IRC log of musicbrainz on 2004-02-17
Timestamps are in UTC.
- 00:16:28 [djce]
- djce has quit
- 01:03:14 [elinenbe]
- elinenbe has quit
- 01:07:06 [Necronom]
- ruaok: When an album is merged, can the old AlbumID be an "alias" to the new one?
- 01:07:33 [ruaok]
- It should be, but it is not right now.
- 01:07:46 [ruaok]
- That's on the list of things to fix...
- 01:07:54 [Necronom]
- Okay..
- 01:08:23 [Necronom]
- I've got about 30 albums that I need to retag for that reason.
- 01:08:58 [ruaok]
- :(
- 01:10:04 [Necronom]
- Do you know of a perl module that groks id3v2 tags?
- 01:10:15 [Necronom]
- Err. id3v2.4.0.
- 01:10:38 [ruaok]
- no.
- 01:10:57 [ruaok]
- Did I mention that I want to shoot the author of id3v2?
- 01:11:12 [ruaok]
- non-backward-compatible versions SUCK ASS.
- 01:11:19 [Necronom]
- No. Need a gun?
- 01:11:23 [ruaok]
- :-)
- 01:11:33 [ruaok]
- I dunno if I could hit .se from here
- 01:11:36 [ruaok]
- :-)
- 01:12:04 [Necronom]
- I'll mod up tagged, then. It seems to read the tags, but doesn't understand all of the fields.
- 01:12:37 [ruaok]
- its worse than that.
- 01:12:47 [ruaok]
- The internal structure of 2.4 is different from 2.3
- 01:12:56 [ruaok]
- it will require some code changes to handle that as well.
- 01:13:09 [Necronom]
- I'll buy you a plane ticket to .se?
- 01:13:18 [ruaok]
- PLEASE!
- 01:13:50 [Necronom]
- Hmm. Maybe I'll swig the id3v2 library for perl?
- 01:14:47 [ruaok]
- Ick, no.
- 01:14:55 [ruaok]
- The reference lib is chock full of bugs.
- 01:15:02 [ruaok]
- The pure perl code is more solid.
- 01:15:17 [ruaok]
- * ruaok hears Necronom browsing for plane tix to .se
- 01:15:44 [Necronom]
- No, that's me browsing through the tunepimp code to see how it uses id3v2..
- 01:16:04 [ruaok]
- see libid3 dir.
- 01:16:12 [ruaok]
- or id3tag, yeah that's it
- 01:16:21 [ruaok]
- it uses underbit's id3 library.
- 01:16:28 [ruaok]
- With lots of patches so that it works.
- 01:16:55 [ruaok]
- I've submitted a number of patches to the author, but he hasn't accepted any of them or released a new version.
- 01:17:02 [ruaok]
- So I'm using my own copy. Grrr.
- 01:17:20 [Necronom]
- Hmm.
- 01:17:48 [ruaok]
- perl wrapping that lib might be useful.
- 01:17:56 [Necronom]
- That's what I'm thinking..
- 01:42:20 [jwatk03]
- jwatk03 has joined #musicbrainz
- 01:43:34 [spooge]
- spooge has joined #musicbrainz
- 02:12:59 [jwatk03]
- jwatk03 has quit
- 02:18:33 [jwatk03]
- jwatk03 has joined #musicbrainz
- 02:25:27 [jwatk03]
- In the MBHttp class, how come the Progress method is empty? Was it something that was being worked on?
- 02:26:29 [ruaok]
- it was used when we ganked the code from freeamp, but we're not using it.
- 02:26:59 [jwatk03]
- ok
- 02:27:25 [spooge]
- is anyone else still having those random xp crashes when doing lookups/analyze?
- 02:28:03 [ruaok]
- * ruaok doesn't use XP
- 02:28:40 [ruaok]
- but, it appears the crashes have not gone away entirely.
- 02:28:46 [ruaok]
- just a lot fewer.
- 02:29:04 [ruaok]
- is it possible to install a clean copy of IE??
- 02:29:38 [jwatk03]
- i have xp, but it hasn't crashed so far
- 02:30:15 [spooge]
- hmmm, even the latest client can bearly stay open 1-2 min before crashing... not nearly usable.
- 02:30:32 [ruaok]
- wow.
- 02:30:42 [ruaok]
- try installing a clean IE.
- 02:30:57 [ruaok]
- and completly uninstalling the tagger.
- 02:31:07 [spooge]
- hmm, ok
- 02:31:09 [ruaok]
- ok, try this in the given order
- 02:31:14 [ruaok]
- 1. uninstall the tagger
- 02:31:21 [ruaok]
- 2. remove all tagger created dirs
- 02:32:00 [ruaok]
- 3. remove the registry keys for the tagger (LOCAL MACHINE|CURRENT USER\Software\MusicBrainz)
- 02:32:09 [ruaok]
- 4. Install a clean copy of IE
- 02:32:14 [ruaok]
- 5. Install the latest Tagger
- 02:32:19 [ruaok]
- 6. Report
- 02:32:21 [ruaok]
- :-)
- 02:33:55 [spooge]
- but i must say the error seems to be unrelated to browsing (it happens if i browse or not) and also does not happen when i am exclusively browsing (without any tracks being analyzed)
- 02:33:59 [spooge]
- but i'll try it
- 02:34:15 [ruaok]
- well, when does it crash?
- 02:34:54 [spooge]
- after attempting to identify roughly 1-50 tracks
- 02:35:14 [ruaok]
- grrr.
- 02:35:19 [spooge]
- i've watched the status line to see what file/action it appears to be doing at the time, and...
- 02:35:25 [ruaok]
- ok, that's not IE related.
- 02:35:58 [spooge]
- it hasn't been the same any of the say 5 times i just tested
- 02:36:17 [spooge]
- although it does appear to go through the same files each time in the same order
- 02:36:40 [ruaok]
- and crashes in different places??
- 02:36:46 [spooge]
- yes
- 02:36:55 [ruaok]
- lovely.
- 02:37:05 [ruaok]
- and not related to the set of music you're analyzing?
- 02:37:27 [spooge]
- usually the error pane is empty, but maybe 15% of the time it is filled with 10-15 error messeges with "Decoder Crashed"
- 02:37:50 [ruaok]
- ahh, that is a start.
- 02:37:58 [spooge]
- those come quite quickly after one appears and then it crashes shortly after
- 02:38:20 [ruaok]
- ok, thats making it more clear now.
- 02:38:37 [ruaok]
- One crash is causing instability.
- 02:38:49 [ruaok]
- I don't have many bad files laying around.
- 02:38:52 [spooge]
- unfortunately that doesn't happen every time, but it does seem at least slightly related
- 02:39:04 [ruaok]
- likely.
- 02:39:10 [spooge]
- lemme reinstall and try something brb
- 02:39:16 [ruaok]
- can you stash 100 - 200 of the files on the net for me?
- 02:39:23 [ruaok]
- hold on
- 02:39:29 [ruaok]
- I doubt it will help.
- 02:39:35 [ruaok]
- Its likely more file related.
- 02:43:58 [spooge]
- spooge has quit
- 02:44:15 [spooge]
- spooge has joined #musicbrainz
- 02:44:24 [jwatk03]
- jwatk03 has quit
- 03:03:08 [lgonze]
- lgonze has joined #musicbrainz
- 03:03:20 [lgonze]
- hey boyz.
- 03:03:27 [lgonze]
- it is my impression that mb totally rocks
- 03:03:31 [lgonze]
- correct me if I'm wrong.
- 03:03:48 [ruaok]
- I hope so.,
- 03:03:54 [ruaok]
- Missed ya at Etech!
- 03:03:57 [lgonze]
- :)
- 03:04:09 [lgonze]
- wish I'd gone.
- 03:04:09 [ruaok]
- so, what makes you say that?
- 03:04:17 [ruaok]
- 'twas fine.
- 03:04:54 [ruaok]
- I was talking to the Mike L and Danbri when Bob Frankston butted into the conversation.
- 03:05:09 [ruaok]
- Mind you the creator of the spreadsheet butted into the conversation. :-)
- 03:05:22 [ruaok]
- I liked that moment...
- 03:05:51 [lgonze]
- very awesome.
- 03:05:55 [ruaok]
- lgonze: you seen http://test.musicbrainz.org/album/8f468f36-8c7e-4fc1-9166-50664d267127.html
- 03:06:01 [ruaok]
- ?
- 03:06:08 [lgonze]
- checking it out...
- 03:06:43 [lgonze]
- has that changed? what's new?
- 03:07:02 [ruaok]
- the main site doesn't have cover art.
- 03:07:08 [lgonze]
- the image is quite nice.
- 03:07:13 [lgonze]
- hmmmmmmmmmm very nice indeed.
- 03:07:14 [ruaok]
- helps the page lots.
- 03:08:09 [lgonze]
- where's the image come from? is that from amazon?
- 03:08:21 [ruaok]
- yep.
- 03:08:54 [lgonze]
- hey, and you have the "buy this" link.
- 03:08:56 [lgonze]
- very nice.
- 03:09:00 [lgonze]
- how do you do that?
- 03:09:23 [ruaok]
- its just a form submit link that posts a form to the amazon web pages.
- 03:09:36 [ruaok]
- normally we'd earn money for it, but we can't right now.
- 03:10:02 [ruaok]
- We don;t have the non-profit status yet and we'd be violating the CCCP charter if we 'earned' money.
- 03:10:27 [jwatk03]
- jwatk03 has joined #musicbrainz
- 03:11:16 [lgonze]
- do you get the amazon ID by doing a text lookup at amazon?
- 03:11:33 [ruaok]
- yes, but we do that offline.
- 03:11:41 [ruaok]
- we have a matching script that runs daily.
- 03:11:54 [lgonze]
- well, it's quite a cool thing.
- 03:11:57 [ruaok]
- :-)
- 03:12:34 [lgonze]
- there is a sad thing about it, though, which is that if there is any money involved then you increase your legal risk. never mind if the point of it all is to help convert p2p sharing to money for the rights holders.
- 03:13:37 [ruaok]
- some money has legal risks -- not all forms of income.
- 03:13:55 [ruaok]
- This amazon thing is totally legit and we make money from it. May not be much, but all helps.
- 03:14:08 [lgonze]
- it's all good. brings cash in for everybody.
- 03:14:41 [lgonze]
- getting anywhere near the music industry, which is crazed with lawyers, is freaky.
- 03:15:31 [lgonze]
- webjay is in beta 1.
- 03:15:47 [lgonze]
- thinking about how to make money eventually...
- 03:15:59 [lgonze]
- that is: http://webjay.org.
- 03:17:18 [ruaok]
- ohh. interesting.
- 03:17:26 [ruaok]
- are you going to be around in 1-2 hours?
- 03:17:41 [ruaok]
- I want to chat about that and FOAF music profiles and portable playlists.
- 03:18:19 [lgonze]
- L)
- 03:18:26 [lgonze]
- all good things, but I'll be over on the sofa by then. it's 10pm here.
- 03:18:36 [ruaok]
- ah, right.
- 03:18:39 [ruaok]
- hmmm. one sec.
- 03:19:09 [lgonze]
- ian mentioned that conversation. I was happy to hear it.
- 03:19:46 [ruaok]
- ian -- exellent.
- 03:19:55 [ruaok]
- I didn't know you two knew each other.
- 03:20:10 [ruaok]
- OK, I shirked my responsibilities and I can chat.
- 03:20:23 [ruaok]
- how do you know ian?
- 03:20:45 [lgonze]
- awesome. that's cool.
- 03:20:55 [lgonze]
- he introduced himself via email.
- 03:21:04 [ruaok]
- cool.
- 03:21:07 [lgonze]
- a fellow maven of playlists...
- 03:21:17 [ruaok]
- excellent.
- 03:21:35 [lgonze]
- it's a good feeling to meet people with the same incredibly narrow interests.
- 03:21:41 [ruaok]
- lemme look at webjay real quick.
- 03:21:43 [ruaok]
- ding.
- 03:26:00 [ruaok]
- you need to mirror the MB database and host it locally and let people search for tracks to stuff into playlists
- 03:26:19 [lgonze]
- I've thought about that.
- 03:26:19 [ruaok]
- or does every playlist track have to have an accessible URL?
- 03:26:24 [lgonze]
- yes.
- 03:26:31 [ruaok]
- Ouch.
- 03:26:38 [lgonze]
- that's an odd application feature of webjay -- everything is on the public web.
- 03:26:45 [ruaok]
- Great for Creative Commons music.
- 03:27:09 [ruaok]
- not good for non-free music, unlike you like lawyers.
- 03:27:13 [lgonze]
- also for very old music, music from countries without strong copyright, electronica.
- 03:27:29 [lgonze]
- yes, it is not good for non-free music, but that's intentional.
- 03:28:11 [lgonze]
- once you hang around in the free (but obscure) music scene for a while you find a huge amount of new stuff. but you have to let your taste adjust.
- 03:28:11 [ruaok]
- I hear your motivations loud and clear.
- 03:28:28 [lgonze]
- exactly.
- 03:28:57 [ruaok]
- However your approach limits the scope of applicability, which will make it difficult to gain critical mass.
- 03:29:05 [lgonze]
- that's a big problem.
- 03:53:25 [lgonze]
- lgonze has quit
- 04:04:48 [jwatk03]
- jwatk03 has quit
- 04:13:52 [Knio1]
- Knio1 is now known as Knio
- 05:11:20 [elinenbe]
- elinenbe has joined #musicbrainz
- 05:23:40 [Necronom]
- Bummer.
- 05:24:24 [Necronom]
- 489/1261 of the albums I tagger are incomplete. :(
- 05:24:36 [Necronom]
- s/tagger/tagged/
- 05:26:20 [Necronom]
- 27 of those are AlbumIds that have gone away.
- 05:26:35 [ruaok]
- :-(
- 05:26:47 [ruaok]
- So, can I invite you to do something about that?
- 05:26:49 [ruaok]
- :-)
- 05:27:08 [ruaok]
- We sorely need a supercession table that keeps track of all this stuff.
- 05:27:28 [ruaok]
- some perl server hacking sound appealing?
- 05:27:43 [ruaok]
- how about if we throw in an unpaid trip to florida?
- 05:27:59 [ruaok]
- * ruaok needs more sleep or caffeine
- 05:29:05 [Necronom]
- Hmm. If I got a copy of the server running, I would need to abuse the main one doing bigass queries.
- 05:29:25 [Necronom]
- s/running/running locally/
- 05:29:49 [ruaok]
- er?
- 05:30:03 [ruaok]
- Yes, get a local copy running and then you have a sandbox to play in.
- 05:30:14 [ruaok]
- you can download a fully functioning dataset to tinker with.
- 05:30:21 [Necronom]
- Oooh. Supper.
- 05:30:21 [ruaok]
- we only strip private info.
- 05:30:36 [ruaok]
- food or excitement?
- 05:30:47 [ruaok]
- I'd prefer supper, really.
- 05:31:42 [ruaok]
- bbiab
- 05:40:32 [elinenbe]
- elinenbe has quit
- 05:41:17 [elinenbe]
- elinenbe has joined #musicbrainz
- 06:16:23 [Necronom]
- Nifty. I now have a perl script that tells me what tracks I am missing from my albums. :)
- 06:28:00 [Necronom]
- And what tracks got spurriously attributed to the wrong albums.
- 06:40:30 [Knio]
- Knio is now known as Knio-sleep
- 08:32:38 [orogor]
- orogor has joined #musicbrainz
- 11:23:24 [RJ]
- RJ has joined #musicbrainz
- 11:41:49 [Mutiny]
- Mutiny has joined #musicbrainz
- 11:50:31 [yusuke]
- yusuke has joined #musicbrainz
- 12:04:48 [mp3man]
- mp3man has joined #musicbrainz
- 12:05:17 [mp3man]
- mp3's?
- 12:15:04 [mp3man]
- mp3man has quit
- 13:23:51 [elinenbe__]
- elinenbe__ has joined #musicbrainz
- 13:38:33 [elinenbe]
- elinenbe has quit
- 13:38:33 [elinenbe__]
- elinenbe__ is now known as elinenbe
- 13:54:01 [wolfsong]
- wolfsong has joined #musicbrainz
- 14:50:17 [Knio-sleep]
- Knio-sleep is now known as Knio
- 15:28:23 [jwatk03]
- jwatk03 has joined #musicbrainz
- 15:28:51 [Knio]
- Knio is now known as Knio-school
- 15:38:47 [alus]
- alus has quit
- 15:38:47 [yusuke]
- yusuke has quit
- 15:38:59 [yusuke]
- yusuke has joined #musicbrainz
- 15:38:59 [alus]
- alus has joined #musicbrainz
- 15:58:47 [yalaforge]
- yalaforge has joined #musicbrainz
- 15:59:45 [yalaforge]
- wow. pretty crowded today
- 16:01:50 [wolfsong]
- Does MB Tagger actually overwrite other portions of the ID3 tag such as lyrics, pics, etc.?
- 16:02:32 [yalaforge]
- I'll check the tunepimp sources, wait a minute ...
- 16:02:56 [wolfsong]
- I noticed last night that some of the tracks i'd tagged after adding pics & lyrics to no longer had the lyrics and the pic was corrupted
- 16:03:20 [yalaforge]
- could very well be a bug.
- 16:03:38 [yalaforge]
- not necessarily in tunepimp/mb_tagger. perhaps the tag was corrupted right from the beginning
- 16:04:25 [elinenbe]
- elinenbe has quit
- 16:05:14 [elinenbe]
- elinenbe has joined #musicbrainz
- 16:05:16 [wolfsong]
- well it seems unlikely
- 16:05:23 [yalaforge]
- in tunepimp it is configurable if the old tag is cleared or not
- 16:05:35 [wolfsong]
- i created the MP3s with EAC
- 16:05:41 [wolfsong]
- then ran MBT
- 16:05:47 [yalaforge]
- I don't know EAC
- 16:05:51 [wolfsong]
- then added pics & lyrics
- 16:06:00 [yalaforge]
- using which tool?
- 16:06:02 [wolfsong]
- then reran MBT
- 16:06:22 [wolfsong]
- EAC - http://www.exactaudiocopy.de/
- 16:07:29 [wolfsong]
- MP3BookHelper - http://mp3bookhelper.sourceforge.net/
- 16:08:04 [yalaforge]
- which backend library does it use for tagging?
- 16:08:48 [yalaforge]
- heh, there must be a 'clearTags' setting in the preferences dialog of mb_tagger
- 16:09:28 [wolfsong]
- when I reran MBT I had 2 retag everything so I think there may be an issue where MP3BookHelper updates something but when I reran MP3BookHelper after retagging the corrupted pics showed up and the lyrics were gone
- 16:09:54 [wolfsong]
- there is but I don't have that selected
- 16:10:15 [yalaforge]
- you could have told me *before* I searched that in the source
- 16:10:23 [wolfsong]
- not sure about the tagging libray
- 16:10:37 [wolfsong]
- sorry
- 16:11:05 [wolfsong]
- i was still trying 2 explain the steps I took 2 reproduce the problem
- 16:12:32 [yalaforge]
- hm. I really don't think this is a problem with the tagger. libid3tag is pretty close to the standard
- 16:13:12 [yalaforge]
- if the lyrics or picture frames had some errors already it's possible, however, that libid3tag/mb_tagger messed them up
- 16:13:31 [wolfsong]
- how would i test that?
- 16:13:50 [yalaforge]
- using the standard and a hex editor ;-)
- 16:13:56 [yalaforge]
- or ask ruaok what to do
- 16:14:13 [elinenbe]
- elinenbe has quit
- 16:14:22 [elinenbe]
- elinenbe has joined #musicbrainz
- 16:14:37 [yalaforge]
- sorry, this isn't very helpful but id3v2 and all the implementations out there is a mess
- 16:14:39 [wolfsong]
- who's ruaok? is he the developer?
- 16:14:43 [yalaforge]
- yes
- 16:14:47 [wolfsong]
- ok
- 16:15:04 [wolfsong]
- looks like he's signed in but must be away
- 16:15:09 [yalaforge]
- yup
- 16:15:28 [wolfsong]
- yes the implementation stuff does seem to be a bit chaotic
- 16:15:48 [yalaforge]
- that's because id3v2 is too complex. nobody gets it right
- 16:16:21 [wolfsong]
- oic
- 16:16:30 [yalaforge]
- * yalaforge is away: I'm busy
- 16:20:21 [elinenbe]
- elinenbe has quit
- 16:20:47 [jwatk03]
- jwatk03 has quit
- 17:18:58 [Necronom]
- yalaforge,wolfsong; NB: tunepimp doesn't use the id3v2 reference library, it uses a derived version of the id3v2 library that comes with madplay.
- 17:19:15 [yalaforge]
- yup, I know
- 17:31:13 [sbw_]
- sbw_ has joined #musicbrainz
- 17:31:24 [sbw]
- sbw has quit
- 18:16:36 [ruaok]
- morning kids!
- 18:21:55 [yalaforge]
- * yalaforge is back (gone 02:05:23)
- 18:21:59 [yalaforge]
- hi
- 18:22:07 [ruaok]
- hi
- 18:22:25 [yalaforge]
- still worrying about the cheap little join?
- 18:23:01 [ruaok]
- LOL
- 18:23:14 [ruaok]
- I guess that says all about your view on the matter.
- 18:23:18 [yalaforge]
- :-)
- 18:23:57 [ruaok]
- kleinvieh macht auch mist.
- 18:24:03 [yalaforge]
- so isses
- 18:24:10 [Necronom]
- ruaok, is there an RFE open to have the tagger support "affinity
- 18:24:21 [Necronom]
- " to albums when matching tracks?
- 18:24:29 [ruaok]
- yes.
- 18:24:30 [yalaforge]
- the HTML pages join so much, I guess we should optimize on that front first
- 18:24:36 [ruaok]
- I'm keenly aware of the request. :-)
- 18:24:51 [Necronom]
- Heh. How were you planning on implementing it?
- 18:25:02 [ruaok]
- yalaforge: I dunno. We should see which gets more hits.
- 18:25:21 [ruaok]
- Necronom: I was going to read all the suggestions from mb-users and figure something out.
- 18:25:25 [ruaok]
- :0(
- 18:25:29 [ruaok]
- er
- 18:25:31 [ruaok]
- :-)
- 18:25:40 [yalaforge]
- if postgres supported proper hash indices, the joins would be a lot cheaper. oh well
- 18:25:41 [ruaok]
- any suggestions Necronom?
- 18:25:58 [Necronom]
- Glad you asked.. :)
- 18:26:51 [Necronom]
- If we make the assumption that people won't be tagging "duplicate" albums at the same time, we can probably figure out which album "fits" the tracks to be tagged better.
- 18:27:43 [ruaok]
- seems like a reasonable assumption.
- 18:27:55 [ruaok]
- I'm hoping for an automatic detection scheme
- 18:28:04 [Necronom]
- Mmm. Me too.
- 18:28:29 [ruaok]
- my first approach was going to take all the files that came in and read the metadata from them.
- 18:28:42 [ruaok]
- then feed all this data to a clustering algorithm of sorts.
- 18:29:02 [ruaok]
- then analyze clusters for 'albums'
- 18:29:07 [Necronom]
- Mmm. And iterate through possible album matches..
- 18:29:17 [ruaok]
- yup.
- 18:29:31 [ruaok]
- that could actually drive the recognition rates much higher.
- 18:29:48 [ruaok]
- it will certainly make the results more appealing.
- 18:30:22 [Necronom]
- The problems are going to be:
- 18:30:25 [ruaok]
- spooge: any chance of making some of your files available to me??
- 18:30:38 [Necronom]
- 1) If someone tries to tag their albums which include a "greatest hits" album.
- 18:30:55 [Necronom]
- 2) If someone has both a domestic and "import" album.
- 18:31:50 [Necronom]
- 3) tracks common to VA/regular album being tagged at the same time..
- 18:32:02 [ruaok]
- if the metadata for case #1 is ok, it should sort itself out.
- 18:32:21 [ruaok]
- #2 will be a good challenge. May not tackle that right away.
- 18:32:36 [ruaok]
- #3 yup.
- 18:32:41 [Necronom]
- A strong enough clustering algorithm should be able to handle all of that..
- 18:32:48 [ruaok]
- I'm thinking.
- 18:32:50 [Necronom]
- Or at least make reasonable guesses. :)
- 18:33:06 [ruaok]
- Plus we should encourage users to tag sets non-overlapping albums.
- 18:33:10 [ruaok]
- at the same time.
- 18:33:25 [ruaok]
- I see this happening in two stages:
- 18:33:26 [Necronom]
- Mmm.
- 18:33:35 [ruaok]
- 1. find albums
- 18:33:40 [Necronom]
- * Necronom usually just dumps a few thousand files in the tagger, and comes back a while later...
- 18:33:46 [ruaok]
- 2. match albums
- 18:34:13 [ruaok]
- and i think it is reasonable to say that if you do that you may get some overlap.
- 18:34:26 [ruaok]
- and may need to tweak your album groupings.
- 18:34:36 [Necronom]
- I wrote a perl script to do #1, and report albums with missing tracks, a "best fit" algorithm shouldn't be too hard..
- 18:34:57 [ruaok]
- unfortunately this is a big change and I may need to finish my grant work before attemping this.
- 18:35:11 [Necronom]
- :)
- 18:35:33 [Necronom]
- Maybe if I stop sleeping, I could write a prototype perl script. :)
- 18:37:51 [ruaok]
- heh.
- 18:38:10 [ruaok]
- I'd rather you take some server tasks on and free me of those so I can do tagger stuff.
- 18:38:57 [ruaok]
- * ruaok wonders if Necronom knows ruaok wants him to do some server development
- 18:39:58 [ruaok]
- brb
- 19:30:10 [ruaok]
- yalaforge: do you think you will have a chance to point wendell in the right direction with the report stuff??
- 19:30:43 [yalaforge]
- nope, I don't think so
- 19:30:55 [KyleB]
- ruaok, if you were writing a new tagger, would you use the call back or polling method in tp?
- 19:30:58 [yalaforge]
- subscriptions aren't shipped with our metadata
- 19:31:33 [yalaforge]
- if he wants the reports to be "personalized" (using the user's subscriptions) the report backend has to be changed
- 19:31:51 [ruaok]
- yalaforge: right. he doesn't need live data for that/.
- 19:31:59 [ruaok]
- he can setup his own subs and test with that
- 19:32:19 [ruaok]
- I was more asking you to tell him where in the server code to look at and how we have things setup
- 19:32:22 [ruaok]
- KyleB: polling
- 19:32:42 [yalaforge]
- wait a minute ...
- 19:32:51 [KyleB]
- ruaok, cool, that's what im using
- 19:34:05 [yalaforge]
- ruaok: I don't see how that would help him. Or does he want to patch mb_server?
- 19:35:08 [ruaok]
- he says: Hi Guys! Any chance of getting relevant duplicate suggestions somehow
- 19:35:08 [ruaok]
- connected to subscribed artists? Is there something I could do to help this
- 19:35:08 [ruaok]
- happen?
- 19:35:37 [ruaok]
- I'm not entirely certain I know what he wants -- I was hoping you had a better clue.
- 19:35:47 [yalaforge]
- nope, I don't
- 19:36:08 [yalaforge]
- but I think he means the report display routine. it should display the user's subscribed artists first or something
- 19:36:23 [ruaok]
- oh, I see.
- 19:36:29 [yalaforge]
- pretty useless, IMHO
- 19:36:41 [ruaok]
- so that involves futzing with an existing report?
- 19:36:48 [yalaforge]
- yes
- 19:37:03 [ruaok]
- ok, I'll drop him a line and tell him where to look.
- 19:37:10 [yalaforge]
- ok, thanks.
- 19:37:59 [djce]
- djce has joined #musicbrainz
- 19:38:29 [yalaforge]
- the display stuff is in mb_server/htdocs/reports/*.inc
- 19:39:22 [yalaforge]
- anyway. night!
- 19:39:25 [ruaok]
- nite!
- 19:39:26 [yalaforge]
- yalaforge has quit
- 19:39:26 [djce]
- bye
- 19:39:31 [ruaok]
- evening djce!
- 19:39:34 [djce]
- hi
- 19:39:50 [djce]
- (just as I was about to say hi to yalaforge... off he goes again!)
- 19:43:14 [sbw_]
- sbw_ has quit
- 20:10:29 [orogor_]
- orogor_ has joined #musicbrainz
- 20:15:55 [orogor]
- orogor has quit
- 20:16:17 [kgk]
- kgk has joined #musicbrainz
- 20:21:52 [djce]
- * djce is away: but will be back
- 20:24:59 [kgk]
- kgk has left #musicbrainz
- 20:29:34 [sbw]
- sbw has joined #musicbrainz
- 20:44:46 [RJ]
- RJ has quit
- 20:46:35 [sbw]
- sbw has quit
- 20:46:41 [sbw]
- sbw has joined #musicbrainz
- 21:01:29 [crw]
- crw has joined #musicbrainz
- 21:01:38 [crw]
- good afternoon, all
- 21:01:46 [ruaok]
- hey
- 21:02:02 [crw]
- i'm a new musicbrainz user. :)
- 21:02:07 [ruaok]
- welcome!
- 21:02:14 [crw]
- fed up with allmusic.com's timeouts and server overloads.
- 21:02:28 [crw]
- thanks :)
- 21:02:35 [ruaok]
- and closed nature, I hope. :-)
- 21:03:15 [crw]
- eh, as long as the info is relatively accurate. i've also used cddb to locate which album a track is on before looking it up on allmusic.
- 21:03:40 [crw]
- but the big question on my mind, is how do i get musicbrainz to show me which year the album was released?
- 21:04:14 [ruaok]
- we've just rolled out that feature, and thus we haven't collected much data yet.
- 21:04:35 [crw]
- aaah.
- 21:04:45 [crw]
- at least the functionality is there.
- 21:05:17 [ruaok]
- http://musicbrainz.org/album/eef23979-839b-43d9-b832-2c1233acda20.html
- 21:05:25 [ruaok]
- see that page. This album has the info.
- 21:05:33 [crw]
- groovy.
- 21:05:39 [ruaok]
- if you find an album that doesn't have it, please add it.
- 21:05:50 [ruaok]
- shouldn't be hard to find those. ;-)
- 21:05:56 [crw]
- * crw nods
- 21:06:14 [crw]
- looks like i'm stuck with allmusic for a while longer.
- 21:07:04 [real]
- doing dates now!?!
- 21:07:58 [ruaok]
- crw: don't give up on us yet.
- 21:08:10 [ruaok]
- We need your help to get the data fleshed out.
- 21:08:25 [crw]
- ruaok: pitching in even as we speak
- 21:09:14 [ruaok]
- sweet!
- 21:10:39 [crw]
- so uhh, which edit link do i hit to submit the year data?
- 21:11:17 [ruaok]
- what are your choices?
- 21:11:41 [crw]
- there are about 20 edit links on the page :P
- 21:11:52 [crw]
- http://musicbrainz.org/showalbum.html?albumid=38642
- 21:12:06 [crw]
- but none near where it says "no release data available"
- 21:12:14 [ruaok]
- are you logged in?
- 21:12:32 [crw]
- yeah, otherwise i wouldn't get the edit links
- 21:12:42 [ruaok]
- right.
- 21:12:42 [crw]
- though there is some sort of cookie issue with safari
- 21:12:44 [ruaok]
- erm. one sec.,
- 21:12:53 [ruaok]
- yes, there is.
- 21:13:04 [ruaok]
- Safari has a number of little niggling bugs.
- 21:13:11 [crw]
- * crw nods
- 21:13:12 [ruaok]
- I'm in the process of switching to firefox.
- 21:13:13 [soccos]
- soccos has joined #musicbrainz
- 21:13:16 [crw]
- it's still a 1.x browser
- 21:13:17 [ruaok]
- soccos: hey!
- 21:13:26 [crw]
- yeah, i should switch too
- 21:13:38 [ruaok]
- above the album there is an 'edit releases' button
- 21:13:45 [ruaok]
- that's the one you want.
- 21:14:07 [crw]
- okey doke.
- 21:14:13 [crw]
- lemme fire up firefox
- 21:16:00 [wolfsong]
- ruaok: i hear you're the one to pose MBT questions to
- 21:16:28 [wolfsong]
- does MBT do anything to the pics and lyrics portions of ID3 tags?
- 21:16:37 [soccos]
- ruoak: hey hey!
- 21:16:53 [ruaok]
- wolfsong: it shouldn't. does it?
- 21:17:07 [wolfsong]
- seems 2
- 21:17:25 [ruaok]
- i know that there is a bug there it messes up some tag types.
- 21:17:39 [wolfsong]
- i tagged an album & then used another app 2 add the pic lyrics
- 21:17:59 [ruaok]
- do you have an unadulterated file that you know would have issues when the tagger got done with it?
- 21:18:13 [wolfsong]
- i think it stripped the year out as well but i'm not @ home & can't remember
- 21:18:27 [ruaok]
- you could help me out a bit.
- 21:18:33 [wolfsong]
- hmmm not sure
- 21:18:43 [ruaok]
- If you could find 1 file that:
- 21:18:52 [wolfsong]
- i generally rip with EAC, tag with MTB & then add pics & lyrics
- 21:18:53 [ruaok]
- working fine
- 21:19:11 [ruaok]
- one file that is gotten garbled by the tagger or some other program that followed it.
- 21:19:17 [ruaok]
- preferrably the same file.
- 21:20:01 [wolfsong]
- so you want me 2 add pics & lyrics, copy that file & then tag it?
- 21:20:28 [ruaok]
- kinda.
- 21:20:42 [ruaok]
- first off pick a file that hasn't been touched by the tagger. save a clean copy.
- 21:21:00 [ruaok]
- go through your tagging process and determine that something in the file got borked.
- 21:21:08 [ruaok]
- send me both files
- 21:21:39 [ruaok]
- but only if you're certain that the problem occurred.
- 21:22:34 [crw]
- sweet. added my first bit 'o content
- 21:23:22 [ruaok]
- crw: thanks!
- 21:25:21 [crw]
- s'all good
- 21:26:57 [djce]
- * djce is back (gone 01:05:06)
- 21:27:12 [wolfsong]
- ruaok: i also have 2 interface suggestions
- 21:27:47 [wolfsong]
- 1st it would be nice to save only the selected tracks that have been identifyed
- 21:28:35 [wolfsong]
- 2nd it would be nice to have a collapsable 'browser' window without having to drag
- 21:28:46 [wolfsong]
- kinda like the sidebars in mozilla
- 21:29:44 [ruaok]
- we've got an RFE for #1...
- 21:29:53 [ruaok]
- I can see #2. I'll think about it.
- 21:30:08 [wolfsong]
- RFE?
- 21:31:21 [djce]
- request for enhancement
- 21:31:39 [wolfsong]
- oic
- 21:31:54 [wolfsong]
- where does one view them?
- 21:32:10 [djce]
- you know the "bugs" page?
- 21:32:21 [djce]
- go there, and then see the "RFE" link above the list of bugs
- 21:32:24 [ruaok]
- http://sourceforge.net/tracker/?group_id=19506&atid=369506
- 21:33:03 [wolfsong]
- ahhh
- 21:33:24 [wolfsong]
- i figured there was a sf page but hadn't found it yet
- 21:33:42 [wolfsong]
- once i found the wiki i figured there wasn't one
- 21:59:08 [ruaok]
- djce: what should we do about album release dates?
- 21:59:19 [djce]
- RDF you mean?
- 21:59:25 [ruaok]
- Making one change to the RDF stuff to return them is not a big deal.
- 21:59:38 [ruaok]
- Adding more stuff to the album meta table is a bit more invasive.
- 21:59:46 [ruaok]
- I was hoping to get this feature out on friday.
- 21:59:48 [djce]
- what more stuff do you mean?
- 22:00:04 [djce]
- albummeta already has firstreleasedate
- 22:00:06 [ruaok]
- adding the earliest release date/country to the albummeta table
- 22:00:12 [ruaok]
- oh.
- 22:00:17 [ruaok]
- really?
- 22:00:20 [djce]
- yup
- 22:00:28 [ruaok]
- you're a god.
- 22:00:45 [djce]
- ok, my Q now. :-)
- 22:00:58 [djce]
- I've got libfoo.so installed locally (in $HOME/lib)
- 22:01:08 [djce]
- how do I get gcc etc to "just find" it?
- 22:01:12 [ruaok]
- so, then I'd be inclined to send that for all album queries and then add a new query to get all the release dates. how does that sound?
- 22:01:22 [djce]
- that sounds good
- 22:01:37 [ruaok]
- add the path to /etc/ld.conf.so
- 22:01:47 [djce]
- no way of doing it as non-root?
- 22:02:02 [ruaok]
- that's one way.
- 22:02:04 [djce]
- I'm trying to avoid messing with root config for stuff which is meant to be local.
- 22:02:20 [ruaok]
- You can also set LD_CONFIG_SOMETHING env var to accomplish the same.
- 22:02:40 [ruaok]
- it gives the dynamic linker the path to search
- 22:02:53 [ruaok]
- lemme look for the real var name
- 22:03:12 [djce]
- LD_LIBRARY_PATH isn't it? that rings a bell.
- 22:03:25 [ruaok]
- yes!
- 22:03:39 [djce]
- hmm. that wasn't it then... I'll keep looking.
- 22:03:46 [ruaok]
- set that to include the path to your .so file and gcc/ld should find it.
- 22:03:59 [ruaok]
- You can also specify the .so file on the link line explicitly.
- 22:04:12 [ruaok]
- what is the problem? Gimme some more details!
- 22:05:18 [crw]
- ruaok: one thing i've noticed in the past hour or two is that if i do a quick search using artist and track strings, it more frequently times out
- 22:06:10 [ruaok]
- crw: djce is the main server guy -- he might be able to give you some insight into that.
- 22:06:27 [djce]
- be right with you..
- 22:06:35 [crw]
- ruaok: groovy, thanks
- 22:06:49 [crw]
- i may just be misusing the quick search function
- 22:07:47 [djce]
- crw: what search are you trying to do?
- 22:07:59 [djce]
- which page, which search form, which fields?
- 22:08:46 [crw]
- djce: any page, using the quick search in the sidebar.
- 22:08:56 [crw]
- i've tried two different searches
- 22:09:02 [djce]
- ah, well those are three separate forms
- 22:09:05 [crw]
- artist: clash track: london calling
- 22:09:19 [crw]
- gotcha, so i *was* misusing it. bugger.
- 22:09:19 [djce]
- you can quick-search for artist OR album OR track, but all separately
- 22:09:30 [crw]
- djce: that's not immediately apparent. :)
- 22:09:31 [djce]
- try the "tagger" search form on /search.html
- 22:09:37 [djce]
- no, I agree
- 22:09:40 [djce]
- :-(
- 22:11:43 [crw]
- any plans to change the quick search to AND's instead of OR's?
- 22:13:22 [djce]
- We'd certainly like to. But there are no plans to do that in a hurry, because it's hard.
- 22:13:39 [djce]
- However, the tagger search form (aka "Advanced" search) does something quite similar to that.
- 22:13:53 [crw]
- just takes an extra click to get there :P
- 22:14:56 [ruaok]
- ok. time for me to hide and get some stuff done. I'll be back in a couple of hours!
- 22:14:58 [ruaok]
- ruaok has left #musicbrainz
- 22:23:01 [orogor_]
- orogor_ has quit
- 22:32:11 [soccos]
- soccos has quit
- 22:32:40 [Knio-school]
- Knio-school is now known as Knio
- 22:51:35 [blizzow]
- blizzow has joined #musicbrainz
- 23:02:54 [fish1209]
- fish1209 has joined #musicbrainz
- 23:03:04 [fish1209]
- fish1209 has left #musicbrainz
- 23:10:06 [jwatk03]
- jwatk03 has joined #musicbrainz
- 23:49:26 [wolfsong13]
- wolfsong13 has joined #musicbrainz
- 23:50:51 [wolfsong]
- wolfsong has quit
- 23:50:51 [wolfsong13]
- wolfsong13 is now known as wolfsong