Discussion:
[mb-style] Re: Musicbrainz-style Digest, Vol 34, Issue 86
(too old to reply)
Brian Schweitzer
2008-02-27 00:02:30 UTC
Permalink
b) is simply "Allegro"?
Show me one release where a track is identified as "Allegro", without
mentioning "Symphony No. XY" and the composer's name on the front cover.
c) is in multiple languages, or the same release has been reissued
multiple times with different liners? (very common, in fact, the
majority)
http://cover-paradies.to/?Module=ViewElement&ID=40762
Nothing specific to classical.
If I gave you just one, you'd assume I cherry picked it. Here's a
dozen or so, just pulling from the first 30 on my "messy releases
list". As for languages, I never claimed it to be. However, it's
very much a classical problem more so than other types of releases.
I'd rather not re-lay it all out again; I suggest instead you read the
lengthy email where I laid it all out and broke it down, and pointed
out why this is primarily a classical issue, and quite an issue within
classical, sent to this list a few weeks back.

Brian

http://musicbrainz.org/show/release/?releaseid=57554

http://musicbrainz.org/show/release/?releaseid=60836
"Contredanse in C" - which of the 100+?
"Andantino (from Concerto in C)" - could be any of 50 movements/works

http://musicbrainz.org/show/release/?releaseid=84582
"Organ Music"? "The Toy Symphony II "? "The Toy Symphony III "?

http://musicbrainz.org/show/release/?releaseid=80854
Which movement of EKN? 5:49 isn't long enough to hold more than one movement

http://musicbrainz.org/show/release/?releaseid=84067
"Adagio & Fugue" "Theme & 5 Variations" "Adagio & Allegro

http://musicbrainz.org/show/release/?releaseid=75640
The 6:31 "Don Giovanni" :D

http://musicbrainz.org/show/release/?releaseid=77979
Each of these 3 could be any of a dozen or two works.

http://musicbrainz.org/show/release/?releaseid=82741
Not just 1, but 9 tempo-only-no-work-ID tracks

http://musicbrainz.org/show/release/?releaseid=87415

http://musicbrainz.org/show/release/?releaseid=79928

http://musicbrainz.org/show/release/?releaseid=88769
"Floetenrondo D-Dur"

And best for last,
http://musicbrainz.org/show/release/?releaseid=19236
The 63:56 "Canon in D"
Leiv Hellebo
2008-02-27 02:36:20 UTC
Permalink
Post by Brian Schweitzer
b) is simply "Allegro"?
Show me one release where a track is identified as "Allegro", without
mentioning "Symphony No. XY" and the composer's name on the front cover.
c) is in multiple languages, or the same release has been reissued
multiple times with different liners? (very common, in fact, the
majority)
http://cover-paradies.to/?Module=ViewElement&ID=40762
Nothing specific to classical.
If I gave you just one, you'd assume I cherry picked it. Here's a
dozen or so, just pulling from the first 30 on my "messy releases
list". As for languages, I never claimed it to be. However, it's
very much a classical problem more so than other types of releases.
I'd rather not re-lay it all out again; I suggest instead you read the
lengthy email where I laid it all out and broke it down, and pointed
out why this is primarily a classical issue, and quite an issue within
classical, sent to this list a few weeks back.
Brian
http://musicbrainz.org/show/release/?releaseid=57554
http://www.amazon.com/Mozart-Relaxation-Richard-Stoltzman/dp/B00000I9M0
Post by Brian Schweitzer
http://musicbrainz.org/show/release/?releaseid=60836
"Contredanse in C" - which of the 100+?
"Andantino (from Concerto in C)" - could be any of 50 movements/works
http://www.amazon.com/Essential-Mozart-His-Greatest-Masterpieces/dp/B00005A8JZ
Post by Brian Schweitzer
http://musicbrainz.org/show/release/?releaseid=84582
"Organ Music"? "The Toy Symphony II "? "The Toy Symphony III "?
http://www.amazon.com/Mozart-Effect-Babies-Playtime-Sleepytime/dp/B00000DA1O
Post by Brian Schweitzer
http://musicbrainz.org/show/release/?releaseid=80854
Which movement of EKN? 5:49 isn't long enough to hold more than one movement
My money's on the first.
And it's the first mvt of Symph. No. 40
And first mvt of Concerto No. 17.
Post by Brian Schweitzer
http://musicbrainz.org/show/release/?releaseid=84067
"Adagio & Fugue" "Theme & 5 Variations" "Adagio & Allegro
http://www.amazon.com/The-Mozart-Album/dp/B0013AZN4E
Post by Brian Schweitzer
http://musicbrainz.org/show/release/?releaseid=75640
The 6:31 "Don Giovanni" :D
Overture:
http://www.amazon.com/Mozart-Dummies-Jack-Brymer/dp/B000002S8Z
Post by Brian Schweitzer
http://musicbrainz.org/show/release/?releaseid=77979
Each of these 3 could be any of a dozen or two works.
http://musicbrainz.org/show/release/?releaseid=82741
Not just 1, but 9 tempo-only-no-work-ID tracks
tough one, we'wouldn't let this pass in today.
Post by Brian Schweitzer
http://musicbrainz.org/show/release/?releaseid=87415
http://www.rhino.com/store/productdetail.lasso?number=72947&P=upPage
Post by Brian Schweitzer
http://musicbrainz.org/show/release/?releaseid=79928
http://www.amazon.com/Mozart-Baby-Quiet-Richard-Stoltzman/dp/B000066ASX
Post by Brian Schweitzer
http://musicbrainz.org/show/release/?releaseid=88769
"Floetenrondo D-Dur"
http://musik.ciao.de/Gehirnjogging_mit_Mozart_Various__1519428#productdetail
Post by Brian Schweitzer
And best for last,
http://musicbrainz.org/show/release/?releaseid=19236
The 63:56 "Canon in D"
No clue.


Still, hope that helps ;)

Leiv
Jim DeLaHunt
2008-02-27 13:05:04 UTC
Permalink
Post by Leiv Hellebo
...
Post by Brian Schweitzer
If I gave you just one, you'd assume I cherry picked it. Here's a
dozen or so, just pulling from the first 30 on my "messy releases
list".
....
Post by Brian Schweitzer
http://musicbrainz.org/show/release/?releaseid=57554
http://www.amazon.com/Mozart-Relaxation-Richard-Stoltzman/dp/B00000I9M0
etc.

Leivhe, a virtuoso bit of research! But I think it misses the larger point.

The point really ought to be, are the track titles Brian points out really
the quality level we want to settle for in MB? I think they aren't good
enough for my satisfaction, and I imagine even many tagging consumers would
be dissatisfied. The purpose of the CSG is to tell contributors what
quality level and style to aim for.

"Andante" isn't good enough. "Andante K. 315", as Amazon seems to indicate
is on the Release package, is much better, because of the catalog number.
The CSG gives a contributor encouragement to add in that "K. 315".


-----
-- http://jdlh.com/ Jim DeLaHunt , Vancouver, Canada ?
http://wiki.musicbrainz.org/JimDeLaHunt
--
View this message in context: http://www.nabble.com/Re%3A-Musicbrainz-style-Digest%2C-Vol-34%2C-Issue-86-tp15696280s2885p15707227.html
Sent from the Musicbrainz - Style mailing list archive at Nabble.com.
Lauri Watts
2008-02-27 14:00:35 UTC
Permalink
Post by Jim DeLaHunt
Post by Leiv Hellebo
...
Post by Brian Schweitzer
If I gave you just one, you'd assume I cherry picked it. Here's a
dozen or so, just pulling from the first 30 on my "messy releases
list".
....
Post by Brian Schweitzer
http://musicbrainz.org/show/release/?releaseid=57554
http://www.amazon.com/Mozart-Relaxation-Richard-Stoltzman/dp/B00000I9M0
etc.
Leivhe, a virtuoso bit of research! But I think it misses the larger point.
The point really ought to be, are the track titles Brian points out really
the quality level we want to settle for in MB? I think they aren't good
enough for my satisfaction, and I imagine even many tagging consumers would
be dissatisfied. The purpose of the CSG is to tell contributors what
quality level and style to aim for.
The audience for this album already doesn't care for all the detail,
or they wouldn't have bought it in the first place. And these albums
are _huge_ sellers, they regularly break into the US and UK album
charts and can come in pretty high (when was the last time you saw a
Mahler album in the top 40 album chart?)
Post by Jim DeLaHunt
"Andante" isn't good enough. "Andante K. 315", as Amazon seems to indicate
is on the Release package, is much better, because of the catalog number.
The CSG gives a contributor encouragement to add in that "K. 315".
You cannot seriously believe that the people who buy this album, or
it's like, actually give a damn if the catalog number (which may as
well be a chemical formula if you don't know what it is) is not in the
title?

If they are looking up their album here, either to tag it, or for any
other purpose, the people who buy this kind of album, want what's on
the cover. Perhaps with the level of normalisation we apply to all the
other track titles.

Which is what MB records, track titles. As in, what's on the cover.
Not work titles, which are great and all, and please, keep making
lists of them, and working out the fancy style guide for work titles,
because when we get NGS we'll need them. But we're not there yet, and
work titles are way overblown for albums like this. _Please_ read
lukz posts of yesterday.

But just because you find it useless, doesn't make it useless. It
makes it useless for you, that's all. Just as full work titles are
useless to the consumers of _this_ kind of album, because they are
meaningless mumbo jumbo. And there's a whole spectrum in between.
Useless is as useless does.
--
Lauri Watts
Jim DeLaHunt
2008-02-27 14:54:30 UTC
Permalink
Post by Jim DeLaHunt
The point really ought to be, are the track titles Brian points out really
the quality level we want to settle for in MB? I think they aren't good
enough for my satisfaction, and I imagine even many tagging consumers would
be dissatisfied. The purpose of the CSG is to tell contributors what
quality level and style to aim for.
"Andante" isn't good enough [for my satisfaction]. "Andante K. 315", as
Amazon seems to indicate
is on the Release package, is much better, because of the catalog number.
The CSG gives a contributor encouragement to add in that "K. 315".
You cannot seriously believe that the people who buy this album, or it's
like, actually give a damn if the catalog number (which may as well be a
chemical formula if you don't know what it is) is not in the
title?... the people who buy this kind of album, want what's on the
cover....
Point 1: are you arguing that the cover didn't have "K. 315"? What makes you
so sure? Circumstantial evidence: I see three vendor sites which have track
titles as detailed as the CSG's aspirations. I haven't seen the the real
evidence yet, which is the Release cover itself.

"Andante K. 315" and "Andante for flute & orchestra in C major, K. 315 (K.
285e) "
<http://www.amazon.com/Mozart-Relaxation-Richard-Stoltzman/dp/B00000I9M0>
"Andante for Flute and Orchestra in C major, K 315 (285e) "
<http://www.cduniverse.com/productinfo.asp?pid=1131963&style=classical&cart=687779324>
"Andante for Flute and Orchestra in C major, K 315 (285e)"
<http://www.arkivmusic.com/classical/album.jsp?album_id=429>

Brian's point is that there are many track titles with too little
information to identify the work. He gives an example where the MB
TrackTitle is "Andante". If the cover has a track title of "Andante K.
315", then wouldn't it be better if the MB TrackTitle read at least
"Andante K. 315"?

Point 2. You seem very confident that the purchasers of these albums don't
care about whether "K. 315" is in the TrackTitle. But that record in MB
isn't for those users alone. It's for other users too, for reasearchers, for
anyone else who wants an open encyclopedia of music metadata.

Suppose some contributer is encouraged by the CSG ? as *currently*
practised, mind you ? to copy the metadata from the vendor sites: "Andante
for Flute and Orchestra in C major, K. 315 / K. 285e" (modified per CSG).
Who would be dissatisfied by this TrackTitle? Would the innocent purchaser
of the CD, who found this Title string in their digital music player, be
upset?

What's the ceiling on the quality and detail we want in our data?
Post by Jim DeLaHunt
But just because you find it useless, doesn't make it useless. ...
Useless is as useless does.
Well, there you go again. "Useless" is your word. I said "dissatisfied".

Lauri, I appreciate the reality check. You are right, the people who buy
(and likely enter) such a CD probably have different motivations and desires
than those who buy a fine DG release.

But really. Is "Andante K. 315" worse than "Andante"? Is "Andante for Flute
and Orchestra in C major, K. 315 / K. 285e" worse than that? Should MB
reject such contributions? Should MB make no effort to guide contributors
who add them towards consistency and best practices? That's what's at issue
with Brian's examples, I think.

-----
-- http://jdlh.com/ Jim DeLaHunt , Vancouver, Canada ?
http://wiki.musicbrainz.org/JimDeLaHunt
--
View this message in context: http://www.nabble.com/Re%3A-Musicbrainz-style-Digest%2C-Vol-34%2C-Issue-86-tp15696280s2885p15708823.html
Sent from the Musicbrainz - Style mailing list archive at Nabble.com.
Leiv Hellebo
2008-02-27 15:02:02 UTC
Permalink
Post by Jim DeLaHunt
Lauri, I appreciate the reality check. You are right, the people who buy
(and likely enter) such a CD probably have different motivations and desires
than those who buy a fine DG release.
Not only is Lauri right, but DG also makes these kinds of albums. (Try
the "Mad About Bach" etc. series, and others)
Post by Jim DeLaHunt
But really. Is "Andante K. 315" worse than "Andante"?
I don't think so, no

Is "Andante for Flute
Post by Jim DeLaHunt
and Orchestra in C major, K. 315 / K. 285e" worse than that?
IMO, yes.

Should MB
Post by Jim DeLaHunt
reject such contributions? Should MB make no effort to guide contributors
who add them towards consistency and best practices?
IMO, we shouldn't pester submitters to add that data, and there is
little need to add it ourselves. If someone comes along and says it's on
the liners (and I think it very often is) and add ", K. 315" to a
tracktitle, I'would vote yes, as I think the medium packaging should be
the prime arbiter.
Lauri Watts
2008-02-27 15:55:30 UTC
Permalink
Post by Jim DeLaHunt
Point 1: are you arguing that the cover didn't have "K. 315"? What makes you
so sure? Circumstantial evidence: I see three vendor sites which have track
titles as detailed as the CSG's aspirations. I haven't seen the the real
evidence yet, which is the Release cover itself.
Which is the point.

You want to add it whether it's on the cover or not.

I say, if it's not on the cover, we have no place adding it _to the track title_
Post by Jim DeLaHunt
Brian's point is that there are many track titles with too little
information to identify the work. He gives an example where the MB
TrackTitle is "Andante". If the cover has a track title of "Andante K.
315", then wouldn't it be better if the MB TrackTitle read at least
"Andante K. 315"?
Not if it's not on the cover. You are talking about _work titles_ again.

It's identifiable to people who listen album by album. It's
identifiable to people
who own the album. _When we have NGS_ it'll be identifiable by it's connection
to the complete, non-ambiguous, work title.

Until then, when does it matter if an album you don't own, has a useless-to-you
track title on it in the MB database? Why do you need to be
identifying it anyway?
Post by Jim DeLaHunt
Point 2. You seem very confident that the purchasers of these albums don't
care about whether "K. 315" is in the TrackTitle. But that record in MB
isn't for those users alone. It's for other users too, for reasearchers, for
anyone else who wants an open encyclopedia of music metadata.
And that is what NGS is about.

What I am very confident about is that MB is _not_ _only_ for serious
researchers, anymore
than it's for casual users.

And clearly, when NGS _is_ here, someone's going to have to go over
all these albums
again and put back the track titles anyway, are you volunteering for that too?
Post by Jim DeLaHunt
Suppose some contributer is encouraged by the CSG ? as *currently*
practised, mind you ? to copy the metadata from the vendor sites: "Andante
for Flute and Orchestra in C major, K. 315 / K. 285e" (modified per CSG).
Who would be dissatisfied by this TrackTitle? Would the innocent purchaser
of the CD, who found this Title string in their digital music player, be
upset?
Not if that's what's on the cover. But if it's not why does it
possibly matter that
it's not added?

Here's one I have in front of me:
http://www.megastore.se/template/next%2CProduct.vm?itemid=943570

and our entry for it:
http://musicbrainz.org/release/13be4c33-63a3-4d5e-9093-f99825d416ec.html

For me, that's plenty enough information _in this context_ to identify the
works well enough. It's a soundtrack album, which happens to have a couple
of Bach pieces on it.

I'm quite sure those two need some punctuation correction, and well
the typo in the
second is obvious and I'll fix that myself in a second. But in total,
they match what
the cover says, and they match what a pretty reliable store says.

What's there is plenty enough information for my mother in law (I
bought it to send to
her :). I seriously doubt any music researcher gives a damn about
this album. So who is
it actually helping to do anything other than clean up the punctuation
and caps here?
Post by Jim DeLaHunt
What's the ceiling on the quality and detail we want in our data?
Track titles should be the titles of the tracks, as on the cover, post standard
MB 'tidying-up' normalisation. I guess we largely disagree on what is
'tidying-up' but
to me, adding more than what's on the cover, doesn't fit.

Work titles, what you want (and I do understand why, I do), are a
completely different beast,
and hounding people who are innocently adding their random soundtrack
album to go research
the intricacies of Bach, is just not fair.
Post by Jim DeLaHunt
Well, there you go again. "Useless" is your word. I said "dissatisfied".
Brian used the word useless, not you, sorry.
Post by Jim DeLaHunt
Should MB
reject such contributions? Should MB make no effort to guide contributors
who add them towards consistency and best practices? That's what's at issue
with Brian's examples, I think.
I 100% agree that we should have complete, unambiguous, work titles,
and use them
in the appropriate place, when the DB has a way to enter the work
lists. I don't agree
that track titles are the place to use them.

I also think there is a hell of a lot of work to be done making those
lists, and
a "CSG for work titles" is a grand thing to have. I just don't think
it should be
applied wholesale to the database entries we have now, we should go with the
covers as much as possible.

As for rejecting contributors, it is my firm opinion, MB should
_never_ reject a valid
album addition on style issues. Reject it for being a duplicate, for
having too many
or too few tracks, for being a homebrew bittorrent release. But not
ever, ever, because
it's got bad punctuation. Sadly, that isn't the situation just now.
--
Lauri Watts.
Brian Schweitzer
2008-02-27 23:20:45 UTC
Permalink
Post by Lauri Watts
Post by Jim DeLaHunt
Point 1: are you arguing that the cover didn't have "K. 315"? What makes you
so sure? Circumstantial evidence: I see three vendor sites which have track
titles as detailed as the CSG's aspirations. I haven't seen the the real
evidence yet, which is the Release cover itself.
Which is the point.
You want to add it whether it's on the cover or not.
I say, if it's not on the cover, we have no place adding it _to the track title_
And the track title is? If it's listed on the liner as
Bach (sonata Nr 1 G-moll Bwv 1001 Presto)
you seriously would argue that that is exactly how we ought to enter
it? I have a release of Mozart's Requiem K. 626 which has no track
titles whatsoever - none. Should I really enter it as a 14 track
[untitled]?
Post by Lauri Watts
Post by Jim DeLaHunt
Brian's point is that there are many track titles with too little
information to identify the work. He gives an example where the MB
TrackTitle is "Andante". If the cover has a track title of "Andante K.
315", then wouldn't it be better if the MB TrackTitle read at least
"Andante K. 315"?
Not if it's not on the cover. You are talking about _work titles_ again.
It's identifiable to people who listen album by album. It's
identifiable to people
who own the album. _When we have NGS_ it'll be identifiable by it's connection
to the complete, non-ambiguous, work title.
Until then, when does it matter if an album you don't own, has a useless-to-you
track title on it in the MB database? Why do you need to be
identifying it anyway?
You seem to be focused on MB as a tagging/CD lookup database. Perhaps
I'm the radio station streaming work information based on
MB-identification while playing that soundtrack. Perhaps I'm the
music researcher who wants to do a comparative study on the frequency
with which each movement is used on soundtracks in the MB database.
Perhaps I'm last.fm, trying to pair that track up with the identical
track released as a full non-soundtrack release elsewhere. Point is,
you seem to be quite concerned about one subset of tagger users, while
completely ignoring the fact that there are many other uses for the
data than just satisfying the person who is tagging this soundtrack.
(Not to even question that if the person tagging such a soundtrack or
"budget" classical release really cares as little as you say about
details, do they even really care if it says "Sonata" or "Sonata BWV
869" or whatever else?)

As for it being an album I don't own, and thus somehow ought to not
care about, perhaps it's simply a difference between you and I, but I
quite frequently help clean up or verify both incoming and existing
releases, many of which I have no interest in, have never seen, don't
plan to buy, and don't own.
Post by Lauri Watts
Post by Jim DeLaHunt
Point 2. You seem very confident that the purchasers of these albums don't
care about whether "K. 315" is in the TrackTitle. But that record in MB
isn't for those users alone. It's for other users too, for reasearchers, for
anyone else who wants an open encyclopedia of music metadata.
And that is what NGS is about.
No, NGS is about (in this respect) linking the work with the master
with the track. You keep referring to the concept as some magic
panacea. Should we just completely ignore classical then, as every
other music database seems to do, and treat it like a "big mess not
worth trying to fix", just dumping however much or little or wrong
data happens to be on a liner into the database, just because some
tagger users might actually want that?
Post by Lauri Watts
And clearly, when NGS _is_ here, someone's going to have to go over
all these albums
again and put back the track titles anyway, are you volunteering for that too?
Considering that the people on this list you're saying this to are the
same ones who've been making the CSGS lists and who've been actually
doing a lot of the editing with regards to trying to help editors
entering classical currently, and are also the principle ones who've
been actively going through and fixing typos and such in classical,
then yes, I would quite expect that, when NGS does finally come, we'll
also be quite happy and willing to do that part too... if, that is,
track titles haven't reverted to a sea of unidentifiable Andantes and
Allegros, as you would seem to want.
Post by Lauri Watts
Post by Jim DeLaHunt
Suppose some contributer is encouraged by the CSG ? as *currently*
practised, mind you ? to copy the metadata from the vendor sites: "Andante
for Flute and Orchestra in C major, K. 315 / K. 285e" (modified per CSG).
Who would be dissatisfied by this TrackTitle? Would the innocent purchaser
of the CD, who found this Title string in their digital music player, be
upset?
Not if that's what's on the cover. But if it's not why does it
possibly matter that
it's not added?
http://www.megastore.se/template/next%2CProduct.vm?itemid=943570
http://musicbrainz.org/release/13be4c33-63a3-4d5e-9093-f99825d416ec.html
For me, that's plenty enough information _in this context_ to identify the
works well enough. It's a soundtrack album, which happens to have a couple
of Bach pieces on it.
I'm quite sure those two need some punctuation correction, and well
the typo in the
second is obvious and I'll fix that myself in a second. But in total,
they match what
the cover says, and they match what a pretty reliable store says.
And Lauri, pray tell, just from the information at that site and in
our listing, what is the serious difference to you between the
present:

Sonata Nr. 1 G-moll, BWV 1001, Presto

(which is already changed from that site's liner listed version of
"sonata Nr 1 G-moll Bwv 1001 Presto")

and the CSG:

Sonata f?r violine solo No. 1 in G-moll, BWV 1001: IV.Presto

Shall we also not enter performer ARs if they're not in the liner,
should that particular recording's performance info happen to be
identified? How far do you suggest we take this concept?
Post by Lauri Watts
Post by Jim DeLaHunt
What's the ceiling on the quality and detail we want in our data?
Track titles should be the titles of the tracks, as on the cover, post standard
MB 'tidying-up' normalisation. I guess we largely disagree on what is
'tidying-up' but
to me, adding more than what's on the cover, doesn't fit.
And for classical, please, define which liner you refer to, as many
releases are released multiple times. Please define which language on
that liner. And when you say liner, are you referring to the
"Andante" on the back cover, or the full work ID in the booklet? Just
what, to you, is the definitive "only possible acceptable official
track title" for that track on that release, amidst all the
possibilities?
Post by Lauri Watts
Post by Jim DeLaHunt
Well, there you go again. "Useless" is your word. I said "dissatisfied".
Brian used the word useless, not you, sorry.
Yes, though what I said was:

" a) Titles just as they are on liners, helpful or not
...
A, to me is useless. "Allegro" or "Movement from a symphony by
Mozart" as track titles are absolutely useless in identifying the work"

or perhaps you meant when I followed up with

"be done a year from now. But what we will have between now and then
is people using the database and finding what are essentially useless
listings, such as "Allegro" for track titles, thus giving up on MB's
classical, for the same reasons such data is essentially useless
everywhere else. What we will have is a large mass of either
unidentifiable or only semi-identifiable tracks."

whereas you turned that statement into
Post by Lauri Watts
But just because you find it useless, doesn't make it useless. It
makes it useless for you, that's all. Just as full work titles are
useless to the consumers of _this_ kind of album, because they are
meaningless mumbo jumbo. And there's a whole spectrum in between.
Useless is as useless does.
And please, find me the consumer (I assume you mean taggers only,
since you seem to be ignoring all other possible uses for the data)
who seriously argues that "Allegro" is preferable to "Symphony No. 1
for Orchestra in E-flat major, K. 16: I. Allegro molto".
Post by Lauri Watts
I also think there is a hell of a lot of work to be done making those
lists, and
a "CSG for work titles" is a grand thing to have. I just don't think
it should be
applied wholesale to the database entries we have now, we should go with the
covers as much as possible.
I think the people actually using the lists, such as
http://musicbrainz.org/release/1af2287c-428c-4042-a862-cb1c39e8dfb2.html
might disagree.
Post by Lauri Watts
As for rejecting contributors, it is my firm opinion, MB should
_never_ reject a valid
album addition on style issues. Reject it for being a duplicate, for
having too many
or too few tracks, for being a homebrew bittorrent release. But not
ever, ever, because
it's got bad punctuation. Sadly, that isn't the situation just now.
This seems to me an argument for turning MB's classical section into
freedb. So long as someone dumps some tempos in, you're happy?

Anyhow, I find it interesting that, though the box set has been out
for 20 years, and is quite popular, the Philips complete Mozart set
had been barely entered until recently. We had perhaps 30 or 40 CDs
out of the 180 entered. Since the CSGS list for Mozart, the box has
come a long way towards actually being entered - and I've only entered
a half-dozen myself. I think at this point, it's only some of the
operas which are left to be entered. That's 100-some releases, all
entered per CSG, in just over a month, whereas the prior 30-40
releases had taken the entire history of MB to be entered. Some users
seem to be quite happy to have CSGS Mozart available, and seem quite
happy with the results - and I'd assume, all those entering these
releases have been entering them primarily to tag from them.

Brian
Lauri Watts
2008-02-28 00:00:14 UTC
Permalink
On Wed, Feb 27, 2008 at 6:20 PM, Brian Schweitzer
Post by Brian Schweitzer
And the track title is? If it's listed on the liner as
Bach (sonata Nr 1 G-moll Bwv 1001 Presto)
you seriously would argue that that is exactly how we ought to enter
In the context of a movie soundtrack, why not? Do your little
punctuation fixes,
just like you would any other track, take the artist name into the
appropriate field, just like you would any other track. Why does
_this_ album for _this_ audience need more than that?
Post by Brian Schweitzer
it? I have a release of Mozart's Requiem K. 626 which has no track
titles whatsoever - none. Should I really enter it as a 14 track
[untitled]?
No, because just like any rock album, or any other kind, COD takes
precedence, and for _this_ audience, for _this_ album, the complete
titles are entirely appropriate.
Post by Brian Schweitzer
You seem to be focused on MB as a tagging/CD lookup database. Perhaps
I'm the radio station streaming work information based on
MB-identification while playing that soundtrack. Perhaps I'm the
music researcher who wants to do a comparative study on the frequency
with which each movement is used on soundtracks in the MB database.
Perhaps I'm last.fm, trying to pair that track up with the identical
track released as a full non-soundtrack release elsewhere. Point is,
you seem to be quite concerned about one subset of tagger users, while
completely ignoring the fact that there are many other uses for the
data than just satisfying the person who is tagging this soundtrack.
(Not to even question that if the person tagging such a soundtrack or
"budget" classical release really cares as little as you say about
details, do they even really care if it says "Sonata" or "Sonata BWV
869" or whatever else?)
Whereas, you're concerned about one other set of users, and don't give a
damn about any of the others. Pot, Kettle, Black.
Post by Brian Schweitzer
Post by Lauri Watts
Post by Jim DeLaHunt
Point 2. You seem very confident that the purchasers of these albums don't
care about whether "K. 315" is in the TrackTitle. But that record in MB
isn't for those users alone. It's for other users too, for reasearchers, for
anyone else who wants an open encyclopedia of music metadata.
And that is what NGS is about.
No, NGS is about (in this respect) linking the work with the master
with the track. You keep referring to the concept as some magic
And you seem quite determined to fix everything tomorrow, in one fell
swoop, without regard to the fact the database and the users are not able
to cope with it. And that NGS with work titles vs track titles _will_ fix this
exact issue to both our satisfaction.
Post by Brian Schweitzer
panacea. Should we just completely ignore classical then, as every
other music database seems to do, and treat it like a "big mess not
worth trying to fix", just dumping however much or little or wrong
data happens to be on a liner into the database, just because some
tagger users might actually want that?
Strawmen much? We should correct _errors_ in identification, normalise
punctuation and spelling, just as we do everywhere else. Stop overreacting.
Post by Brian Schweitzer
Post by Lauri Watts
And clearly, when NGS _is_ here, someone's going to have to go over
all these albums
again and put back the track titles anyway, are you volunteering for that too?
Considering that the people on this list you're saying this to are the
same ones who've been making the CSGS lists and who've been actually
doing a lot of the editing with regards to trying to help editors
entering classical currently, and are also the principle ones who've
been actively going through and fixing typos and such in classical,
then yes, I would quite expect that, when NGS does finally come, we'll
also be quite happy and willing to do that part too... if, that is,
track titles haven't reverted to a sea of unidentifiable Andantes and
Allegros, as you would seem to want.
You done insulting my intelligence yet? If you are trying to make me mad
I assure you it's failing.
Post by Brian Schweitzer
Post by Lauri Watts
Not if that's what's on the cover. But if it's not why does it
possibly matter that
it's not added?
http://www.megastore.se/template/next%2CProduct.vm?itemid=943570
http://musicbrainz.org/release/13be4c33-63a3-4d5e-9093-f99825d416ec.html
For me, that's plenty enough information _in this context_ to identify the
works well enough. It's a soundtrack album, which happens to have a couple
of Bach pieces on it.
I'm quite sure those two need some punctuation correction, and well
the typo in the
second is obvious and I'll fix that myself in a second. But in total,
they match what
the cover says, and they match what a pretty reliable store says.
And Lauri, pray tell, just from the information at that site and in
our listing, what is the serious difference to you between the
Sonata Nr. 1 G-moll, BWV 1001, Presto
(which is already changed from that site's liner listed version of
"sonata Nr 1 G-moll Bwv 1001 Presto")
Basic capitalisation and punctuation fixes (so they're wrong in the
current state)
Post by Brian Schweitzer
Sonata f?r violine solo No. 1 in G-moll, BWV 1001: IV.Presto
I'm saying, how is a non-classical-elitist listener supposed to know
that? It's not
even on your magic list that I can see. How does my mother-in-law
know it's 'f?r violine'? If she wanted to enter the album, and was
told to 'fix your track titles' and 'do some damn googling', you can
bet she wouldn't return, or try any others. IV of what? How does she
know? She's not interested, to her it's 'that pretty piece on that
movie soundtrack'.
Post by Brian Schweitzer
Shall we also not enter performer ARs if they're not in the liner,
should that particular recording's performance info happen to be
identified? How far do you suggest we take this concept?
AR's are by definition not the track title. Strawmen popping up all
over the place here, it's an epidemic.
Post by Brian Schweitzer
Post by Lauri Watts
Track titles should be the titles of the tracks, as on the cover, post standard
MB 'tidying-up' normalisation. I guess we largely disagree on what is
'tidying-up' but
to me, adding more than what's on the cover, doesn't fit.
And for classical, please, define which liner you refer to, as many
releases are released multiple times. Please define which language on
There is active discussion going on elsewhere, bringing it up here in
a thread most people long ago consigned to the bin, is wasting my
time. Discuss it in the other thread.
Post by Brian Schweitzer
that liner. And when you say liner, are you referring to the
I don't say liner. I say cover, you are the one constantly switching
back and forth.
Post by Brian Schweitzer
"Andante" on the back cover, or the full work ID in the booklet? Just
what, to you, is the definitive "only possible acceptable official
track title" for that track on that release, amidst all the
possibilities?
We covered that only about a thousand times. I get that you disagree.
You get that I disagree. I am _allowed_ to disagree with you, you
know, it's not a capital crime.
Post by Brian Schweitzer
And please, find me the consumer (I assume you mean taggers only,
Don't assume.
Post by Brian Schweitzer
freedb. So long as someone dumps some tempos in, you're happy?
No. We apply the usual standard of MB editing, and _fix it_. If you
(the generic you) know the album is valid and you can't be arsed
fixing it yourself, you have no business voting no on it.

You are overreacting again. Just because I don't care if every single
instance of a work has the exact same title, doesn't mean I don't have
any standards at all. Nothing is as black and white as you make it.
Post by Brian Schweitzer
happy with the results - and I'd assume, all those entering these
releases have been entering them primarily to tag from them.
Who's focused on it being a tagging database?

The people using the list is a self-selected sample, of people who
like the list. And since they're the audience for most of those
discs, fine. I don't care, do whatever you like.

But those people, are not the ones who are having trouble with all
this. It's the folks adding things like "The Classical Album 2005"
who are being bullied into submission.
--
Lauri "I'm really done here" Watts
Brian Schweitzer
2008-02-28 00:32:13 UTC
Permalink
Post by Lauri Watts
I'm saying, how is a non-classical-elitist listener supposed to know
that? It's not
even on your magic list that I can see. How does my mother-in-law
know it's 'f?r violine'? If she wanted to enter the album, and was
told to 'fix your track titles' and 'do some damn googling', you can
bet she wouldn't return, or try any others. IV of what? How does she
know? She's not interested, to her it's 'that pretty piece on that
movie soundtrack'.
Umm, it's not on the list I was working on because I was working on WA
Mozart... this is JS Bach. :D
As for the JS Bach list, no, it's not there yet, though there is a
clear space for it: "Werke f?r ein Soloinstrument - BWV 1001-1013" -
no one has ever claimed the lists are springing from nowhere Lauri.
However, I am subscribed to
http://wiki.musicbrainz.org/CSGStandard/JSBach , and can easily see
that symphonick (with some contributions from Aaron and myself) has
been doing his best to fill in all of the remaining blanks - 1000+
works does take some time to list. As for Googling, umm, all I had to
do to identify the instrument to form that title was google on "BWV
1001"; it was in the very first hit, at wikipedia. It really wasn't
difficult.

And so not all the information matters *to your mother*. She might
not also care if the titles were capitalized in a language correct
fashion, or whether there were endashes or emdashes or what have you.
The point is, for those who *do* care, it is there.

Now, I am not trying to make you mad. You on the other hand, have
myself (and I'm quite sure other classical editors) quite ticked, as,
to be honest, I have no idea what your point is, and you seem to keep
trying to waste time and drag the discussion off on debates about the
entire existence of CSG. CSG exists, it's already an official
guideline, get over it already.

You seem to be against the entire notion of CSG, but when it is
suggested that that is actually the meaning of what you're saying, you
say its not. You want the titles "fixed", but you don't want "work
identification". Lauri, I have no idea, even after all this time,
what you are saying you *do* want. Any suggestion any classical
editor makes you seem to be against; any suggestion of modifying from
track titles you seem to be against; CSG you seem to be against...
and yet you seem quite willing to make capitalization changes,
punctuation changes, changes moving the composer from the title to the
artist field, etc. In short, you seem to be willing to do anything at
all to the title to make it look "clean", but you seem to be against
anything that might actually make that title meaningful.
Post by Lauri Watts
No. We apply the usual standard of MB editing, and _fix it_. If you
(the generic you) know the album is valid and you can't be arsed
fixing it yourself, you have no business voting no on it.
I have to disagree. If something is added with incomplete titles, no
attention to the style guidelines, etc, it is the responsibility of
the editor who added that release to fix it, else have it voted down.
It is not the responsibility of the voters to clean up serious missing
information and style issues in release adds. I'm not saying we vote
something out because it has some bad capitalization. But you seem to
forget - if no one votes no on that add edit with just tempos, and no
one fixes it, it expires in and defacto becomes part of the database.
We have plenty enough "not even basic info" classical release listings
like that in the database already, thank you very much.

Brian

* Who would very much like to stop seeing emails to the list that
debate CSG without actually having any comprehensible point *
Lauri Watts
2008-02-28 00:51:53 UTC
Permalink
On Wed, Feb 27, 2008 at 7:32 PM, Brian Schweitzer
Post by Brian Schweitzer
Post by Lauri Watts
No. We apply the usual standard of MB editing, and _fix it_. If you
(the generic you) know the album is valid and you can't be arsed
fixing it yourself, you have no business voting no on it.
I have to disagree. If something is added with incomplete titles, no
attention to the style guidelines, etc, it is the responsibility of
the editor who added that release to fix it, else have it voted down.
It is not the responsibility of the voters to clean up serious missing
information and style issues in release adds. I'm not saying we vote
something out because it has some bad capitalization.
I've seen exactly that happen. I voted yes on an album addition
_yesterday_ that had a no vote on a style issue, and would have
otherwise failed.'
Post by Brian Schweitzer
But you seem to
forget - if no one votes no on that add edit with just tempos, and no
one fixes it, it expires in and defacto becomes part of the database.
We have plenty enough "not even basic info" classical release listings
like that in the database already, thank you very much.
I prefer to fix them and explain why.

I think voting no without doing so is elitist and unfair, especially
when said no votes are not even accompanied with a comment. It's a
bully tactic, and has no place here.

--
Lauri Watts
Brian Schweitzer
2008-02-28 01:15:26 UTC
Permalink
Post by Lauri Watts
Post by Brian Schweitzer
I have to disagree. If something is added with incomplete titles, no
attention to the style guidelines, etc, it is the responsibility of
the editor who added that release to fix it, else have it voted down.
It is not the responsibility of the voters to clean up serious missing
information and style issues in release adds. I'm not saying we vote
something out because it has some bad capitalization.
I've seen exactly that happen. I voted yes on an album addition
_yesterday_ that had a no vote on a style issue, and would have
otherwise failed.'
Yes, name anything, we've all seen it happen. But *that is not what
we are talking about here* and is *totally irrelevant*.
Post by Lauri Watts
Post by Brian Schweitzer
But you seem to
forget - if no one votes no on that add edit with just tempos, and no
one fixes it, it expires in and defacto becomes part of the database.
We have plenty enough "not even basic info" classical release listings
like that in the database already, thank you very much.
I prefer to fix them and explain why.
I think voting no without doing so is elitist and unfair, especially
when said no votes are not even accompanied with a comment. It's a
bully tactic, and has no place here.
Talk about straw men! When did I suggest that one would vote no and
not explain why? For that matter, when did I suggest anyone would
vote no? My own practice, and my experience with the other classical
editors is that, like just about anywhere else in the database, we
tend to link to the relevant guidelines, point out the things that
need correction, and, except in the really ugly cases, abstain for a
week to give the adding editor time to fix it.

But anyhow, let's assume the editors are all fixing the issues in
classical releases. Doesn't that go right back to exactly what I said
two months ago, re: CSGS lists, in that the entire point is that
classical editors, the few we do have, end up wasting all their time
doing just this fixing you're talking about, which means they have
little time nor desire to do the other things, like cleaning up
existing releases, ARs, etc?

Brian

* Seriously tempted to bring up some random reference, just so I can
invoke Godwin's law and end this seemingly pointless thread *
Leiv Hellebo
2008-02-28 01:52:31 UTC
Permalink
Post by Brian Schweitzer
Post by Lauri Watts
I'm saying, how is a non-classical-elitist listener supposed to know
that? It's not
even on your magic list that I can see. How does my mother-in-law
know it's 'f?r violine'? If she wanted to enter the album, and was
told to 'fix your track titles' and 'do some damn googling', you can
bet she wouldn't return, or try any others. IV of what? How does she
know? She's not interested, to her it's 'that pretty piece on that
movie soundtrack'.
As for Googling, umm, all I had to
Post by Brian Schweitzer
do to identify the instrument to form that title was google on "BWV
1001"; it was in the very first hit, at wikipedia. It really wasn't
difficult.
(Sorry for entering late here, I know you've already moved on, but I've
been busy putting the kid to sleep :)

Actually one could argue that its name is

"Sonata 1ma ? Violino Solo senza Ba?o"

(Loading Image...)

It has been transcribed a number of times for other instruments. From
Wikipedia: "Although this work was intended for violin, Bach himself
transcribed portions for other instruments, and the entire set has been
transcribed by others for guitar, viola, cello and piano."

(And that's not so many. There are others that are "worse" in this respect.)

I've already said that I agree with Lauri, and there's no point in
repeating the arguments. But I'm wondering what you have in store for
Lauri's mother-in-law, Brian, were she to get a Soundtrack where e.g.
Nigel North's transcriptions for guitar of BWV 1002, Partita in B minor
is transposed to A minor.

A final note, Brian: About a year ago there was full agreement between
people interested in classical at MB that best-of compilations, "Tune
Your Brain With Debussy" et al, should be allowed to remain outside the
jurisdiction of the CSG. Just because you keep repeating the same
arguments over and over, I don't think you'll succeed in convincing
many. I think you'll have to come up with something better than you've
done so far.


So, goodbye until things cool off,

Leiv
Brian Schweitzer
2008-02-28 02:17:37 UTC
Permalink
Post by Leiv Hellebo
A final note, Brian: About a year ago there was full agreement between
people interested in classical at MB that best-of compilations, "Tune
Your Brain With Debussy" et al, should be allowed to remain outside the
jurisdiction of the CSG. Just because you keep repeating the same
arguments over and over, I don't think you'll succeed in convincing
many. I think you'll have to come up with something better than you've
done so far.
Actually, this is the first I've ever heard of this, nor it is
something I've seen anywhere in any docs or discussions. Do you
recall quite where this "full agreement" was?

Brian
Leiv Hellebo
2008-02-29 05:05:25 UTC
Permalink
Post by Brian Schweitzer
Post by Leiv Hellebo
A final note, Brian: About a year ago there was full agreement between
people interested in classical at MB that best-of compilations, "Tune
Your Brain With Debussy" et al, should be allowed to remain outside the
jurisdiction of the CSG. Just because you keep repeating the same
arguments over and over, I don't think you'll succeed in convincing
many. I think you'll have to come up with something better than you've
done so far.
Actually, this is the first I've ever heard of this, nor it is
something I've seen anywhere in any docs or discussions. Do you
recall quite where this "full agreement" was?
Sorry again :(, an overstatement. "Full agreement" in MB would be rather
rare, wouldn't it?

So thanks for following up on that quote.

I believe coopera, mll, me and others agreed on that. (It certainly was
said by more people than myself, and it was never contradicted or even
discussed.) If coopera has another take on this, then I have painted the
wrong picture.

Leiv

Aaron Cooper
2008-02-28 00:48:05 UTC
Permalink
Post by Lauri Watts
On Wed, Feb 27, 2008 at 6:20 PM, Brian Schweitzer
Post by Brian Schweitzer
And please, find me the consumer (I assume you mean taggers only,
Don't assume.
As much as I've tried to ignore some of your more asinine comments,
this one threw me over the edge. This sort of thing contributes
absolutely nothing to the discussion and wastes all of our time.
Please let's try to leave our personal issues on the shelf when trying
to communicate on the mailing list...

-Aaron (cooperaa)
Loading...