Ok to sum up: I hear some voices against incomplete edits, the (only?) argument seems to be a waiting period when adding more info to those releases. But there is no offical policy on that topic. Correct?
2014-01-13 01321, 2014
hawke1
Correct.
2014-01-13 01318, 2014
nikki
incomplete tracklists, you mean.
2014-01-13 01334, 2014
_flow_
I am not conviced that they are really that bad and should be disallowed. Like CallerNo6 said "an edit that improves the quality and breadth of the database is a good edit" But well, that's what we have the voting system for. :)
2014-01-13 01340, 2014
Freso
If adding something known to be incomplete, noting it in the annotation or, at the very least, edit note would probably be a decent thing to do, so people investigating it in the future can tell off-the-bat that it's incomplete.
2014-01-13 01344, 2014
CallerNo6
Well, the official policy is that voters ultimately decide what does/doesn't improve the quality of the database.
2014-01-13 01344, 2014
hawke1
_flow_: I would say it’s fine as long as you have the right number of tracks.
2014-01-13 01304, 2014
_flow_
s/incomplete edits/incomplete tracklists
2014-01-13 01306, 2014
hawke1
If you don’t, it’s incorrect as well as incomplete.
2014-01-13 01310, 2014
nikki
the question is whether an incorrect tracklist is an improvement or not :)
2014-01-13 01330, 2014
hawke1
IMO, if there are placeholders for unknown tracks, that’s acceptable.
2014-01-13 01338, 2014
CallerNo6
A few years ago, there was a movement to delete classical releases that were so incomplete as to be un-identifiable. I agreed with those removals. Not everybody did.
2014-01-13 01304, 2014
reodroid joined the channel
2014-01-13 01345, 2014
CallerNo6
Now I'm wondering if the same editors who opposed the removals would have voted 'yes' on the original adds.
2014-01-13 01308, 2014
Jozo joined the channel
2014-01-13 01331, 2014
jesus2099
bye bye
2014-01-13 01327, 2014
Freso joined the channel
2014-01-13 01357, 2014
Hadora joined the channel
2014-01-13 01337, 2014
Jozo has left the channel
2014-01-13 01303, 2014
JoeLlama joined the channel
2014-01-13 01356, 2014
Mineo joined the channel
2014-01-13 01348, 2014
reodroid joined the channel
2014-01-13 01354, 2014
zas
nikki: incorrect tracklists are what we have mostly (incorrect titles, incorrect durations, incorrect order, ...), when a release is very bad, and nobody bothers to fix it, it disappears at some point, if a release is interesting enough, people will fix it, and it will become something better
2014-01-13 01307, 2014
reosarevok
CallerNo6: almost certainly not (re: vote yes on adds)
2014-01-13 01323, 2014
reosarevok
zas: that's very optimistic, but ok
2014-01-13 01345, 2014
reosarevok
(we still have looots of 10-year old very bad releases :p)
2014-01-13 01308, 2014
zas
reosarevok: sure, can we do anything about it ?
2014-01-13 01326, 2014
zas
more active editors will do for sure
2014-01-13 01338, 2014
reosarevok
Not really, just saying "it disappears" isn't true, it stays there looking evily at you :)
2014-01-13 01354, 2014
reosarevok
I don't think it's a huge problem though
2014-01-13 01309, 2014
zas
well, we (editors) suggest to merge or delete releases on regular basis
2014-01-13 01335, 2014
zas
the thing that isn't working imho, is the quality indicator
2014-01-13 01348, 2014
nikki
and people add new crap on a regular basis :P
2014-01-13 01356, 2014
zas
but the question was more about "release with _optional_ tracklist"
2014-01-13 01358, 2014
zas
entering data in MB is very time-consuming, and being able to divide the time is a must imho
2014-01-13 01303, 2014
zas
let me explain
2014-01-13 01300, 2014
xhienne joined the channel
2014-01-13 01333, 2014
zas
when i receive a new CD from some obscure label, i have to enter the new release (and release group), often the band too, and links and infos about them, and label, and tracklist, dates, isrcs, acoustids, cover art scans (damn a 52-pages booklet), etc...
2014-01-13 01357, 2014
zas
so i can't do it in one time, i have to do it steps by steps
2014-01-13 01322, 2014
zas
i will enter release info, then scans cd for discid and isrcs, then extract audio, calculate acoustids, and next day search info on label, complete artist links, and next week scan cover art, etc...
2014-01-13 01320, 2014
zas
if i can't split those tasks, i will skip some parts, or just not use MB, simple as that, not because of "lazyness" but instead because the fact time is limited for any task
2014-01-13 01311, 2014
simukis_
↑ and that gets worse with compilations of obscure artists’ songs.
2014-01-13 01314, 2014
zas
so if someone wants to enter part of data concerning a release, i see no reason to prevent him to do so, but we can still require that data to be correct
2014-01-13 01326, 2014
zas
i hate entering compilations....
2014-01-13 01354, 2014
zas
excellent exemple as i skipped few in my collection for this reason
2014-01-13 01324, 2014
zas
just looking the back cover, and "later, much later, ... or never" appeared in my mind
2014-01-13 01347, 2014
zas
because the process _is_ time-consuming
2014-01-13 01343, 2014
srotta joined the channel
2014-01-13 01350, 2014
zas
still i try to do a good job when entering a new release (each scan is cropped, rotated, etc...), each bit of info checked if possible, etc...
2014-01-13 01342, 2014
xhienne
CallerNo6: this interesting behviour can be observed on Wikipedia as well. A few days ago, I came across an article which was 90% made by an anonymous users. It didn't contain any reference whatsoever. Recently, another anonymous added an information that was already on the page, and corrected (with reason) a date elsewhere. They were quickly reverted with the reason that... the change was not sourced.
2014-01-13 01329, 2014
CallerNo6
One reasonable test of "quality" might be "would it take less time to add this properly than to identify, track down and fix the existing entry?"
2014-01-13 01353, 2014
zas
CallerNo6: i agree
2014-01-13 01314, 2014
zas
some entries are such a mess than no one will ever fix them
2014-01-13 01338, 2014
CallerNo6
xhienne: I should do some wikipedia editing for comparison.
2014-01-13 01334, 2014
xhienne
I bet you would be disappointed
2014-01-13 01347, 2014
CallerNo6
hmm. I think I just shifted the task from "define: quality" to "define : properly"
2014-01-13 01301, 2014
xhienne
if you define "properly" as "matching our quality guidelines", you can go back defining "quality"
2014-01-13 01325, 2014
hawke1
I don’t think defining quality is really relevant as long as we can’t really see any effect of setting quality
2014-01-13 01302, 2014
CallerNo6
hawke1: I'm only talking about it in the "does this edit improve the quality of the database?" sense
2014-01-13 01341, 2014
CallerNo6
(not sure how that would overlap with 'release quality', if at all)
2014-01-13 01334, 2014
xhienne
maybe reliability would be a better term?
2014-01-13 01327, 2014
xhienne
or comprehensiveness?
2014-01-13 01304, 2014
xhienne
What am I supposed to do when I am facing a mistake in the track list of an album? Actually, this is a compilation where the artist has been replaced with the artists's album from which the track was borrowed. Should I simply credit the artist alone? Or should credit him with "as <album title>" ?
2014-01-13 01326, 2014
hawke1
xhienne: Assuming this is a concrete example, you should credit the artist alone…but I can’t really say why other than that the artist credit field is for artists.
2014-01-13 01309, 2014
xhienne
Fine, I'll do it that way (yes, this is a concrete example)
2014-01-13 01309, 2014
hawke1
If there is an error (like they credited the wrong artist), that should be corrected., of course.
2014-01-13 01357, 2014
CatQuest joined the channel
2014-01-13 01358, 2014
xhienne
ok, thanks
2014-01-13 01316, 2014
jcazevedo joined the channel
2014-01-13 01345, 2014
Jozo joined the channel
2014-01-13 01301, 2014
Flip_ joined the channel
2014-01-13 01319, 2014
Flip_
Um... just a question about the database dump?
2014-01-13 01326, 2014
Flip_
Is there some way to just import it as a PostrgreSQL database -- without having to run the whole server et. al.?
2014-01-13 01344, 2014
Flip_
If so, how? If not, why not?
2014-01-13 01334, 2014
Jozo
Flip_: Yeah, just follow README
2014-01-13 01346, 2014
chirlu`
Yes, but you need scripts distributed with the server.
Question, though: I'm on Windows, with Cygwin right now. It /looks/ like it should work all the same, but do you happen to know if there's anything I should look out for?
2014-01-13 01335, 2014
Flip_
(Also, why is it so complicated? Why not just have a download that can be imported normally?)
2014-01-13 01351, 2014
Jozo has left the channel
2014-01-13 01333, 2014
zas joined the channel
2014-01-13 01337, 2014
nikki
Flip_: by "a download that can be imported normally", do you mean the output of pgdump? (or whatever the command is)
