbrainz-devel
10:20:00 [ocharles]
ocharles has joined #musicbrainz-devel
10:20:04 [ocharles]
g'morning
10:20:17 [nikki]
moin
10:27:25 [SultS]
SultS has joined #musicbrainz-devel
10:27:25 [SultS]
SultS has joined #musicbrainz-devel
10:34:29 [jesus2099]
jesus2099 has joined #musicbrainz-devel
10:35:07 [jesus2099]
and then i suddenly understood something… it rarely happens to have DiscID release with mistake track order from the editor but when it’s the case i guess having attached TOC prevents from reordering tracks isn’t it ? http://tickets.musicbrainz.org/browse/MBS-6981
10:35:33 [jesus2099]
weren’t there a text saying that in the release editor before ?
10:36:09 [jesus2099]
it’s like i seem to mrememebnt there were a message like thies… maybe the reorder track arrows should be invisible when attached TOC (as it forbids the reorder anyway) + message, no ?
10:38:38 [jesus2099]
i changed the ticket message in this way…
10:45:08 [nikki]
it would be more consistent to make the arrows behave like the track time fields (i.e. visible, but disabled with a tooltip)
10:49:06 [nikki]
ocharles: oh, what do you want me to do about issues with the work attributes stuff?
10:50:10 [ocharles]
nikki: tickets would be fine
10:51:57 [nikki]
and speaking of which...
10:52:00 [nikki]
alastairp: ping
10:58:39 [andreypopp]
andreypopp has joined #musicbrainz-devel
10:59:23 [ocharles]
nikki: btw, aware of ISEs on /edit/19227832 on beta at the moment
10:59:32 [ocharles]
something seems borked there so I'm trying to fix it
10:59:46 [nikki]
ok
11:07:27 [nikki]
anyway http://tickets.musicbrainz.org/browse/MBS-6989 and http://tickets.musicbrainz.org/browse/MBS-6991
11:09:04 [ruaok]
ruaok has joined #musicbrainz-devel
11:14:53 [ruaok]
feh. rainy barcelona is rainy.
11:23:51 [nikki]
ocharles: and also http://tickets.musicbrainz.org/browse/MBS-6992 since iirc you said it's not right now
11:24:12 [nikki]
moin ruaok
11:24:16 [ruaok]
moin!
11:25:21 [ocharles]
nikki: thanks
11:27:18 [nikki]
ruaok: I was wondering earlier if it'd be ok to add commons to our score whitelist (right now we only have imslp)
11:27:50 [ruaok]
yeah, makes total sense.
11:28:27 [nikki]
heh, so I guess I just have to get round to making the proposal then
12:08:55 [mstnorris]
mstnorris has joined #musicbrainz-devel
12:09:08 [mstnorris]
: Does anyone have an up-to-date schema for the database?
12:09:17 [mstnorris]
I have just set up the Database on my own server and I've tried using schemaSpy but it doesn't pick up any relationships
12:09:35 [ruaok]
mstnorris: because the relationships do not use FKs.
12:09:49 [ruaok]
look at the l_<entity>_<entity> tables
12:11:45 [mstnorris]
Is there a listing somewhere for standard simple queries to pull lets say all albums from an artist and their associated tracks?
12:11:52 [mstnorris]
I'm sort of flying blind at the moment
12:12:25 [ruaok]
what is your goal mstnorris?
12:12:41 [ruaok]
we recommend that people use or WS to work with MB.
12:12:45 [ocharles]
well, the relationships do use FKs, it's more that the specific relationship types in MusicBrainz are tucked away in an entity-attribute-value type schema
12:12:56 [ocharles]
that too :)
12:13:13 [nikki]
it doesn't look like mstnorris is looking for relationships though
12:13:26 [mstnorris]
I am using the MB database for my final year university project to create a music tagging system
12:13:47 [ruaok]
then use our WS and save yourself a load of hassle.
12:13:50 [mstnorris]
The MB database will be untouched, but I need a way to reference whether the artist, album/track is being tagged
12:14:07 [ruaok]
thats what our MBIDs are for.
12:15:49 [mstnorris]
i've already set up the database and I was hoping to use that as it was a learning curve doing it that way.
12:16:08 [mstnorris]
Your WS uses those queries
12:16:16 [ruaok]
do you want a learning curve or do you want to finish your project?
12:16:20 [ruaok]
depends on what your goal is?
12:16:44 [mstnorris]
Well, the project will never be finished, but the goal is to learn as much as I can.
12:17:15 [ruaok]
if finishing isn't the point, then have at it. :)
12:17:27 [mstnorris]
I would like to get a working prototype of my idea, but it certainly doesn't have to be finished
12:17:40 [ruaok]
http://musicbrainz.org/doc/MusicBrainz_Database/Schema
12:19:17 [mstnorris]
Are the diagrams up to date?
12:19:29 [ruaok]
* ruaok isn't sure.
12:19:37 [nikki]
yes
12:19:43 [mstnorris]
Thanks
12:20:09 [nikki]
also if you follow the link to the wiki at the bottom of the page, you can click on the image there to get a bigger version
12:20:39 [nikki]
(unless you like squinting :P)
12:27:57 [mstnorris]
Much better :)
12:40:15 [ruaok]
4,249,916 edit in 2012. we had 1/4 of our total edit count in one year. really?
12:40:39 [ruaok]
https://gist.github.com/mayhem/7527149
12:40:42 [ruaok]
sanity check, please.
12:45:04 [nikki]
sounds about right to me
12:45:17 [ruaok]
thats insane.
12:45:54 [nikki]
well, the tools for editing make it easier than ever to make loads of edits and we have more bots than ever too
12:46:43 [ruaok]
I'll query next how many edits we already have for this year.
12:47:38 [nikki]
https://beta.musicbrainz.org/statistics/timeline/main#+c-edit-information+-v-+-editor.votelastweek+-editor.activelastweek+-editor.editlastweek+-vote.perweek+-edit.open+-c-core-entities heh, you can see when we started running bots :P
12:48:37 [ruaok]
holy shit, our edit count for 2012 has got to be intense.
12:48:41 [ruaok]
er 2013.
12:49:03 [ruaok]
hmm. total number of editors is down from 2011.
12:49:29 [ruaok]
20540 as compared to 23637 from 2012.
12:49:43 [nikki]
what about excluding puid edits?
12:50:00 [nikki]
* nikki notes she hasn't told ian to add that statistic yet
12:50:19 [ruaok]
that would be a good one, true.
12:51:47 [nikki]
looking at http://tickets.musicbrainz.org/browse/MBS-6624 it seems the edit types are 77 and 113, if that helps
12:52:20 [ruaok]
I won't bother for now. I need to finish this report and then finally do october finances.
12:52:25 [nikki]
k
13:00:53 [ruaok]
16976 unique editors so far in 2013. the number is dropping. hmmm.
13:03:19 [ocharles]
you mean editors who made edits, I guess?
13:03:55 [ruaok]
yep
13:05:11 [nikki]
feh wrong db
13:05:15 [nikki]
* nikki reruns her query
13:05:31 [nikki]
excluding puids I get 17881 for 2010 and 18756 for 2011
13:05:44 [nikki]
but now I need to wait for this query for 2012 and part of 2013
13:06:01 [navap]
ruaok: You had "5.0 billion hits of which 4.73 billion were web hits" You meant 4.73 being web service right?
13:06:15 [ruaok]
yeah.
13:06:19 [ruaok]
* ruaok goes to fix
13:06:22 [navap]
Okay good, I changed it
13:06:27 [ruaok]
ah, thanks.
13:06:55 [ruaok]
I now just need the words of appreciation and I'm done.
13:08:01 [ruaok]
only 2,247,150 edits in 2011.
13:08:02 [reosarevok]
reosarevok has joined #musicbrainz-devel
13:08:04 [ruaok]
wow.
13:08:57 [ruaok]
nikki: are you doing the PUID adjusted queries?
13:09:00 [nikki]
yes
13:09:04 [nikki]
still waiting for it to finish
13:09:05 [ruaok]
cool.
13:10:10 [reosarevok]
Staaaatss
13:12:22 [nikki]
https://pastee.org/7kycn the dump that was used appears to be from the end of october
13:13:27 [nikki]
so there was a slight drop last year, a bigger one this year
13:13:52 [ruaok]
its clearly time to do some promotion and get some more people editing. :)
13:14:07 [nikki]
(assuming we're not gonna get another 3k people editing in november and december)
13:14:51 [nikki]
hmm, actually, might work
13:15:01 [snartal]
Maybe the db is so clean now it needs less edits :)
13:15:11 [nikki]
nah, people are forever releasing new music
13:16:31 [reosarevok]
Yeah
13:16:33 [navap]
Not to mention the massive amount of music not in the DB
13:16:37 [reosarevok]
Pesky people
13:16:39 [nikki]
18046 is roughly 50 people per day, and november and december are 61 days, and 50*61 is 3050, so I think it's probably going to be similar by the end of the year
13:16:39 [navap]
er, released music
13:16:42 [reosarevok]
Not letting us finish our job
13:16:46 [rvedotrc]
rvedotrc has joined #musicbrainz-devel
13:17:13 [ruaok]
ha. I had someone ask me, if MB was "finished". It seems all the music if there.
13:17:22 [ruaok]
#eyeroll
13:17:22 [ruaok]
*is
13:17:26 [reosarevok]
hahaha
13:17:53 [reosarevok]
Yeah, maybe we should go out of beta, call it done
13:18:03 [reosarevok]
From now on, we'll only patch the content
13:19:33 [navap]
Sounds like a good april 1st thing to do
13:19:55 [nikki]
like we've ever done anything for then
13:20:07 [reosarevok]
We've always considered doing something!
13:20:13 [reosarevok]
That's doing something, in a way :p
13:20:17 [nikki]
considering and doing isn't the same thing
13:20:24 [nikki]
if it were, all my proposals would've passed by now :P
13:20:49 [navap]
Well usually we start discussing april 1 at the end of march.... heh
13:22:27 [nikki]
and how long does it really take to replace the index page with something funny and add a link to bypass it? :P
13:23:20 [reosarevok]
Well it has to go into beta, and...
13:23:21 [reosarevok]
:p
13:25:01 [navap]
:)
13:25:21 [ruaok]
ok, the report is done: http://wiki.musicbrainz.org/MetaBrainz:Annual_Report/2012
13:25:34 [ruaok]
please have a read and ping my on whatever mistakes, typos you might find.
13:25:59 [ruaok]
I'l go fetch some noms now and when I'll get back I'll fix up the mistakes and finally push out the report.
13:26:08 [ruaok]
* ruaok will be really glad to get this off his plate
13:26:37 [nikki]
* nikki wonders why "approximately twice a month" instead of just fortnightly
13:27:02 [ruaok]
because its not a perfect fortnightly schedule...
13:27:38 [nikki]
it's what we're aiming for though
13:27:41 [navap]
And "fortnight" isn't as widely used a term
13:27:55 [nikki]
fortnight is a great word and if people don't know it, it's about time they learnt it! :P
13:28:02 [navap]
heh
13:28:31 [reosarevok]
Oh, so it comes from fourteen nights
13:29:43 [nikki]
anyway "every two weeks" works if you're not convinced about "fortnightly" :P
13:32:25 [Mineo]
I always thought our release schedule depends on people finding new ways to describe heat and cold :P
13:32:50 [nikki]
"we launched the Cover Art Archive, which we'll talk about more later." "in a moment" would be better imo, given that there's only sentence between that and the start of the caa section
13:35:04 [navap]
Or just omit that bit entirely since there's a big header called CAA right after
13:35:13 [nikki]
or not
13:35:14 [nikki]
er
13:35:15 [nikki]
or that
13:38:11 [nikki]
shouldn't the mb: links be using mbdoc:?
13:40:10 [nikki]
also we have editor: for linking to editors
13:40:55 [navap]
Ah, mb: is because that's what I used on my wiki and I forgot to change that to mbdoc: when I moved it over
13:41:09 [nikki]
there's also one link that's just a full url to the doc page
13:41:50 [navap]
There's also an edit link up top :)
13:42:07 [nikki]
but maybe you're editing it and are gonna cause a conflict!
13:43:12 [navap]
I'm not, I'm in a class and already spending too much time getting distracted with irc :)
13:54:01 [Freso]
Freso has joined #musicbrainz-devel
13:57:55 [fractalizator]
fractalizator has joined #musicbrainz-devel
14:11:08 [Freso]
reosarevok: I already have one URLCleanup.js patch pending... should I build the Bandcamp thing on top of that to prevent conflicts?
14:11:42 [reosarevok]
Is it on the exact same lines?
14:12:07 [reosarevok]
In any case, it's quite easy to just merge beta into your repo and update the PR if needed, so personally I'd say make 2
14:12:31 [Freso]
Alrighty then.
14:14:52 [mstnorris]
mstnorris has left #musicbrainz-devel
14:26:18 [ruaok]
* ruaok makes the "in a moment" change
14:26:26 [ruaok]
any other problems that people have found?
14:26:33 [reosarevok]
Oh
14:26:36 [reosarevok]
* reosarevok reads
14:26:49 [Freso]
Problems?
14:26:59 [ruaok]
in the 2012 annual report.
14:27:03 [Freso]
Oh.
14:27:06 [reosarevok]
http://wiki.musicbrainz.org/MetaBrainz:Annual_Report/2012
14:27:21 [reosarevok]
Does "it's Nov 2013" count as a problem?
14:27:24 [reosarevok]
* reosarevok hides
14:27:26 [ruaok]
nikki made one suggestion. only one. which means to my that she probably didn't finish reading. :)
14:27:32 [reosarevok]
hahaha
14:29:11 [Freso]
Oh. That's a decent income from the Amazon associates program.
14:29:34 [Freso]
$.13 from bank credits. \o/
14:31:21 [reosarevok]
"After two years with no significant updates, this is a welcome change!"
14:31:22 [reosarevok]
WORD! :D
14:31:35 [reosarevok]
* reosarevok stops impersonating Caller :)
14:32:37 [ruaok]
oh and congrats on being the top editor, reosarevok. by more than a 100% margin over the next editor.
14:32:40 [reosarevok]
"In July we announced that we're working on a revamp of our edit system" shouldn't it say "we were"? :p
14:32:46 [reosarevok]
hahaha
14:32:53 [navap]
* navap made some updates
14:32:57 [ruaok]
at the end of 2012 we were.
14:33:02 [ruaok]
thanks navap
14:33:09 [reosarevok]
"congrats on being absolutely nuts, reosarevok, we'll pay for your psychiatrist"
14:33:27 [ruaok]
no WAI.
14:33:30 [reosarevok]
hahaha
14:33:33 [ruaok]
we don't have that much money.
14:35:03 [reosarevok]
Amazon Associates $829.54
14:35:04 [reosarevok]
hmm
14:35:14 [reosarevok]
That's about $800 more than I was expecting :)
14:35:23 [ruaok]
its its mostly nikki's work. :)
14:36:18 [reosarevok]
Payroll Taxes $7,602.47
14:36:25 [reosarevok]
Huh, I thought we didn't have to pay taxes?
14:36:37 [reosarevok]
* reosarevok bets he asks the same every year
14:36:37 [ruaok]
we
14:36:46 [ruaok]
're not exempt from all taxes.
14:36:50 [ruaok]
only income taxes.
14:37:05 [ruaok]
but when you pay a paycheck, you need to pay taxes.
14:37:26 [reosarevok]
Oh
14:38:33 [reosarevok]
Looks fine to me anyway
14:38:46 [Freso]
reosarevok: Can you, by chance, tell me the ids of the two new relationships?
14:39:00 [ruaok]
thanks reosarevok
14:39:01 [reosarevok]
Good question! :p
14:39:05 [kepstin-laptop]
kepstin-laptop has joined #musicbrainz-devel
14:39:07 [reosarevok]
Yes, I can!
14:39:11 [Freso]
\o/
14:39:26 [Freso]
reosarevok: What are the ids of the two new relationships?
14:39:32 [reosarevok]
But not by chance, but because of my own patch which I can finally use for something \o/
14:40:29 [reosarevok]
ID: 718 for artist and ID: 719 for label, it says :)
14:40:45 [reosarevok]
(that seems very high...)
14:41:13 [Freso]
It does.
14:41:25 [Freso]
Hm.
14:41:29 [Freso]
otherdb for places is 561
14:41:43 [Freso]
blog for place i 627
14:41:49 [Freso]
So perhaps not.
14:41:51 [reosarevok]
Seems fine
14:41:55 [reosarevok]
(looking at the code)
14:41:59 [reosarevok]
No idea *why* though
14:42:23 [reosarevok]
nikki, any idea what the ID number depends on?
14:42:37 [nikki]
no idea
14:42:57 [ruaok]
report posted: http://metabrainz.org/doc/Annual_Report/2012
14:43:29 [ruaok]
ha. I get to used the my favorite very infrequently used blog category. :)
14:44:07 [Freso]
Speaking of, I would like a blog category for schema change releases.
14:44:51 [navap]
Don't we already have one?
14:44:51 [Freso]
I think I have permissions to just go and add it myself, but I wouldn't want to do that without being given a "yay" for it.
14:44:59 [Freso]
navap: If we do, it's not being used.
14:45:25 [ruaok]
nope
14:45:28 [ruaok]
I'll add one
14:45:31 [Freso]
At least not for the last couple of schema change releases.
14:45:46 [ruaok]
added
14:49:24 [Freso]
\o/
14:52:17 [Freso]
What part of the code is it that figures out whether a link should be http://, https://, or // ?
14:53:12 [ocharles]
Freso: not sure what you mean
14:53:22 [ocharles]
link in what context?
14:54:20 [nikki]
you're probably looking for url_is_scheme_independent
14:55:04 [ruaok]
reosarevok: will you please tweet about the annual report?
14:55:11 [reosarevok]
Oh, sure!
14:55:12 [Freso]
nikki: Yep! Cheers! :D
14:55:56 [Freso]
And it is already set for Bandcamp urls.
14:55:58 [Freso]
(y)
15:01:37 [reosarevok]
ruaok: what do you say, link to the blog or the report directly?
15:01:42 [ruaok]
* ruaok groans at the latest info@ email
15:01:45 [ruaok]
report
15:06:59 [reosarevok]
Done!
15:07:06 [ruaok]
thx
15:08:14 [ruaok]
ha. thanks reo. you couldn't resist, I see. :)
15:09:03 [reosarevok]
I couldn't :p
15:09:25 [nikki]
* nikki checked and sees that the comment is about exactly what she thought it'd be :P
15:10:41 [mb-chat-logger]
New post: blog: Annual report for 2012 finally posted <http://blog.musicbrainz.org/2013/11/18/annual-report-for-2012-finally-posted/>
15:15:47 [Freso]
reosarevok btw: https://bitbucket.org/metabrainz/musicbrainz-server/pull-request/598 ;)
15:18:16 [marco-oliveira]
marco-oliveira has joined #musicbrainz-devel
15:28:15 [marcooliveira]
marcooliveira has joined #musicbrainz-devel
15:34:57 [navap]
So next Monday my school is having a lockdown simulation of a shooter in the building which includes the theatre students acting as victims and police firing blanks inside the school
15:35:31 [kuno]
oh, fun
15:35:47 [reosarevok]
wtf
15:35:51 [kuno]
who gets to play the shooter?
15:35:59 [reosarevok]
North America is insane
15:36:35 [navap]
kuno: I'm not sure. I wonder how you'd hold auditions for a school shooter
15:36:55 [kuno]
navap: pick whichever student has the highest call-of-duty ranking ofcourse.
15:37:00 [navap]
haha :)
15:37:05 [snartal]
* snartal chuckles
15:37:13 [ocharles]
I was in a top division 2 team when I played Team Fortress 2!
15:37:37 [ruaok]
navap: speaking of canuckistan... any thoughts on that illustrious mayor down the road? :)
15:37:42 [navap]
hah
15:37:51 [navap]
It's surreal
15:38:14 [ruaok]
what a shit-show that is
15:40:54 [navap]
Here's a quick recap for those who don't know: https://www.youtube.com/watch?v=5j6m3Zvfsjs
15:46:11 [ocharles]
is this the crack pipe thing?
15:46:17 [ruaok]
yep.
15:46:23 [ocharles]
haha, yea that was weird
15:46:25 [ruaok]
and the guy won't refuse to step down.
15:46:26 [ruaok]
classy
15:53:33 [reosarevok]
Won't refuse?
15:54:16 [reosarevok]
Also, I'm not sure why having done drugs at some point should invalidate anyone for public duty (if the guy's an idiot, which he might well be, then that's something else :p)
15:54:33 [ruaok]
refuses.
15:54:48 [ruaok]
it seems this guy is a freaking yard.
15:54:59 [ruaok]
damn, you autocorrect.
15:55:01 [ruaok]
tard!
15:55:37 [ocharles]
uhhh
15:55:37 [reosarevok]
A yard wide, yes :p
15:55:38 [ocharles]
CRITICAL: statistics last collected 2013-10-31, 18 days ago
15:55:41 [ocharles]
is that true?
15:56:02 [navap]
reosarevok: He saidHe didn't, then He said He did but Only because Hewas drunk at The time
15:56:14 [ocharles]
you're looking quite drunk there, navap
15:56:21 [ocharles]
or you're talking about God
15:56:25 [reosarevok]
hahaha
15:56:34 [navap]
haha
15:57:02 [navap]
i don't know why Swype keeps capitalizing He
15:57:27 [ocharles]
looks like our nagios checks are broken
15:57:36 [ocharles]
Last updated: 2013-11-18 is what http://musicbrainz.org/statistics has to say
15:57:51 [navap]
anyway, He is definitely a complete idiot and names a fool Of himself each time he's on camera
15:57:53 [reosarevok]
phew
16:02:18 [ruaok]
nice. we need about $7000 of income to have this be our best year ever.
16:02:28 [ruaok]
and google is about to send a $9k payment.
16:02:41 [ruaok]
and who knows maybe amazon will actually send money too!
16:02:51 [reosarevok]
So you need to get rid of $2k not to be too rich, right? You have my paypal? :D
16:03:03 [reosarevok]
I mean, we don't want to show off...
16:03:04 [reosarevok]
:p
16:03:12 [hawke]
hawke has joined #musicbrainz-devel
16:03:15 [ruaok]
nice try. :)
16:03:22 [reosarevok]
Awww
16:03:32 [Freso]
Haha.
16:04:25 [Freso]
http://img.photobucket.com/albums/v694/Cro_ludesi/well-played-sir.jpg
16:09:40 [derwin]
$90 on chocolate.. I heard that non-profits were reckless with donor money but never seen it before!!
16:09:52 [derwin]
(but seriously, awesome yearly report)
16:10:07 [marco-oliveira]
marco-oliveira has joined #musicbrainz-devel
16:10:11 [derwin]
can't wait to see what happened in 2013!
16:10:20 [ruaok]
:-)
16:10:27 [ruaok]
2013 is looking pretty good so far.
16:11:14 [derwin]
grammar error - "Sadly, we have no way of knowing how many total MusicBrainz requests were handled in total in 2012"
16:11:27 [hawke_1]
hawke_1 has joined #musicbrainz-devel
16:12:32 [nikki]
clearly we should start on 2013's now, so we can have it done in the first half of 2013 :P
16:12:36 [nikki]
er
16:12:37 [nikki]
2014
16:12:50 [ruaok]
* ruaok is over this joke now
16:13:29 [nikki]
* nikki wasn't entirely joking.
16:29:13 [ruaok_]
ruaok_ has joined #musicbrainz-devel
16:35:21 [JesseW_not_logge]
JesseW_not_logge has joined #musicbrainz-devel
16:38:07 [ruaok]
ok, I declare December 2nd as "Amazon Cake Day"
16:40:09 [reosarevok]
hahahaha
16:40:58 [ruaok]
they keep not taking me serious.
16:41:04 [ruaok]
so, fuck it. the only way for them to fix it, is to pay us.
16:44:47 [Freso]
+1
16:44:53 [Freso]
I want to see that cake.
16:45:05 [reosarevok]
I want to eat that cake!
16:45:08 [reosarevok]
But you're far!
16:45:10 [ruaok]
it will be posted to twitter and reddit. no worries, you'll see it.
16:45:23 [ruaok]
the cake is not for us to eat.
16:45:35 [ruaok]
in fact, I'm not sure if anyone is going to eat it.
16:45:35 [ruaok]
which is a shame.
16:45:57 [ruaok]
chances are, the people who the cake is destined for may loose their jobs before the cake arrives.
16:46:46 [ruaok]
-lose
16:46:58 [ruaok]
damn, I'll never teach my fingers to stop making that mistake.
16:49:58 [voiceinsideyou]
voiceinsideyou has joined #musicbrainz-devel
16:50:05 [nikki]
k
16:52:37 [voiceinsideyou1]
voiceinsideyou1 has joined #musicbrainz-devel
16:53:51 [reosarevok]
* reosarevok goes to the museum, back in a while
16:59:38 [ocharles]
I think I can sum up today's experience with three words
16:59:44 [ocharles]
Fuck. Mutable. State.
16:59:53 [ocharles]
* ocharles ARGHS at Sql.pm
17:12:24 [jseet]
jseet has joined #musicbrainz-devel
17:16:50 [reosarevok]
reosarevok has joined #musicbrainz-devel
17:17:12 [jseet]
I'm attempting to set up MBS on a shared host server and have gotten (I think) about halfway through. I'm at the database setup stage now and it occurs to me that perhaps I'm missing a key piece of information: is root privilege required to install this from source? The Postgre extensions seem to require direct loading by the database to work...(and hence, installing them locally would make that problematic). Unless, there's anoth
17:17:13 [jseet]
way...
17:19:55 [ocharles]
you don't strictly need root for anything
17:19:56 [Freso]
AFAIK, root access isn't needed for anything.
17:20:13 [ocharles]
if you installed postgresql outside your userspace, then you'll probably need root to install the extensions
17:21:11 [bitmap]
http://tickets.musicbrainz.org/browse/MBS-6986 <- anyone know what "ARTIS" means in the last comment?
17:21:39 [Freso]
jesus2099: ^
17:21:41 [reosarevok]
Artist, but I haven't experienced that problem
17:22:25 [bitmap]
oh, for some reason the missing T made that completely incomprehensible to me :P
17:22:42 [jseet]
ocharles, Freso: thanks. I'll see if it's possible for me to install a private instance...
17:24:49 [bitmap]
since it was uppercased I thought it was some strange acronym like "advanced relationship track inline search"
17:26:56 [reosarevok]
ahahaha
17:27:11 [reosarevok]
Nah, that's just jesus :)
17:27:38 [reosarevok]
bitmap: any chance of fixing the copyable AC script so that I can stop crying every time I edit classical? :(
17:28:12 [bitmap]
reosarevok: yeah I'll look into that today
17:28:14 [reosarevok]
(if you're too busy you're too busy but worth asking :) )
17:28:19 [reosarevok]
\o/ thanks
17:49:46 [ruaok]
ruaok has joined #musicbrainz-devel
17:54:40 [misterswag_]
misterswag_ has joined #musicbrainz-devel
18:02:35 [DWSR]
Quick q: I'm working on a vm image that's 1 version older than the latest one. I was behind enough that I'm 1 schema sequence behind the latest fullexport. I was under the impression that droping the database and starting over with a fresh import would be the quickest way to get back up to speed, but it's still telling me that my codebase is at version #18. I checked out the latest git tag, what's up?
18:03:11 [navap]
DWSR: You'll need to manually update lib/DBDefs.pm and update the schema value to 19
18:03:16 [DWSR]
Oh, that's all?
18:03:29 [navap]
I think so. Try it and lets see :)
18:04:43 [DWSR]
odd that checking out the latest tag doesn't do that? Though I suppose it's a stupidity check for people like me?
18:05:24 [DWSR]
Anyway, seems to be importing now, thanks.
18:05:55 [DWSR]
Last time I did a full database import it took literally 3 days, let's hope that this one runs a little faster.
18:06:09 [navap]
DBDefs.pm isn't in the codebase so that local changes aren't overwritten. There's a default DBDefs.pm that gets updated when you checkout the latest tag, but you'd need to manually copy over any updates
18:06:15 [DWSR]
ah.
18:07:18 [DWSR]
Going through all this trouble just so I can run headphnoes.
18:07:22 [DWSR]
headphones, rather.
18:19:23 [Freso]
DWSR: I usually run vimdiff every now and again against the default and "local" DBDefs. Esp. after "large" updates.
18:20:54 [DWSR]
meh, makes sense.
18:33:29 [misterswag]
misterswag has joined #musicbrainz-devel
18:48:28 [derwin]
derwin has joined #musicbrainz-devel
18:51:46 [reodroid]
reodroid has joined #musicbrainz-devel
18:54:28 [derwin]
derwin has joined #musicbrainz-devel
19:01:19 [derwin]
derwin has joined #musicbrainz-devel
19:17:55 [misterswag]
misterswag has joined #musicbrainz-devel
19:33:48 [LordSputnik]
LordSputnik has joined #musicbrainz-devel
19:40:49 [marcooliveira]
marcooliveira has joined #musicbrainz-devel
19:50:43 [ocharles]
kepstin-laptop: order placed for a shiney T440P :)
19:50:51 [kepstin-laptop]
heh
19:51:07 [kepstin-laptop]
I think I'm leaning towards the T440s
19:51:07 [Freso]
ocharles: \o/
19:51:21 [ianmcorvidae]
ocharles: aware of the ISE on http://beta.musicbrainz.org/edit/19227832 ? that looks like it's related to your sql stuff
19:51:28 [kepstin-laptop]
but either way, not available in canada yet (and the US store doesn't ship here)
19:51:30 [ocharles]
ianmcorvidae: yes, it's in review at the moment
19:51:34 [ianmcorvidae]
okay
19:51:47 [ocharles]
I maxed out just about everything, making it the most expensive laptop i've ever bought :P
19:51:51 [kepstin-laptop]
heh
19:51:59 [ocharles]
but i bought it through my business so I get it 0% rated on VAT
19:52:05 [ocharles]
which comes out about the same as a maxed out T440S
19:52:06 [kepstin-laptop]
quad core, 16gb ram, etc? :)
19:52:11 [ocharles]
yep
19:52:21 [kepstin-laptop]
pity you have to go up to the 15" model to get >1920x1080 screen
19:52:24 [ocharles]
i7 4800MQ, 16gb RAM, fhd display
19:52:41 [ocharles]
my 24" is 1920x1080, I think that same resolution will suffice in a 14" screen ;)
19:52:54 [kepstin-laptop]
quite true :)
19:53:04 [fractalizator2]
fractalizator2 has joined #musicbrainz-devel
19:53:17 [ocharles]
hopefully it will weigh less than my t400
19:53:21 [ocharles]
but I don't expect much difference
19:53:29 [ocharles]
that was the only thing that tempted me about the t440s
19:53:49 [ocharles]
the power consumption is appealing, but 90% of the time I'm using it docked or connected to power anyway
19:54:06 [kepstin-laptop]
yeah, the T440p is really the successor to the previous "normal" T-series laptops
19:54:11 [ocharles]
and it being my primary computer, thought I'd shell out a bit more for quad core (and I'm interested in parallel/concurrent programming)
19:54:14 [ocharles]
yep
19:54:16 [kepstin-laptop]
the T440 is just a cheaper version of the T440s
19:55:02 [ocharles]
Freso: so yea, not long for you to wait!
19:55:21 [Freso]
:)
20:00:10 [ruaok]
<BANG>
20:00:12 [ruaok]
meeting time.
20:00:25 [ruaok]
ianmcorvidae: nawlins? didja have a good time?
20:00:30 [ianmcorvidae]
I did
20:00:51 [ruaok]
good good. its hard to have a bad time there.
20:00:53 [ruaok]
any good noms?
20:00:55 [ianmcorvidae]
mostly hanging out with family, but still :)
20:01:11 [ruaok]
ah, didn't know you had family there.
20:01:13 [ianmcorvidae]
yeah -- we went to a pretty good seafood dive-y place
20:01:16 [ianmcorvidae]
I don't, really
20:01:24 [ianmcorvidae]
my cousin, who was getting married, went to college there
20:01:30 [ianmcorvidae]
so she decided to have the wedding there
20:01:41 [ruaok]
lots of travel for weddings this year. :)
20:01:46 [ianmcorvidae]
haha, indeed
20:01:51 [ruaok]
anyways, welcome back.
20:02:03 [ruaok]
general discussion on the past week?
20:02:22 [ruaok]
I went to a networking event of "the new music industry" on monday.
20:02:36 [uk]
uk has joined #musicbrainz-devel
20:02:37 [ruaok]
at a company called ROLI. they make an interesting instrument called the seaboard.
20:02:45 [ruaok]
* ruaok is unsure if he spelled that right
20:03:18 [ruaok]
it resembles a keyboard, but has a silicone skin with lots of sensors under the skin.
20:03:25 [ruaok]
keyboard++.
20:03:28 [ruaok]
interesting stuff.
20:03:34 [ruaok]
ocharles: I'm sure you've heard of them, no?
20:03:41 [ocharles]
nope, but it sounds like a ribbon keyboard
20:03:51 [ocharles]
ish
20:03:55 [ocharles]
but cool :)
20:04:07 [ocharles]
money is irrelevent when it comes to musical instruments, it seems :)
20:04:08 [ruaok]
I think you'd dig it. interesting stuff.
20:04:16 [ocharles]
i think so too!
20:04:38 [ruaok]
agreed. some people have no problem dropping the equivalent cost of a car on an instrument.
20:04:55 [ruaok]
the rest of my week was more boring.
20:05:00 [ianmcorvidae]
* ianmcorvidae is always amazed how much people spend on guitars, and those aren't even something weird and particular :P
20:05:09 [fractalizator2_]
fractalizator2_ has joined #musicbrainz-devel
20:05:29 [uk_]
uk_ has joined #musicbrainz-devel
20:05:39 [ruaok]
oh, and I really like playing with logstash/elasticsearch and kibana.
20:05:50 [ruaok]
it seems to be aiming squarely for splunk.
20:05:57 [ruaok]
and for our needs I think it will do our trick.
20:06:09 [ocharles]
kibana looks great, is that new?
20:06:18 [ocharles]
i'd never heard of it, but i've been wanting it for ages
20:06:41 [ruaok]
it seems that it was standalone and that elasticsearch did some integrating of bits and pieces.
20:07:01 [ianmcorvidae]
hm
20:07:55 [ruaok]
anyways, I hope to have something useful to show off in a day or so.
20:08:03 [ianmcorvidae]
cool
20:08:06 [ruaok]
shall we proceed with the agenda?
20:08:12 [ianmcorvidae]
* ianmcorvidae wonders how this interacts/overrides graphite and such
20:08:28 [ianmcorvidae]
but maybe that's better asked when we have something to look at
20:08:28 [ocharles]
that would be worth investigating
20:08:36 [ruaok]
ianmcorvidae: graphite can be configured as an output of logstash
20:09:04 [ruaok]
so logstash can do all the collecting/parsing and then it can be handed off to graphite.
20:09:11 [ruaok]
as well as elasticsearch.
20:09:34 [ruaok]
agenda: spam accounts (nikki)
20:09:46 [ianmcorvidae]
okay; it seems like kibana and elasticsearch is slightly parallel to graphite and carbon then, but we can see what it looks like when we have something real to look at :)
20:10:03 [ruaok]
ianmcorvidae: yeah.
20:10:06 [nikki]
so... we have a couple of thousand new spam accounts being created every day
20:10:16 [ruaok]
once we have a play with it we should figure out an overall plan.
20:10:51 [nikki]
which is obviously ridiculous and so I was wondering what we can do about it
20:10:53 [ianmcorvidae]
or at least presumed spam accounts, since we definitely don't have that many people and they're all doing nothing to prove them spammers or not
20:11:06 [ruaok]
one date point to add.
20:11:26 [ruaok]
we've had one felllow mail me personally as the owner of the musicbrainz.org domain and ask me to take links to their site down.
20:11:56 [ruaok]
they said they used a service that was supposed to do harmelss SEO crap, but ended up spamming tons of site for googlejuice.
20:12:11 [ianmcorvidae]
do we actually have spammy links then? I hadn't seen the spam accounts literally posting anything
20:12:16 [ruaok]
the realized that they were spammers and asked us to take down the links.
20:12:27 [ruaok]
they are links in user profiles.
20:12:48 [ruaok]
so, low visibility.
20:12:52 [ianmcorvidae]
* ianmcorvidae wonders if you have an example; I hadn't seen them do anything even there
20:12:58 [kepstin-laptop]
ah, so stuff we could clean up by deleting spam accounts?
20:13:03 [ruaok]
ianmcorvidae: I do. I will forward you the mail I got.
20:13:08 [ianmcorvidae]
okay, cool
20:13:08 [ruaok]
kepstin-laptop: yes.
20:13:20 [ianmcorvidae]
this of course does nothing re: the question of what to do
20:13:20 [ruaok]
so, we need to do some account cleanup.
20:13:30 [nikki]
one of the things I wondered about was deleting completely unused accounts, spammers or not
20:13:41 [ruaok]
I told the chap that a $1000 donation to cover our work would help matters along. no response yet.
20:13:52 [ianmcorvidae]
even before we switched to bcrypt we had a bunch of them, which is why I'd grabbed a dump of the editor table to process -- but I haven't done so
20:13:56 [ruaok]
nikki: define completely unused.
20:14:16 [nikki]
no edits, edit notes, votes, tags, collections, ratings or subscriptions
20:14:30 [ruaok]
what if they have a verified email address?
20:14:32 [Freso]
nikki: Verified e-mail?
20:14:34 [Freso]
ruaok: :)
20:14:51 [ocharles]
out of curiousity, what problem would that solve? The only problem I see at the moment is name-holding
20:14:54 [hawke_1]
hawke_1 has left #musicbrainz-devel
20:15:03 [ruaok]
ocharles: it dilutes our statistics.
20:15:11 [hawke_1]
hawke_1 has joined #musicbrainz-devel
20:15:14 [ruaok]
we can't tell if our community is growing on shrinking.
20:15:15 [nikki]
I wouldn't count verifying an email address as using the account. that doesn't interact with the mb data in any way
20:15:17 [ianmcorvidae]
ocharles: well, that's why I was asking for examples of links in profiles, that's a real use of the accounts
20:15:27 [ocharles]
ruaok: presumably the stat we care about is editors who make edits
20:15:30 [ocharles]
not registrations?
20:15:55 [ruaok]
yes, but I hate that our stats are getting spammed including our accounts.
20:15:57 [ocharles]
registrations and no edits doesn't imply spam, it could also imply our site is awfully hard to use
20:16:11 [ruaok]
and there is potential for many accounts that are dormant go explode and spam everywhere.
20:16:13 [ocharles]
though the knee is step, so i agree - it's probably more likely spam
20:16:25 [ruaok]
I
20:16:27 [ruaok]
feh.
20:16:37 [ruaok]
I'd like to consider a few things: first, a new capcha.
20:16:47 [nikki]
well, if our site is so hard to use that the person can't figure out how to enter a single edit (correct or not), they're not going to have much use for the account anyway :P
20:16:59 [ruaok]
second, running user bios/links and edit notes through a spam checker.
20:17:15 [reodroid]
reodroid has joined #musicbrainz-devel
20:17:16 [nikki]
in real time?
20:17:19 [ocharles]
nikki: that's my point. if we have registrations and no edits, then we should reach out to these people
20:17:25 [ianmcorvidae]
we should also ensure links in profiles/etc. are marked with rel=nofollow
20:17:28 [ocharles]
ruaok: i tried to do that
20:17:30 [ocharles]
i didn't have much luck
20:17:30 [ruaok]
nikki: no
20:17:30 [nikki]
we can't, because they're mostly spammers
20:18:04 [ianmcorvidae]
(it won't stop people making the accounts, but it'll ensure that it doesn't actually benefit the people spamming with those tools)
20:18:07 [ruaok]
ok, captcha is an important one.
20:18:21 [ruaok]
the google one is clearly really fucking broken.
20:18:29 [ruaok]
there are other tools out there.
20:18:39 [ocharles]
most captchas are broken these days, once you have a high enough google rank
20:18:40 [ruaok]
what if we all had a look around for alternatives to put in its place?
20:18:47 [ianmcorvidae]
http://areyouahuman.com/ is at least semi-popular
20:18:51 [ocharles]
and with the bbc and twitter linking to us, that rank is only going to go up
20:18:51 [ianmcorvidae]
and yeah
20:18:59 [ruaok]
ianmcorvidae: I was considering that.
20:19:09 [ianmcorvidae]
if nothing else, it's very very easy to just farm out captcha solving to a farm of humans
20:19:15 [ocharles]
exactly
20:19:37 [ocharles]
areyouhuman looks fun :)
20:19:59 [ruaok]
ocharles: shall we try it?
20:20:06 [Freso]
We could also add a HTTP auth like we have for the wiki...
20:20:07 [ruaok]
it probably won't take much effort to put it in place.
20:20:07 [Freso]
7me hides
20:20:21 [ocharles]
I don't see any captcha as a particularly robust solution. what about considering better reporting tools for users?
20:20:38 [ocharles]
buttons to report users, and an admin panel to process the reports
20:20:56 [ocharles]
you could also generate reports dynamically
20:21:01 [ocharles]
(such as editors with no data)
20:21:05 [reodroid]
for a thousand users?
20:21:27 [ocharles]
reodroid: yes
20:21:43 [ianmcorvidae]
* ianmcorvidae corrects: for a thousand users *per day*
20:21:54 [ocharles]
ok, that is a lot
20:22:10 [nikki]
we're getting around 2000 new users a day
20:22:14 [reodroid]
I mean, most people won't even see the stuff
20:22:15 [nikki]
so probably closer to twice that :P
20:22:32 [ianmcorvidae]
by my estimate I'd guess about 1/2-3/4 of our total editors are likely to be spammers
20:22:51 [reodroid]
I thought it was more an issue of not having all those potential sleeping accounts around
20:22:52 [ruaok]
fun.
20:23:01 [ruaok]
nikki's idea is starting to sound better.
20:23:17 [ruaok]
but even then, they'll make one edit note.
20:23:23 [ruaok]
and bam the account stays.
20:23:27 [reodroid]
they can't
20:23:39 [reodroid]
they can make an edit though...
20:23:41 [ruaok]
previous sockpuppet rules?
20:23:47 [derwin]
what do spammers do with musicbrainz accounts?
20:23:55 [ruaok]
that could be inviting real trouble.
20:24:09 [ocharles]
derwin: the normal forum type spam. links to cheap watches, etc
20:24:11 [ruaok]
derwin: googlejuice in account bios and spammy edit notes
20:24:12 [ianmcorvidae]
derwin: the same thing they do with anything, post links
20:24:13 [ianmcorvidae]
yeah
20:24:15 [ocharles]
that's the majority i've seen so far
20:24:53 [ruaok]
bios would could fix by saying that we wont show the users' bio until they have x accepted edits
20:25:20 [reodroid]
that sounds like overkill
20:25:20 [ianmcorvidae]
no policy will stop them from making accounts
20:25:30 [reodroid]
just kill empty accounts
20:25:43 [ruaok]
I'm up for giving that a try.
20:25:44 [ocharles]
We could have another look at spam detection on edit notes and bios, and if we see a submission that is spammer, show a captcha to confirm the submission (unless the user has already confirmed at least one false-positive in the past)
20:25:47 [ruaok]
any objections?
20:25:54 [ianmcorvidae]
that potentially prevents the benefit to them, but most of these things don't do anything anyway
20:25:58 [reodroid]
we can worry about they doing things when they do things
20:26:36 [nikki]
we don't even let google index user pages afaik, and they haven't noticed
20:26:37 [ocharles]
for me personally, touching accounts is the last resort
20:26:39 [ianmcorvidae]
I do think that spam detection is the way we have to go here, otherwise we're fighting a losing battle against people who don't care if they lose :P
20:26:43 [ocharles]
because it's not really my data
20:26:43 [nikki]
so they're probably also not going to notice that the accounts are gone :P
20:26:58 [ruaok]
nikki: these people who asked me to take their links down noticed.
20:27:11 [nikki]
the spammers, I mean
20:27:21 [ruaok]
ah
20:28:28 [ruaok]
how about disallowing links in user bios? we can show the URL, but not make it a link.
20:28:45 [ruaok]
that kills one venue of spamming.
20:28:49 [ocharles]
what about the homepage property?
20:28:55 [ocharles]
or whatever we call it
20:29:06 [ruaok]
same thing.
20:29:08 [ruaok]
show URL
20:29:10 [ianmcorvidae]
I really can't stress enough that something being ineffective for the spammers does not matter an ounce to the spammer
20:29:22 [derwin]
what ianmcorvidae says
20:29:27 [ruaok]
maybe we can show a link for people who are viewing it who have made good edits.
20:29:30 [ocharles]
i'm with ianmcorvidae
20:29:41 [ruaok]
ianmcorvidae:
20:29:44 [ruaok]
feh
20:29:44 [ruaok]
http://musicbrainz.org/user/clamzephyr2
20:29:49 [ruaok]
that *is* effective
20:29:50 [ianmcorvidae]
posting the text for <a href="whatever">cheap watches lol!</a> is without any cost for the spammer
20:30:11 [ocharles]
under my suggestion, that bio would probably not have got through, because it would require a captcha to be posted
20:30:19 [navap]
Stack overflow has a point based system of increasing privileges. Can we borrow from that somehow?
20:30:35 [ianmcorvidae]
as our previous run-in with these people shows, since they were literally just posting in broken HTML that I'm not sure actually even made a link
20:30:39 [ruaok]
navap: we do that in a crude way already.
20:31:00 [ianmcorvidae]
which is why I said if we want to cut off the productivity of the spamming we add rel=nofollow and move on
20:31:11 [nikki]
oh, the other thing I was curious about was only allowing an email to be used once. there's not really any good reasons to have multiple accounts with the same email address (a few cases, but not many, and yeah, I know it can be worked around pretty easily, I'm not claiming it can't, it just seems more like a case of "why do we even allow that?")
20:31:18 [ianmcorvidae]
if we care about the cruft in our database we need to do something more proactive
20:31:46 [navap]
As a matter of pride, I think we should care about cruft in the DB
20:31:54 [ruaok]
nikki: if you enter me a ticket I will run a query to find out the number of unique emails vs accounts
20:32:06 [ianmcorvidae]
ruaok: some emails have as many as thousands of accounts
20:32:13 [reodroid]
I care about the "fake" stats, not much about the accounts themselved
20:32:13 [ianmcorvidae]
we were looking the other week :P
20:32:16 [ruaok]
ianmcorvidae: a good.
20:32:21 [derwin]
as a DBA, I always care about cruft in DBs
20:32:37 [ruaok]
well, one email has more than say, 5, accounts, label them as a spammer and nuke all accounts?
20:32:47 [reodroid]
poor nikki
20:32:49 [ruaok]
maybe make a report from it first?
20:33:27 [ianmcorvidae]
we can't make such a report public, but possibly
20:33:33 [ruaok]
I care about cruft in the DB as well.
20:33:44 [ruaok]
lets pick off the low-hanging fruit.
20:33:46 [navap]
Email the flagged user and tell them to contact us, if they don't in 2 weeks, delete their accounts
20:33:59 [navap]
The flagged email*
20:34:04 [ruaok]
does anyone have any objections to the nofollow= suggestion?
20:34:16 [nikki]
it sounds sensible, whatever else we do
20:34:17 [ocharles]
+1 on nofollow to non-musicbrainz links
20:34:30 [ruaok]
to all links outbound?
20:34:34 [navap]
Are we then adding nofollow to relationship links as well?
20:34:36 [ruaok]
even in edit notes?
20:34:41 [ianmcorvidae]
edit notes yes, relationships no
20:34:44 [navap]
nofollow in edit notes yes
20:34:45 [ruaok]
ianmcorvidae: +1
20:34:46 [ianmcorvidae]
relationships require voting etc.
20:34:59 [ianmcorvidae]
the places for nofollow, for me, are:
20:35:09 [ianmcorvidae]
annotation, edit note, user bio, user homepage
20:35:13 [marcooliveira]
marcooliveira has joined #musicbrainz-devel
20:35:13 [ianmcorvidae]
I think that's all
20:35:18 [hawke_1]
Don’t you have to be logged in to see edit notes?
20:35:24 [reodroid]
what's nofollow?
20:35:25 [derwin]
don't know why we'd want nonfollow links anywhere?
20:35:35 [derwin]
err, not-nofollow
20:35:54 [ianmcorvidae]
nofollow says to search engines that you don't approve of this link, basically
20:36:04 [ianmcorvidae]
i.e. don't use it in calculations of googlejuice, as ruaok would put it
20:36:23 [derwin]
right, my understanding is we generally do not want googlejuice
20:36:24 [ruaok]
so, that helps the first bit, I'll enter a ticket for that.
20:36:24 [navap]
derwin: Inter-MB links should be followed, otherwise nothing would get indexed
20:36:30 [uk_]
https://en.wikipedia.org/wiki/Nofollow
20:36:36 [navap]
Or is that intra-MB*
20:36:36 [derwin]
right, I'm saying out-links, navap.
20:36:38 [ruaok]
hi uk_ !
20:36:38 [ianmcorvidae]
and the reason we'd want relationships to not be nofollow'd is that for those we actually do want googlejuice flowing to official artist homepages, twitters, etc.
20:36:49 [uk_]
hi :)
20:36:54 [ianmcorvidae]
because those are useful outbound links
20:37:03 [ocharles]
* ocharles nods
20:37:06 [derwin]
ah, yes, true. relationship links.
20:37:24 [ruaok]
ok I think we agree on nofollow.
20:37:29 [ianmcorvidae]
cool
20:37:31 [ruaok]
whats our next low hanging fruit?
20:37:50 [navap]
Email sockpuppeting?
20:37:59 [ruaok]
yes, that is a good one.
20:38:07 [ruaok]
lets say its easy to find these.
20:38:07 [ianmcorvidae]
switching captcha, probably, and some sort of investigation of merging accounts that share emails
20:38:23 [ruaok]
what is our course of action when we find them?
20:38:37 [ruaok]
navap's idea of mailing them and if no answer in two weeks, killing the account.
20:38:38 [ianmcorvidae]
I'd say our course of action in general is to support merging editors
20:39:00 [navap]
Merging sounds interesting
20:39:03 [ruaok]
would we have a user_redirect page?
20:39:06 [ocharles]
with a move to making emails unique in the database?
20:39:10 [ianmcorvidae]
ocharles: yes
20:39:13 [reodroid]
huh
20:39:17 [ianmcorvidae]
well, a move towards that
20:39:29 [ruaok]
and prevent sign ups if that email is already in use?
20:39:32 [reodroid]
how is that useful for unused accounts?
20:39:36 [Freso]
Editors as entities!
20:39:43 [ianmcorvidae]
I think we'd want legitimate users to be given a chance to change to a different email or merge, at their discretion
20:39:45 [Freso]
* Freso re-hides
20:39:46 [navap]
I think there are strong cases for having the same email for 2 or 3 accounts, but those are very very rare
20:39:55 [derwin]
frankly, that's a surprising to me. and I dunno about merging, usually hard.
20:39:57 [navap]
Just people in here
20:40:11 [Freso]
Bots, etc.?
20:40:11 [ianmcorvidae]
navap: I think that using gmail's +whatever or a second email that forwards is perfectly fine there though
20:40:22 [ianmcorvidae]
having looked at this
20:40:37 [ianmcorvidae]
most of our duplicate emails are either people who wanted to change their username but couldn't, or probably-spammers
20:40:51 [ianmcorvidae]
even nikki mostly uses different emails on a string comparison basis :P
20:41:42 [ruaok]
ianmcorvidae: do you have any numbers laying around for how common this is?
20:41:48 [ruaok]
if not, maybe collect some for next week's meeting?
20:41:49 [ianmcorvidae]
not laying around, no
20:42:01 [ianmcorvidae]
I sent nikki a report thingy, maybe I can hunt that down again
20:42:01 [ruaok]
and then we can pick this up once we have some numbers to look at.
20:42:02 [nikki]
how would merging work? I'm not sure I agree with having people who are already well-known go around changing their usernames all the time by merging into a new one
20:42:09 [ruaok]
and also to ponder this more.
20:42:26 [ocharles]
i again don't really see what the problem we're trying to solve is
20:42:32 [ocharles]
is this a numbers thing?
20:42:39 [ruaok]
clean up cruft in our db.
20:42:52 [ruaok]
we aim to have a clean db, therefore want the spammer user accounts gone
20:43:04 [derwin]
"why" ?
20:43:20 [derwin]
I mean, as a DBA, I <3 clean DB.. but.. in reality often not justifiable?
20:43:21 [ruaok]
we dot accept clutter elsewhere. why accept it here?
20:43:27 [ianmcorvidae]
there's a bunch of different things we want, I think, and we're helpfully talking about all of them at once
20:43:28 [ocharles]
because it's not public
20:43:39 [ocharles]
we have cruft in the code too, and we live with that :)
20:43:51 [ianmcorvidae]
yeah, I mean, that argument can apply to, say, tags and cdstubs too
20:44:21 [reodroid]
IMO mostly because we don't have n million real editors and it feels misleading to claim that because of spammers
20:44:30 [ruaok]
reodroid: +1
20:44:33 [ocharles]
reodroid: surely we can formulate better queries to get better insight
20:44:38 [marcooliveira]
marcooliveira has joined #musicbrainz-devel
20:44:43 [reodroid]
I might as well keep them if we can not count it for stats
20:44:46 [ocharles]
we have more options than just SELECT count(*) FROM editor
20:44:58 [derwin]
every "[x] has [y] users" stat you have ever seen, ever, has been misleading in the same way.
20:45:03 [ruaok]
ocharles: I'd love for you to spend a couple of hours on this when you can.
20:45:07 [ianmcorvidae]
improving the query for count.editor.valid and maybe adding a count.editor.inactive stat would be reasonable
20:45:09 [ruaok]
see if you can come up with anything interesting.
20:45:15 [ocharles]
ruaok: ok, what bit is the "this"?
20:45:26 [ruaok]
"we have more options than just SELECT count(*) FROM editor"
20:45:27 [ocharles]
ok
20:45:30 [reodroid]
derwin: probably, that doesn't mean we shouldn't try to avoid it :)
20:45:37 [ocharles]
i'd be happy to do some analysis there
20:45:39 [derwin]
it may? heh.
20:45:53 [ruaok]
ok, lets leave it here for now.
20:46:02 [ruaok]
we've spent 40 minutes on this already
20:46:08 [ianmcorvidae]
I'll try to have some numbers on email reuse for next week.
20:46:22 [ruaok]
we have one concrete step (nofollow) and then some investigation by both ianmcorvidae and ocharles
20:46:29 [ruaok]
then next week we look at it again.
20:46:35 [ruaok]
sound reasonable?
20:46:48 [ruaok]
ianmcorvidae: OAuth2 MAC auth
20:46:59 [ianmcorvidae]
I'm also happy to look at stats if ocharles would rather look at his in-progress stuff
20:47:02 [ianmcorvidae]
anyway
20:47:03 [ianmcorvidae]
MAC auth.
20:47:19 [ianmcorvidae]
so we currently support two varieties of authentication with OAuth
20:47:23 [ianmcorvidae]
bearer tokens and MAC tokens
20:47:43 [ocharles]
ianmcorvidae: that might work better, now that i have more work attribute stuff to do
20:47:45 [ianmcorvidae]
bearer is much easier to use and actually standardized; MAC was added, as luks said, because at the time he assumed that using https requests in picard would be a no-no
20:48:05 [ianmcorvidae]
MAC auth is also misimplemented in a way that makes the tests very very angry with me on perl 5.18
20:48:13 [ocharles]
:)
20:48:22 [ianmcorvidae]
we could fix it, but I'd like to propose we just remove MAC auth instead.
20:48:24 [ocharles]
does MAC auth actually work on production?
20:48:33 [ianmcorvidae]
if we had any clients using it we might know
20:48:43 [ianmcorvidae]
the tests work on production, the multiple-nonce protection may not
20:48:44 [ocharles]
we could easily push to graphite to get that info
20:48:57 [ianmcorvidae]
(that being the thing that's broken)
20:49:01 [ocharles]
to see if that path is at least executed
20:49:39 [ianmcorvidae]
we could put in something there for a week or something, but are people okay with us getting rid of that code if, as I expect, it's not used at all?
20:50:06 [ruaok]
sounds like a good approach
20:50:12 [ianmcorvidae]
okay
20:50:28 [ianmcorvidae]
segue, then: ocharles, do we have statsd running somewhere so we can shove something into the code to track this? :)
20:50:37 [ianmcorvidae]
(I assume a statsd thing would be the easiest approach)
20:50:46 [luks]
I think it would be easier to check in the DB if there are any mac tokens
20:50:52 [ianmcorvidae]
oh, that's a good point
20:51:27 [ruaok]
right then.
20:51:33 [ruaok]
that makes that easier.
20:51:37 [ocharles]
ianmcorvidae: we do have statsd somewhere, but you only need carbon-aggregator for this
20:51:43 [ocharles]
luks: oh, heh, that'd be much easier :)
20:51:49 [ruaok]
should we be done or should we be good and do a DR?
20:51:55 [ianmcorvidae]
select * from editor_oauth_token where mac_key is not null; returns no rows
20:52:12 [ruaok]
nuke it then. :)
20:52:37 [luks]
implementing it was a bad decision in the first place, nobody actually uses it
20:52:56 [ianmcorvidae]
okay
20:53:00 [ianmcorvidae]
I'll get rid of it then
20:53:04 [ianmcorvidae]
then I'll be able to run tests again :)
20:53:07 [ianmcorvidae]
and done, onward
20:53:21 [ruaok]
I think onward is… onward with our evenings.
20:53:29 [ruaok]
thanks for your time, everyone.
20:53:31 [nikki]
there was something I wanted to bring up today, but I still can't remember what :P
20:53:38 [ruaok]
that was an enlightening discussion on spam, TBH.
20:53:49 [ruaok]
</BANG>
20:53:49 [MBChatLogger]
MBChatLogger has changed the topic to: http://musicbrainz.org/#devel
20:54:23 [nikki]
can we make MBChatLogger stop doing that?
20:54:37 [navap]
(what's the point of that URL anyway)
20:55:18 [ruaok]
we need to ditch that old logger soon anyway.
20:55:31 [navap]
Question: Have we considered providing musicbrainz-server as an ubuntu package? I attended a talk where some people were handling dependencies by doing that
20:55:53 [reosarevok]
reosarevok has joined #musicbrainz-devel
20:55:53 [ocharles]
navap: we're almost there
20:55:57 [ocharles]
at least in terms of dependencies
20:56:11 [ocharles]
it just needs libmusicbrainz-server-dev to be made
20:56:32 [ocharles]
all of the pieces that make up libmusicbrainz-server-dev exist on my launchpad
20:56:48 [ocharles]
that is in fact how we deploy musicbrainz to our own servers now
20:57:01 [ocharles]
and once we get libmusicbrainz-server-dev, then our servers can be better pruned of unused dependencies
20:58:52 [navap]
Well, that's good news then, I think
21:02:00 [derwin]
derwin has joined #musicbrainz-devel
21:04:48 [ianmcorvidae]
hm, I guess for nofollow I'll need to provide a patch for Text::WikiFormat
21:05:10 [fractalizator2]
fractalizator2 has joined #musicbrainz-devel
21:16:42 [ocharles]
ianmcorvidae: I have maintainership for that
21:16:51 [ianmcorvidae]
ocharles: yeah, I know
21:16:55 [ocharles]
ok :)
21:17:11 [ocharles]
time to study, bye bye!
21:17:25 [ianmcorvidae]
I'm currently putting up an MBS review that monkeypatches it with a note that moving it over is better :)
21:33:08 [fractalizator2]
fractalizator2 has joined #musicbrainz-devel
21:35:18 [marcooliveira]
marcooliveira has joined #musicbrainz-devel
21:44:41 [misterswag]
misterswag has joined #musicbrainz-devel
21:46:01 [fractalizator2_]
fractalizator2_ has joined #musicbrainz-devel
22:05:27 [misterswag]
misterswag has joined #musicbrainz-devel
22:08:25 [DWSR2]
DWSR2 has joined #musicbrainz-devel
22:09:33 [alastairp]
nikki: hello?
22:10:25 [DWSR2]
Hey, getting Caught exception in MusicBrainz::Server::Controller::Root->begin "stash takes a hash or hashref at /usr/lib/perl5/Class/MOP/Method/Wrapped.pm line 162" when accessing my local mirror. Just did a clean import of the database and restarted jetty as well, seems to give me the same thing.
22:12:48 [snartal]
snartal has joined #musicbrainz-devel
22:14:47 [marcooliveira]
marcooliveira has joined #musicbrainz-devel
22:15:56 [Freso]
ianmcorvidae: Sucks about that not breaking tests.
22:16:24 [ianmcorvidae]
yeah, I'm confused about that
22:16:40 [ianmcorvidae]
if it's updated I can still merge it though, I need to fix up one of my patches and there's a chunk of regressions bitmap was working on
22:20:25 [Freso]
ianmcorvidae: I just pushed the fix. Let me update the PR real quick.
22:20:51 [Freso]
ianmcorvidae: Updated.
22:20:57 [ianmcorvidae]
cool, thanks
22:22:07 [Freso]
ianmcorvidae: I also have https://bitbucket.org/metabrainz/musicbrainz-server/pull-request/594 open, btw... :p
22:22:14 [ianmcorvidae]
yeah, I know
22:22:38 [Freso]
:)
22:22:48 [Freso]
Anyway.
22:22:56 [Freso]
The Bandcamp one is the more urgent one.
22:22:59 [ianmcorvidae]
yeah
22:34:21 [ianmcorvidae]
amazing, down to only 11 PRs
22:42:52 [SultS]
SultS has joined #musicbrainz-devel
22:47:41 [Freso]
I think it's Google's time to do some fire-fighting.
22:47:55 [Freso]
https://www.youtube.com/
22:49:46 [warp]
hah :)
22:50:33 [Freso]
* Freso wouldn't want to be a Google server admin right now :p
22:56:26 [DWSR2]
Freso: huh?
22:58:01 [warp]
DWSR2: youtube.com was showing a 500 Internal Server Error.
22:58:15 [DWSR2]
oh, maybe because I'm signed in it wasn't showing for me.
22:58:29 [DWSR2]
Anyway, anyone with a suggestion for my earlier problem?
22:58:33 [DWSR2]
Hey, getting Caught exception in MusicBrainz::Server::Controller::Root->begin "stash takes a hash or hashref at /usr/lib/perl5/Class/MOP/Method/Wrapped.pm line 162" when accessing my local mirror. Just did a clean import of the database and restarted jetty as well, seems to give me the same thing.
22:58:47 [DWSR2]
Tried restarting the entire box as well, didn't seem to do anything.
23:00:07 [Freso]
DWSR2: As warp said, YouTube was utterly and completely down a few minutes ago.
23:00:21 [Freso]
It had probably at least 10 minutes of downtime.
23:05:32 [marcooliveira]
marcooliveira has joined #musicbrainz-devel
23:21:27 [reosarevok_]
reosarevok_ has joined #musicbrainz-devel
23:24:20 [LordSputnik]
LordSputnik has left #musicbrainz-devel
23:40:50 [rvedotrc1]
rvedotrc1 has joined #musicbrainz-devel
23:46:21 [ianmcorvidae]
* ianmcorvidae wonders if anyone who has opinions on editor stats is still around
23:46:47 [ianmcorvidae]
anyway, I'm thinking I'll add three more. At present we have count.editor, count.editor.deleted, and count.editor.valid (where valid = not deleted)
23:47:38 [ianmcorvidae]
the three I'd add would be count.editor.valid.{inactive,verified,active}, where the three steps are a.) no verified email, nothing done on-site, b.) verified email but done nothing, and c.) has taken at least one action, ever