|
Re: How to recognize homebrew discids
On 1/04/2009 4:52 a.m., Brian Schweitzer wrote:
I've been ignoring this thread because Disc IDs are useless to me - I
use PUIDs to identify my digital music.
I agree with Andrew's sentiments above. The odd homebrew Disc ID
isn't going to hurt the database. As long as the track times are
reasonably close to the correct times, who cares. There are better
places our time could be spent than weeding out "homebrew" Disc IDs.
I think this touches on the only point where I would ever care about a
homebrew. If it's within 2 or 4 seconds, on all tracks, I don't see a
reason to remove it. However, if it's way off in some regards - not 2
or 4 seconds, but say 20 seconds too long on the final track, or they
cut a silence-hidden track off and made it a separate track (making it
look like it should be a whole new release), then I would support
removing it, as it's then falsely indicating other things (a different
mastering, a different pressing, etc.), and can lead to editors
searching for data on releases or release events which don't actually
exist.
Brian
Then let's all take these sentiments and instead of merely abstaining,
start voting against such
removals, as I and a couple of others have
been doing for a few months now.
If enough of us are regularly voting against the ones we come across,
people may stop removing them because of the hassle. (alternatively,
they may just organise like-minders to quick-vote their edits, but I'd
like to think that wouldn't happen...!)
http://musicbrainz.org/mod/search/results.html?mod_status=1&automod=0&mod_type=20&moderator_id=137798&moderator_type=0&voter_type=0&voter_id=137798&vote_cast=-2&artist_type=0&orderby=asc&minid=&maxid=&isreset=0
Chad
_______________________________________________
MusicBrainz-users mailing list
MusicBrainz-users@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users
|
|
Re: How to recognize homebrew discids
On Tue, Mar 31, 2009 at 02:58:35PM -0500, Tekno Junky wrote:
> It would seem to be, that the best solution would be to have some sort of
> 'home-brew detection' built in to the clients that submit the disc id's,
> and/or the server when it is submitted.
>
> Even if home-brew detection is not fool-proof, it could simply provide a
> warning and message that home-brew/burned disc id's are not desired for
> music-brainz and for the user to confirm that the disc id is from a real cd.
media detection would be neat, but definitely not 100% accurate. I buy
physical cds from certain netlabels occassionally, these are often
released in such small numbers that they're on CD-R. Ofcourse, the
people dealing in such releases probably are aware of this and would
know to click away any warnings. Just wanted to mention it for
completenesse sake :)
-- kuno / warp.
_______________________________________________
MusicBrainz-users mailing list
MusicBrainz-users@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users
|
|
Re: How to recognize homebrew discids
|
|
Re: How to recognize homebrew discids
2009/4/1 Kuno Woudt < kuno@...>:
> media detection would be neat, but definitely not 100% accurate. I buy
> physical cds from certain netlabels occassionally, these are often
> released in such small numbers that they're on CD-R. Ofcourse, the
> people dealing in such releases probably are aware of this and would
> know to click away any warnings. Just wanted to mention it for
> completenesse sake :)
I don't think there are multiple pressings (c.q. different disc ids)
for these releases? Then there won't be any warnings to click away.
Jan
_______________________________________________
MusicBrainz-users mailing list
MusicBrainz-users@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users
|
|
Re: How to recognize homebrew discids
|
|
Re: How to recognize homebrew discids
On Wed, Apr 1, 2009 at 12:00 AM, Jan van Thiel < janvanthiel@...> wrote:
> 2009/4/1 Kuno Woudt < kuno@...>:
>> media detection would be neat, but definitely not 100% accurate. I buy
>> physical cds from certain netlabels occassionally, these are often
>> released in such small numbers that they're on CD-R. Ofcourse, the
>> people dealing in such releases probably are aware of this and would
>> know to click away any warnings. Just wanted to mention it for
>> completenesse sake :)
>
> I don't think there are multiple pressings (c.q. different disc ids)
> for these releases? Then there won't be any warnings to click away.
I think he was referring to media detection, not multiple DiscIDs. If
Picard say checked for CD-Rs before submitting a DiscID.
-Mark / InSinU8
_______________________________________________
MusicBrainz-users mailing list
MusicBrainz-users@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users
|
|
Re: How to recognize homebrew discids
2009/4/1 Kuno Woudt < kuno@...>:
> On Tue, Mar 31, 2009 at 02:58:35PM -0500, Tekno Junky wrote:
>> It would seem to be, that the best solution would be to have some sort of
>> 'home-brew detection' built in to the clients that submit the disc id's,
>> and/or the server when it is submitted.
>>
>> Even if home-brew detection is not fool-proof, it could simply provide a
>> warning and message that home-brew/burned disc id's are not desired for
>> music-brainz and for the user to confirm that the disc id is from a real cd.
>
> media detection would be neat, but definitely not 100% accurate. I buy
> physical cds from certain netlabels occassionally, these are often
> released in such small numbers that they're on CD-R. Ofcourse, the
> people dealing in such releases probably are aware of this and would
> know to click away any warnings. Just wanted to mention it for
> completenesse sake :)
>
> -- kuno / warp.
>
>
> _______________________________________________
> MusicBrainz-users mailing list
> MusicBrainz-users@...
> http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users>
True, I realised this after my initial mail. If you look through
discogs, there are a number of legitimate promo. releases on CD-R for
example.
--
Andrew :-)
Free Java Software Engineer
Red Hat, Inc. ( http://www.redhat.com)
Support Free Java!
Contribute to GNU Classpath and the OpenJDK
http://www.gnu.org/software/classpathhttp://openjdk.java.netPGP Key: 94EFD9D8 ( http://subkeys.pgp.net)
Fingerprint: F8EF F1EA 401E 2E60 15FA 7927 142C 2591 94EF D9D8
_______________________________________________
MusicBrainz-users mailing list
MusicBrainz-users@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users
|
|
Re: How to recognize homebrew discids
Then let's all take these sentiments and instead of merely abstaining,
start voting against such
removals, as I and a couple of others have
been doing for a few months now.
If enough of us are regularly voting against the ones we come across,
people may stop removing them because of the hassle. (alternatively,
they may just organise like-minders to quick-vote their edits, but I'd
like to think that wouldn't happen...!)
I admit, in the past, I've been convinced to vote yes on such edits. But I think it's been a while since I've done so. However, I would not vote no to such, nor yes, at this point, without some at least semi-official guideline to point to. My sense is that things like this shouldn't be voted no to just based on "this is my opinion of this type of edit", but rather, should have some guidelines basis for that no vote.
There's a small number of other types of edits I'm not too fond of, but they have at least have some basis in guidelines to support them, so I don't vote against them. Isn't that - whether or not to vote down any edit - at least the voting half of why we have guidelines, so that people vote based on a ruleset, and not just personal opinion?
Brian
_______________________________________________
MusicBrainz-users mailing list
MusicBrainz-users@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users
|
|
Re: How to recognize homebrew discids
On 2/04/2009 4:58 p.m., Brian Schweitzer wrote:
> There's a small number of other types of edits I'm not too fond of,
> but they have at least have some basis in guidelines to support them,
> so I don't vote against them. Isn't that - whether or not to vote
> down any edit - at least the voting half of why we have guidelines, so
> that people vote based on a ruleset, and not just personal opinion?
>
> Brian
If I think if there was consensus on this we'd be able to have a
guideline; but then we wouldn't have anyone making such edits so it's
catch 22. There is no consensus, so there is no guideline... so I don't
see that we have any other choice but to vote by opinion?
Chad
_______________________________________________
MusicBrainz-users mailing list
MusicBrainz-users@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users
|
|
Re: How to recognize homebrew discids
Agreed there's a catch-22. But, is the solution really, though, to initiate no-vote campaigns on a one-by-one basis in various edits, where some will pass and some won't, rather than trying to get an RFC passed?
On Thu, Apr 2, 2009 at 12:39 PM, Chad Wilson <chad.wilson@...> wrote:
On 2/04/2009 4:58 p.m., Brian Schweitzer wrote:
> There's a small number of other types of edits I'm not too fond of,
> but they have at least have some basis in guidelines to support them,
> so I don't vote against them. Isn't that - whether or not to vote
> down any edit - at least the voting half of why we have guidelines, so
> that people vote based on a ruleset, and not just personal opinion?
>
> Brian
If I think if there was consensus on this we'd be able to have a
guideline; but then we wouldn't have anyone making such edits so it's
catch 22. There is no consensus, so there is no guideline... so I don't
see that we have any other choice but to vote by opinion?
Chad
_______________________________________________
MusicBrainz-users mailing list
MusicBrainz-users@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users
|
|
Re: How to recognize homebrew discids
2009/4/2 Brian Schweitzer < brian.brianschweitzer@...>:
> Agreed there's a catch-22. But, is the solution really, though, to initiate
> no-vote campaigns on a one-by-one basis in various edits, where some will
> pass and some won't, rather than trying to get an RFC passed?
What would the RFC say? "Don't delete stuff unless you 100% sure it's
incorrect."? Isn't that obvious even without a style guideline?
Lukas
_______________________________________________
MusicBrainz-users mailing list
MusicBrainz-users@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users
|
|
Re: How to recognize homebrew discids
Lukáš Lalinský wrote:
> 2009/4/2 Brian Schweitzer < brian.brianschweitzer@...>:
>
>> Agreed there's a catch-22. But, is the solution really, though, to initiate
>> no-vote campaigns on a one-by-one basis in various edits, where some will
>> pass and some won't, rather than trying to get an RFC passed?
>>
>
> What would the RFC say? "Don't delete stuff unless you 100% sure it's
> incorrect."? Isn't that obvious even without a style guideline?
>
It could remove the references to not adding homebrew discids from the
wiki and/or explicitly allow homebrew ids.
I don't think the 100% thing is that obvious; I know I've trusted a few
editors when they say they're removing a release because they can't find
any proof for it.
- Si
_______________________________________________
MusicBrainz-users mailing list
MusicBrainz-users@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users
|
|
Re: How to recognize homebrew discids
On Thu, 02 Apr 2009 21:20:21 +0200, Lukáš Lalinský < lalinsky@...> wrote:
> What would the RFC say? "Don't delete stuff unless you 100% sure it's
> incorrect."? Isn't that obvious even without a style guideline?
"Suspected homebrew is not grounds for DiscID removal." or something to that effect.
Personally I would've liked to see Picard do media checks, if at all possible. Homebrew duplicates isn't a particularly big problem (and removing them is mostly a cosmetic issue IMO), but it's a bit iffy when they're submitted as the first id, or with a new release; the end result may be a borked tracklist, and unless someone drops by with definite proof of validity and does an extensive search to invalidate the old one, we're going to have false data sitting around that is very hard to get rid of.
You might say this isn't a particularly big problem, and I would agree, but IMO it's the only real problem with homebrews. A dupe doesn't really do any harm, it's just a bit of clutter that doesn't actually do anything.
--
Per // Wizzcat
_______________________________________________
MusicBrainz-users mailing list
MusicBrainz-users@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users
|
|
Re: How to recognize homebrew discids
I wouldn't want Picard or any other tool to be doing media checks and blocking adding Disc IDs only because it thought a CD was a homebrew. It could be wrong, and even if it were a CD-R, see the various points in this thread as to why some releases are official and yet also CD-Rs.
On Thu, Apr 2, 2009 at 3:44 PM, Per Øyvind Øygard <peroyo@...> wrote:
On Thu, 02 Apr 2009 21:20:21 +0200, Lukáš Lalinský < lalinsky@...> wrote:
> What would the RFC say? "Don't delete stuff unless you 100% sure it's
> incorrect."? Isn't that obvious even without a style guideline?
Personally I would've liked to see Picard do media checks, if at all possible. Homebrew duplicates isn't a particularly big problem (and removing them is mostly a cosmetic issue IMO), but it's a bit iffy when they're submitted as the first id, or with a new release; the end result may be a borked tracklist, and unless someone drops by with definite proof of validity and does an extensive search to invalidate the old one, we're going to have false data sitting around that is very hard to get rid of.
You might say this isn't a particularly big problem, and I would agree, but IMO it's the only real problem with homebrews. A dupe doesn't really do any harm, it's just a bit of clutter that doesn't actually do anything.
--
Per // Wizzcat
_______________________________________________
MusicBrainz-users mailing list
MusicBrainz-users@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users
|
|
Re: How to recognize homebrew discids
On Thu, 02 Apr 2009 22:34:26 +0200, Brian Schweitzer < brian.brianschweitzer@...> wrote:
> I wouldn't want Picard or any other tool to be doing media checks and
> blocking adding Disc IDs only because it thought a CD was a homebrew.
> It could be wrong, and even if it were a CD-R, see the various points in
> this thread as to why some releases are official and yet also CD-Rs.
Of course it should never block anything, I'm more than aware of the amount of legit CD-Rs (not that I've seen m(any) DiscIDs for such releases). But I'm quite sure a "did you burn this yourself!?" would do wonders. It's quite possible it's more work than it's worth though.
It's either that or we just forget about the entire issue as far as I can see; which admittedly would save everyone some time, at the possible cost of confusion down the road.
(Or wait, I've actually added a promo CD-R myself for an itunes-only release; God I loathe digital-only releases)
--
Per // Wizzcat
_______________________________________________
MusicBrainz-users mailing list
MusicBrainz-users@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users
|
|
Re: How to recognize homebrew discids
Some parts of this message have been removed.
Learn more about Nabble's security policy.
Nothing was said about blocking disc
id's, it is about recognizing and warning the user and making them confirm they
are submitting a production disc.
Like how when importing releases from
cddb, if a similar release is detected it asks the user to confirm.
I think there is sufficient
acceptance that some real releases are actual cd-r... demo's, dj samplers,
limited editions, etc so globally denying or removing them is not a
solution.
Another idea, if some type of
'home-brew' detection is viable on the server side, would be to highlight or
otherwise indicate suspicious (existing) disc id's (while still allowing
them).
----- Original Message -----
Sent: Thursday, April 02, 2009 3:34
PM
Subject: Re: [mb-users] How to recognize
homebrew discids
I wouldn't want Picard or any other tool to be doing media
checks and blocking adding Disc IDs only because it thought a CD was a
homebrew. It could be wrong, and even if it were a CD-R, see the various
points in this thread as to why some releases are official and yet also
CD-Rs.
On Thu, Apr 2, 2009 at 3:44 PM, Per Øyvind Øygard <peroyo@...>
wrote:
On Thu, 02 Apr 2009 21:20:21 +0200, Lukáš Lalinský < lalinsky@...>
wrote: > What would the RFC say? "Don't delete stuff unless you
100% sure it's > incorrect."? Isn't that obvious even without a style
guideline?
Personally I would've liked to
see Picard do media checks, if at all possible. Homebrew duplicates isn't a
particularly big problem (and removing them is mostly a cosmetic issue IMO),
but it's a bit iffy when they're submitted as the first id, or with a new
release; the end result may be a borked tracklist, and unless someone drops
by with definite proof of validity and does an extensive search to
invalidate the old one, we're going to have false data sitting around that
is very hard to get rid of.
You might say this isn't a particularly
big problem, and I would agree, but IMO it's the only real problem with
homebrews. A dupe doesn't really do any harm, it's just a bit of clutter
that doesn't actually do anything.
--
Per //
Wizzcat
_______________________________________________
MusicBrainz-users
mailing
list
MusicBrainz-users@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users
_______________________________________________
MusicBrainz-users mailing list
MusicBrainz-users@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users
|
|
Re: How to recognize homebrew discids
I think that wouldn't be that good. Legal releases on CD-r, demos and other stuff would highlight either, making them some sort of suspicious releases too. Submitting warning would be enough in this case I think.
2009/4/3 Tekno Junky <teknojnky@...>
Another idea, if some type of
'home-brew' detection is viable on the server side, would be to highlight or
otherwise indicate suspicious (existing) disc id's (while still allowing
them).
_______________________________________________
MusicBrainz-users mailing list
MusicBrainz-users@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users
|
|
Re: How to recognize homebrew discids
|
|
Re: How to recognize homebrew discids
2009/4/2 Tekno Junky <teknojnky@...>
Another idea, if some type of
'home-brew' detection is viable on the server side, would be to highlight or
otherwise indicate suspicious (existing) disc id's (while still allowing
them).
A "home-brew" detection on the server side is not possible, at least not with the TOC data currently available. It's exactly what's making things bad here: People try to do home-brew detection just based on the TOC data (that is on the track length), and that's not reliable at all.
I would support removing that paragraph about not to add home-brew discs.
-- Philipp Wolfer
_______________________________________________
MusicBrainz-users mailing list
MusicBrainz-users@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users
|
|
Re: How to recognize homebrew discids
Some parts of this message have been removed.
Learn more about Nabble's security policy.
Now we have two different things into the discussion:
1. Do we want to allow people to delete suspected homebrew discID's
2. Do we want to have homebrew discID's
On the first point there seems to be a consensus that we do not want to delete any discID's,
as we can never be 100% certain whether it is a homebrew or not.
As for the second point, I am agains allowing homebrew discID's,
it would result in hundreds of discID's for popular releases,
most of which are only usefull for the person who added it.
Musicbrainz does also not allow homemade compilations for exactly this reason,
any data in the database should be relevant for other people too.
Bram / jongetje
Date: Fri, 3 Apr 2009 16:45:37 +0200 From: ph.wolfer@...To: musicbrainz-users@...Subject: Re: [mb-users] How to recognize homebrew discids
2009/4/2 Tekno Junky <teknojnky@...>
Another idea, if some type of 'home-brew' detection is viable on the server side, would be to highlight or otherwise indicate suspicious (existing) disc id's (while still allowing them).
A "home-brew" detection on the server side is not possible, at least not with the TOC data currently available. It's exactly what's making things bad here: People try to do home-brew detection just based on the TOC data (that is on the track length), and that's not reliable at all.
I would support removing that paragraph about not to add home-brew discs.
-- Philipp Wolfer
Express yourself instantly with MSN Messenger! MSN Messenger
_______________________________________________
MusicBrainz-users mailing list
MusicBrainz-users@...
http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users
|
