|
View:
New views
20 Messages
—
Rating Filter:
Alert me
|
| < Prev | 1 - 2 - 3 | Next > |
|
|
Next Generation Database SchemaHi everybody,
Robert and me have met in Hamburg and are trying to do as much preparing work for the summit as possible. We have gathered all the excelent ideas for database changes that are flying around on the wiki, and it looks like we will come up with a pretty cool proposal for the next generation schema. This should solve the major PsITA like SG5 (the dreaded FeatruingArtistStyle), the mess around DontCreateRelationshipClusters, the cludgy special VariousArtistsAlbums, and even a lot of InterNationalization issues. The full proposal is still in our brains, but we will write it up soon. In order to do that, however, WE NEED YOUR HELP ! We need two kinds of things to strain-test the schema: * Examples of Albums, Artists and Tracks with all their weird facts and connections that the current database schema cannot represent, but the new one should. * Usage cases: What do you want MusicBrainz to be able to do? Please give us your examples and usage cases. Reply to this post on mb-users with one example/usage case per mail. If you have described such an example in the past discussions, please dig that mail out and forward it here (e.g Orion had some pretty detailed examples, and the current thread on MakingMusicBrainzUsefulForCollectors contains some important usage cases). I will reply to this mail and add an example and a usage case myself so that you get an idea of what we need. Rob and me will meet again on tuesday, it would be great if we had a set of at least a dozen of Artists/Albums/Tracks that we could try to apply to the new schema. Thanks in advance, DonRedman -- Words that are written in CamelCase refer to WikiPages: Visit http://wiki.musicbrainz.org/ the best MusicBrainz documentation around! :-) _______________________________________________ MusicBrainz-users mailing list MusicBrainz-users@... http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users |
|
|
Re: Next Generation Database SchemaOn Fri, 25 Nov 2005 00:09:55 +0100, Don Redman wrote:
> I will reply to this mail and add an example and a usage case myself so > that you get an idea of what we need. OK, here is an example: I.T.T. by Fela Kuti I have a CD of Fela Anikulapo Kuti, which is a combination of two original albums On the CD itself it says: FELA ORIGINALS ORIGINAL SUFFER HEAD / I.T.T. The cover shows the cover of "Original Suffer Head" on the fron and of "I.T.T." on the back. "Original Suffer Head" is attributed to "FÉLÁ ANIKULAPO" "KUTI AFRIKA 70 ORGANISATION" and "EGYPT 80" on the front and to "FELA ANIKULAPO KUTI & EGYPT 80 BAND" on the back. "I.T.T." has the 'sub title' "International Thief Thief" and is attributed to "FELA Aníkúlápó Kútì & AFRIKA 70" While the accents seem to be there for aesthetical purposes only, "Fela Anikulapo Kuti & Egypt 80" and "Fela Anikulapo Kuti & Afrika 70" (fully sanitized) are different bands. The CD was Produced (p) 1997 and aparently released (c) 1999 by Universal with the catalog number UN900. To complicate matters the tracklisting is as follows: From Original Suffer Head: 1. Original Suffer Head 2. Power Show From I.T.T. 3. I.T.T. (Part 1 & 2) My cousin has the orignal LP of I.T.T. It contains one song which is split over two sides, hence the "(Part 1 & 2)" in the track title of the CD, where the two parts are joined back again. All tracks have been remasterd. Now I want this CD to be related to the two original LPs (and I would enter them into the db just for that). I think being precis, my CD should be listed as a compilation. I futher want it to be clear that track 3 comes from the same recording that the parts 1 and 2 come from. E.g. there is a compilation out there that contains only part 1 of I.T.T. These tracks should somehow be related. Next, I want "Fela Anikulapo Kuti & Egypt 80" and "Fela Anikulapo Kuti & Afrika 70" to be distinct bands in MB. I want - both my compilation and the original "Original Suffer Head" LP to be listed under the discography of "Fela Anikulapo Kuti & Egypt 80", - both my compilation and the original "I.T.T." LP to be listed under the discography of "Fela Anikulapo Kuti & Afrika 70", - all three albums to be listed under the discography of "Fela Anikulapo Kuti". Finally, (and this is already more of a usage scenario) I want to be able to do this: I am on the album details page of this CD. I should be able to find out on which other albums the first track "Original Suffer Head" was released, and especially on which album/single it was released _first_ with no more that two klicks. DonRedman -- Words that are written in CamelCase refer to WikiPages: Visit http://wiki.musicbrainz.org/ the best MusicBrainz documentation around! :-) _______________________________________________ MusicBrainz-users mailing list MusicBrainz-users@... http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users |
|
|
Re: Next Generation Database SchemaOK, since you asked for them and I'm currently waiting for everyone else
to awake from their overeating induced slumber... example 1: http://avexnet.jp/item/3rdxm/disc/product/AVCD-30828.html CD title: 3rd X'mas CD artist: released as 3rd X'mas featuring dream+長澤奈央+SweetS+星井七瀬+嘉陽愛子+ PARADISE GO!! GO!!+斉藤未知. Tracks: 1. Giant X'mas~友情にリボン~ 2. 希望の歌'05 ~Merry Christmas to you~ 3. Giant X'mas~友情にリボン~ (Instrumental) 4. 希望の歌'05 ~Merry Christmas to you~ (Instrumental) It's a collaboration between the groups dream, SweetS, & PARADISE GO!! GO!! with the solo artists 長澤奈央, 星井七瀬, 嘉陽愛子, & 斉藤未知. 23 girls total on the song. Current policy depending on your interpretation would be either randomly select one artist out of that bunch and add the rest as feat. Otherwise add 3rd X'mas featuring dream+長澤奈央+SweetS+星井七瀬+嘉陽愛子+PARADISE GO!! GO!!+斉藤未知 or dream+長澤奈央+SweetS+星井七瀬+嘉陽愛子+PARADISE GO!! GO!!+斉藤未知 or 3rd X'mas and stick all their names in a feat. If any of the latter three were used then collaborated ARs would have to be added from all the groups and solo artists to the collaborative artist, although it's not entirely clear if a single "dream collaborated on 3rd X'mas" AR should be used for the whole group or if the individual members of dream should be ARed to the collaboration one. Also, it's "3rd X'mas" because this is their third year of doing a Christmas super collaboration. The previous two years had similar but smaller (only ~12 girls) ones The label Avex is also confused about how to handle it, with this year's one under "3rd X'mas featuring dream+長澤奈央+SweetS+星井七瀬+嘉陽愛子+ PARADISE GO!! GO!!+斉藤未知" but last year http://avexnet.jp/item/dskxx/disc/product/AVCD-30677.html is under "dream+SweetS+嘉陽愛子" in their artists list and two years ago http://avexnet.jp/item/dfsxx/disc/product/AVCD-30551.html is under "dream+フルーツポンチ+SweetS" even though all three have been released under the "xx X'mas featuring <artists>" formatting. Don Redman wrote: > Hi everybody, > > Robert and me have met in Hamburg and are trying to do as much > preparing work for the summit as possible. We have gathered all the > excelent ideas for database changes that are flying around on the wiki, > and it looks like we will come up with a pretty cool proposal for the > next generation schema. This should solve the major PsITA like SG5 (the > dreaded FeatruingArtistStyle), the mess around > DontCreateRelationshipClusters, the cludgy special > VariousArtistsAlbums, and even a lot of InterNationalization issues. > > The full proposal is still in our brains, but we will write it up soon. > In order to do that, however, > > WE NEED YOUR HELP ! > > We need two kinds of things to strain-test the schema: > > * Examples of Albums, Artists and Tracks with all their weird facts > and connections that the current database schema cannot represent, but > the new one should. > * Usage cases: What do you want MusicBrainz to be able to do? > > Please give us your examples and usage cases. Reply to this post on > mb-users with one example/usage case per mail. If you have described > such an example in the past discussions, please dig that mail out and > forward it here (e.g Orion had some pretty detailed examples, and the > current thread on MakingMusicBrainzUsefulForCollectors contains some > important usage cases). > > I will reply to this mail and add an example and a usage case myself so > that you get an idea of what we need. > > Rob and me will meet again on tuesday, it would be great if we had a > set of at least a dozen of Artists/Albums/Tracks that we could try to > apply to the new schema. > > Thanks in advance, > > DonRedman > _______________________________________________ MusicBrainz-users mailing list MusicBrainz-users@... http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users |
|
|
Re: Next Generation Database SchemaExample 2:
http://www.amazon.co.jp/exec/obidos/ASIN/B000BK6Y1I (linking amazon because their label's website is entirely flash based abomination.) CD title: 青春アミーゴ CD artist: 修二と彰 tracklist: 1.青春アミーゴ 2.カラフル 3.絆 The single is a collaboration between 亀梨和也 from the boyband KAT-TUN and 山下智久 from the boyband NEWS. 修二と彰 means "修二 & 彰" - the main song is the theme song for the show "野ブタ。をプロデュース" where 亀梨和也 plays the character 修二 and 山下智久 plays the character 彰. To make it more fun, tracks 2 and 3 are solo songs - 2 is by 山下智久 and is from the show ドラゴン桜 while 3 is by 亀梨和也 and for the show ごくせん. Issues would be: - whether or not to add the collaboration artist or stick it under one half of the name. - if adding the collaboration artist, do you stick collaboration ARs direct to their real names or do you add those ARs to the 修二 and 彰 character names with further "performs as" ARs to their real names. - clearly indicating that while the single is released and credited under the double name collaboration 2 of the 3 songs aren't - in other words, make it clear that while the release artist of the whole single is 修二と彰 the track artists are 修二と彰, 山下智久, and 亀梨和也 respectively. Don Redman wrote: > Hi everybody, > > Robert and me have met in Hamburg and are trying to do as much > preparing work for the summit as possible. We have gathered all the > excelent ideas for database changes that are flying around on the wiki, > and it looks like we will come up with a pretty cool proposal for the > next generation schema. This should solve the major PsITA like SG5 (the > dreaded FeatruingArtistStyle), the mess around > DontCreateRelationshipClusters, the cludgy special > VariousArtistsAlbums, and even a lot of InterNationalization issues. > > The full proposal is still in our brains, but we will write it up soon. > In order to do that, however, > > WE NEED YOUR HELP ! > > We need two kinds of things to strain-test the schema: > > * Examples of Albums, Artists and Tracks with all their weird facts > and connections that the current database schema cannot represent, but > the new one should. > * Usage cases: What do you want MusicBrainz to be able to do? > > Please give us your examples and usage cases. Reply to this post on > mb-users with one example/usage case per mail. If you have described > such an example in the past discussions, please dig that mail out and > forward it here (e.g Orion had some pretty detailed examples, and the > current thread on MakingMusicBrainzUsefulForCollectors contains some > important usage cases). > > I will reply to this mail and add an example and a usage case myself so > that you get an idea of what we need. > > Rob and me will meet again on tuesday, it would be great if we had a > set of at least a dozen of Artists/Albums/Tracks that we could try to > apply to the new schema. > > Thanks in advance, > > DonRedman > _______________________________________________ MusicBrainz-users mailing list MusicBrainz-users@... http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users |
|
|
Re: Next Generation Database SchemaOn Fri, 25 Nov 2005 00:46:00 +0100, Orion wrote:
> OK, since you asked for them and I'm currently waiting for everyone else > to awake from their overeating induced slumber... Thanks a lot. I am just asking some questions for clarification and adding qotes to your text, because all those japanese characters make me dizzy :-). Please review the quotes. >> example 1: http://avexnet.jp/item/3rdxm/disc/product/AVCD-30828.html CD title: "3rd X'mas" CD artist: released as "3rd X'mas featuring dream+長澤奈央+SweetS+星井七瀬+嘉陽愛子+ PARADISE GO!! GO!!+斉藤未知". Tracks: 1. Giant X'mas~友情にリボン~ 2. 希望の歌'05 ~Merry Christmas to you~ 3. Giant X'mas~友情にリボン~ (Instrumental) 4. 希望の歌'05 ~Merry Christmas to you~ (Instrumental) << Are these just the track titles or are these an artist and a track title? Eg. for track one is it: Track title: "Giant X'mas~友情にリボン~" or Track artist: "Giant X'mas" Track title: "~友情にリボン~" or even Track artist: "Giant X'mas" Track title: "友情にリボン" (tildes removed, just decoration) >> It's a collaboration between the groups "dream", "SweetS", & "PARADISE GO!! GO!!" with the solo artists "長澤奈央", "星井七瀬", "嘉陽愛子", & "斉藤未知 ". 23 girls total on the song. Current policy depending on your interpretation would be either randomly select one artist out of that bunch and add the rest as feat. Otherwise add "3rd X'mas featuring dream+長澤奈央+SweetS+星井七瀬+嘉陽愛子+PARADISE GO!! GO!!+斉藤未知" or "dream+長澤奈央+SweetS+星井七瀬+嘉陽愛子+PARADISE GO!! GO!!+斉藤未知" or "3rd X'mas" and stick all their names in a feat. << Hmm, forget feat. If we had a system that could show an arbitrary string as the album title and still link the album to a lot of different artists, what would _you_ think is the correct 'artist' this should be attributed to? And under which artist(s) should the album be listed? I suppose it should be listed on the discographies of all these: "dream", "SweetS", & "PARADISE GO!! GO!!", "長澤奈央", "星井七瀬", "嘉陽愛子", and "斉藤未知". In all cases it should still retain the correct attributed artist "3rd X'mas featuring dream+長澤奈央+SweetS+星井七瀬+嘉陽愛子 +PARADISE GO!! GO!!+斉藤未知". >> Also, it's "3rd X'mas" because this is their third year of doing a Christmas super collaboration. The previous two years had similar but smaller (only ~12 girls) ones The label Avex is also confused about how to handle it, with this year's one under "3rd X'mas featuring dream+長澤奈央+SweetS+星井七瀬+嘉陽愛子+ PARADISE GO!! GO!!+斉藤未知" but last year http://avexnet.jp/item/dskxx/disc/product/AVCD-30677.html is under "dream+SweetS+嘉陽愛子" in their artists list and two years ago http://avexnet.jp/item/dfsxx/disc/product/AVCD-30551.html is under "dream+フルーツポンチ+SweetS" even though all three have been released under the "xx X'mas featuring <artists>" formatting. << Does this mean this project has become sort of an institution? But IIUC it has no 'name' of its own. Still, I have to ask: Would it make sense to create a collaboration like "xx X'mas" and list all albums there? I suppose not or only additionally to relating each specific album to all the artists that took part on in. DonRedman -- Words that are written in CamelCase refer to WikiPages: Visit http://wiki.musicbrainz.org/ the best MusicBrainz documentation around! :-) _______________________________________________ MusicBrainz-users mailing list MusicBrainz-users@... http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users |
|
|
Re: Next Generation Database SchemaExample 3:
http://www.interq.or.jp/blue/katsuya/1989/coco/cd2.html (search down on PCCA-00472) CD title: CoCoパーソナルベスト (CoCo Personal Best) CD artist: CoCo Album is 10 tracks, 2 solo songs by each of the 5 members of CoCo. Problem again is of indicating that it's a CoCo album when it has 0 songs by CoCo on it, all by the individual members of the group, 1 per track. Don Redman wrote: > Hi everybody, > > Robert and me have met in Hamburg and are trying to do as much > preparing work for the summit as possible. We have gathered all the > excelent ideas for database changes that are flying around on the wiki, > and it looks like we will come up with a pretty cool proposal for the > next generation schema. This should solve the major PsITA like SG5 (the > dreaded FeatruingArtistStyle), the mess around > DontCreateRelationshipClusters, the cludgy special > VariousArtistsAlbums, and even a lot of InterNationalization issues. > > The full proposal is still in our brains, but we will write it up soon. > In order to do that, however, > > WE NEED YOUR HELP ! > > We need two kinds of things to strain-test the schema: > > * Examples of Albums, Artists and Tracks with all their weird facts > and connections that the current database schema cannot represent, but > the new one should. > * Usage cases: What do you want MusicBrainz to be able to do? > > Please give us your examples and usage cases. Reply to this post on > mb-users with one example/usage case per mail. If you have described > such an example in the past discussions, please dig that mail out and > forward it here (e.g Orion had some pretty detailed examples, and the > current thread on MakingMusicBrainzUsefulForCollectors contains some > important usage cases). > > I will reply to this mail and add an example and a usage case myself so > that you get an idea of what we need. > > Rob and me will meet again on tuesday, it would be great if we had a > set of at least a dozen of Artists/Albums/Tracks that we could try to > apply to the new schema. > > Thanks in advance, > > DonRedman > _______________________________________________ MusicBrainz-users mailing list MusicBrainz-users@... http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users |
|
|
Re: Next Generation Database SchemaOn Fri, 25 Nov 2005 00:09:55 +0100, Don Redman wrote:
> I will reply to this mail and add an example and a usage case myself so > that you get an idea of what we need. And here is a usage case: I want to find out how many different recordings of "Star Sprangled Banner" by Jimi Hendrix there are. I want to do no more than enter Artist: "Jimi Hendrix" Title: "Star Sprangled Banner", hit enter and select with max one klick what I am looking for (I am looking for recordings, not for tracks on releases). Then I want MusicBrainz to give me a list of all recordings sorted by date. And not the kind of list it currently gives me at /taglookup.html. This does not tell me at all which of these are the same and which are different. If I am lucky the tilte of a bootleg album might tell me the date and location of the recording, but official albums don't. With one more klick I want to go to the details of one specific recording. I want it to tell me which musicians played on that performance. With one more klick I want to be able to have a listing of all albums, this recording is on. DonRedman -- Words that are written in CamelCase refer to WikiPages: Visit http://wiki.musicbrainz.org/ the best MusicBrainz documentation around! :-) _______________________________________________ MusicBrainz-users mailing list MusicBrainz-users@... http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users |
|
|
Re: Next Generation Database SchemaOn Fri, 25 Nov 2005 01:03:39 +0100, Orion wrote:
> Example 2: > http://www.amazon.co.jp/exec/obidos/ASIN/B000BK6Y1I (linking amazon > because their label's website is entirely flash based abomination.) > > CD title: > 青春アミーゴ > > CD artist: > 修二と彰 > > tracklist: > 1.青春アミーゴ > 2.カラフル > 3.絆 > > The single is a collaboration between 亀梨和也 from the boyband KAT-TUN > and 山下智久 from the boyband NEWS. 修二と彰 means "修二 & 彰" - the > main song is the theme song for the show "野ブタ。をプロデュース" where > 亀梨和也 plays the character 修二 and 山下智久 plays the character 彰. > To make it more fun, tracks 2 and 3 are solo songs - 2 is by 山下智久 > and is from the show ドラゴン桜 while 3 is by 亀梨和也 and for the show > ごくせん. Issues would be: > - whether or not to add the collaboration artist or stick it under one > half of the name. > - if adding the collaboration artist, do you stick collaboration ARs > direct to their real names or do you add those ARs to the 修二 and 彰 > character names with further "performs as" ARs to their real names. > - clearly indicating that while the single is released and credited > under the double name collaboration 2 of the 3 songs aren't - in other > words, make it clear that while the release artist of the whole single > is 修二と彰 the track artists are 修二と彰, 山下智久, and 亀梨和也 > respectively. This one is perfect. And all the issues you rised are real challenges. I am sure, we are going to use this one excessively. So could you please do me a favor and give us _transliterations_ to these names, so that we can _talk_ about them. Babelfish gives me some quite unsatisfactory: 亀梨和也 = turtle pear Kazuya (turtle pear?, how do you pronounce that in Japanese?) 山下智久 = Yamashita Tomohisa 修二と彰 = learning/repairing two and Akira (uh?) 修二 = learning/repairing two (how do you _pronounce_ that?) 彰 = Akira The rest is only context information. I think you cannot imagine how helpful this is. Thanks again DonRedman -- Words that are written in CamelCase refer to WikiPages: Visit http://wiki.musicbrainz.org/ the best MusicBrainz documentation around! :-) _______________________________________________ MusicBrainz-users mailing list MusicBrainz-users@... http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users |
|
|
Re: Next Generation Database SchemaExample 4:
One of these came up during the remix restructuring debate http://lists.musicbrainz.org/pipermail/musicbrainz-style/2005-August/000305.html and in the months since I've run into a similar second case. First example is linked in that mail and is a 2 disc album where disc two contains only a megamix of the first disc into 1 track. Second similar example would be 野宮真貴's album "PARTY PEOPLE" from October where track 10 was titled "PARTY PEOPLE MEGA-MIX" and consisted of a 28 minute megamix of the previous 9 tracks/37 minutes. Possibly more of an AR issue but definitely something that can't be represented at the moment - for the first you can say disc 2 is a mix of disc 1, but for the second you can't say that a disc is a mix of itself. You need something at the track level to indicate that a track is a mix of a disc but even then it becomes tricky when the track is on the disc... Don Redman wrote: > Hi everybody, > > Robert and me have met in Hamburg and are trying to do as much > preparing work for the summit as possible. We have gathered all the > excelent ideas for database changes that are flying around on the wiki, > and it looks like we will come up with a pretty cool proposal for the > next generation schema. This should solve the major PsITA like SG5 (the > dreaded FeatruingArtistStyle), the mess around > DontCreateRelationshipClusters, the cludgy special > VariousArtistsAlbums, and even a lot of InterNationalization issues. > > The full proposal is still in our brains, but we will write it up soon. > In order to do that, however, > > WE NEED YOUR HELP ! > > We need two kinds of things to strain-test the schema: > > * Examples of Albums, Artists and Tracks with all their weird facts > and connections that the current database schema cannot represent, but > the new one should. > * Usage cases: What do you want MusicBrainz to be able to do? > > Please give us your examples and usage cases. Reply to this post on > mb-users with one example/usage case per mail. If you have described > such an example in the past discussions, please dig that mail out and > forward it here (e.g Orion had some pretty detailed examples, and the > current thread on MakingMusicBrainzUsefulForCollectors contains some > important usage cases). > > I will reply to this mail and add an example and a usage case myself so > that you get an idea of what we need. > > Rob and me will meet again on tuesday, it would be great if we had a > set of at least a dozen of Artists/Albums/Tracks that we could try to > apply to the new schema. > > Thanks in advance, > > DonRedman > _______________________________________________ MusicBrainz-users mailing list MusicBrainz-users@... http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users |
|
|
Re: Next Generation Database SchemaDon Redman wrote:
> This one is perfect. And all the issues you rised are real challenges. > I am sure, we are going to use this one excessively. So could you > please do me a favor and give us _transliterations_ to these names, so > that we can _talk_ about them. Babelfish gives me some quite > unsatisfactory: > 亀梨和也 = turtle pear Kazuya (turtle pear?, how do you pronounce that > in Japanese?) > 山下智久 = Yamashita Tomohisa > 修二と彰 = learning/repairing two and Akira (uh?) > 修二 = learning/repairing two (how do you _pronounce_ that?) > 彰 = Akira > The rest is only context information. > > I think you cannot imagine how helpful this is. > > Thanks again > > DonRedman Ah, sorry. I didn't bother at first since they're all names. Babelfishing the definition of Japanese names is as pointless as doing it to American names - I think mine technically means something like "Beloved Son of Fire" if you trace back the origins but people generally don't use that sort of thing. Anyway: 亀梨和也 == last name: 亀梨 Kamenashi; first name: 和也 Kazuya 山下智久 == last name: 山下 Yamashita; first name: Tomohisa 修二と彰 == Shuuji to Akira == Shuuki & Akira - the characters first names seperated by the main Japanese word for "and." According to google you can see http://www.j-fan.com/drama/drama.cgi?action=viewrev&season=Fall_2005&show=NOBUTA._o_PRODUCE for more information than you probably want about the show and it's characters. 修二 == Shuuji 彰 == Akira _______________________________________________ MusicBrainz-users mailing list MusicBrainz-users@... http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users |
|
|
Re: Next Generation Database SchemaOn Fri, 25 Nov 2005 00:09:55 +0100, Don Redman wrote:
> I will reply to this mail and add an example and a usage case myself so > that you get an idea of what we need. I just found the Cassetteboy album which was discussed over a year ago <http://www.nabble.com/forum/Search.jtp?forum=2885&local=y&query=cassetteboy> MusicBrainz album: <http://musicbrainz.org/album/f3e459eb-e17a-4ea9-a8bc-4936301b8c43.html> Discogs: <http://www.discogs.com/artist/Cassetteboy> Here is all the information I could gather from past mails and modnotes (did I already tell you how much I love Opera M2?): On Fri, 19 Aug 2005 22:03:57 +0200, teleGUISE wrote: > Being his first & only release with these 'aliases' it's hard to say > whether > he'd use them again but from reading the meaning behind them I wouldn't > think so. > It appears the suffix to 'Cassetteboy' is just a description of what > the track is about, so i'd say track annotation is more fitting. > > Secondly, does 'Cassetteboy' even belong in the title or shouldn't the > title be split like "Cassetteboy: Mick's Tape" On Mon, 22 Aug 2005 19:30:46 +0200, 'ojnkpjg' has attached a note to moderation #3229576: > here's the record label's web page for the album: > http://www.antidoterecords.co.uk/releases/ANTCD112.htm > > they do list the cassetteboy variations as the performer of the tracks, > however it does seem like playful information that seems more like > additional title info than a performance name. for what it 's worth, > there are other websites whose tracklistings show only 'cassetteboy' as > the performer of those tracks (e.g., > http://www.moviegrooves.com/shop/cassetteboymickstape.htm ). i'm not > sure how to handle this, really. i suppose if an artist declares, "I've > performed track 1 as G, track 2 as H, track 3 as I," etc., who are we to > argue. > > Moderation link: http://musicbrainz.org/showmod.html?modid=3229576 DonRedman -- Words that are written in CamelCase refer to WikiPages: Visit http://wiki.musicbrainz.org/ the best MusicBrainz documentation around! :-) -- Words that are written in CamelCase refer to WikiPages: Visit http://wiki.musicbrainz.org/ the best MusicBrainz documentation around! :-) _______________________________________________ MusicBrainz-users mailing list MusicBrainz-users@... http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users |
|
|
Re: Next Generation Database SchemaExample 5:
CD artist: ZONE with Girlz&Boyz~Run Time All Stars~ CD title: ASTRO Girlz&Boyz Track title and artist info: (copy/pasted from http://tvstation.jp/special/artist/htm/zone.htm) M1.科学の子/西村朝香(コーラス:AYUKA、HITOMI、MAMI) M2.空を越えて/神&瞬矢from BON'Z(コーラス:SANAE) M3.星の彼方/半澤悠&栞 M4.アトムの子/TAKAYO from ZONE&Run Time All Stars M5.誕生/鈴の音(コーラス:Run Time All Stars) M6.鉄腕アトム/ZONE&Run Time All Stars M7.鉄腕アトム(Instrumental) Explanation: Run Time is a production company with various groups and solo artists managed and trained under it. ZONE is pretty much the only big artist they've made so far. Track 1 has Nishimura Tomoka (西村朝香) as the artist with 3 misc younger members of the company as backup but not included in the track artist. To make things fun, she wasn't a member of Zone when this single came out but did join at the beginning of 2004 when one left at the end of 2003 until they broke up completely earlier this year. Track 2 has two members of the group BON'Z credited with the group's name as part of it (神&瞬矢from BON'Z). Some random younger girl thrown in again on the chorus but not credited in the artist. Track 3 is credited to two random girls from Run Time's stable (半澤悠&栞). Track 4 is 1 member of Zone + a whole herd of misc younger members of Run Time that were credited as "Run Time All Stars" Track 5 was a short lived 2 person group (鈴の音) from Run Time with the all stars doing the chorus but not in the artist credit. Track 6 is ZONE & the "Run Time All Stars" herd Track 7 is the instrumental to that. It's often listed as a ZONE single since it starts with their name and they're the only fairly popular artist on it. I remember this CD giving me a headache when I got it... And I think I just got a new one trying to explain it. Specific issues that are problematic: - When one or two members from a band perform solo and are credited with the bands name so that people recognize them, how do we indicate this? - how do you credit the artist the CD was released under "ZONE with Girlz&Boyz~Run Time All Stars~" when that name wasn't used except on this release and wasn't the artist for a single track. - how to deal with collaboration artists like the one on track 3, where, as far as I'm aware and can find on google, neither participant has released anything else. The analogy to the ever useful David Bowie and Queen collaboration would be if neither Bowie nor Queen had ever released anything besides that one song and it were to have been on a Prince album. Don Redman wrote: > Hi everybody, > > Robert and me have met in Hamburg and are trying to do as much > preparing work for the summit as possible. We have gathered all the > excelent ideas for database changes that are flying around on the wiki, > and it looks like we will come up with a pretty cool proposal for the > next generation schema. This should solve the major PsITA like SG5 (the > dreaded FeatruingArtistStyle), the mess around > DontCreateRelationshipClusters, the cludgy special > VariousArtistsAlbums, and even a lot of InterNationalization issues. > > The full proposal is still in our brains, but we will write it up soon. > In order to do that, however, > > WE NEED YOUR HELP ! > > We need two kinds of things to strain-test the schema: > > * Examples of Albums, Artists and Tracks with all their weird facts > and connections that the current database schema cannot represent, but > the new one should. > * Usage cases: What do you want MusicBrainz to be able to do? > > Please give us your examples and usage cases. Reply to this post on > mb-users with one example/usage case per mail. If you have described > such an example in the past discussions, please dig that mail out and > forward it here (e.g Orion had some pretty detailed examples, and the > current thread on MakingMusicBrainzUsefulForCollectors contains some > important usage cases). > > I will reply to this mail and add an example and a usage case myself so > that you get an idea of what we need. > > Rob and me will meet again on tuesday, it would be great if we had a > set of at least a dozen of Artists/Albums/Tracks that we could try to > apply to the new schema. > > Thanks in advance, > > DonRedman > _______________________________________________ MusicBrainz-users mailing list MusicBrainz-users@... http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users |
|
|
Re: Next Generation Database SchemaDon Redman wrote:
> Tracks: > 1. Giant X'mas~友情にリボン~ > 2. 希望の歌'05 ~Merry Christmas to you~ > 3. Giant X'mas~友情にリボン~ (Instrumental) > 4. 希望の歌'05 ~Merry Christmas to you~ (Instrumental) > << > > Are these just the track titles or are these an artist and a track > title? Eg. for track one is it: > Track title: "Giant X'mas~友情にリボン~" > or > Track artist: "Giant X'mas" > Track title: "~友情にリボン~" > or even > Track artist: "Giant X'mas" > Track title: "友情にリボン" (tildes removed, just decoration) Those are the full track titles. Namigata (the ~ mark) are often used for several different random things in Japanese titles so it's a bit confusing. In this particular case the titles are "Giant X'mas~友情に リボン~" (Giant X'mas ~Ribbon of Friendship~) and "希望の歌'05 ~Merry Christmas to you~" ('05 Song of Hope ~Merry Christmas to you~) The extreme end of use of them is stuff like Vo Vo Tau's song "裸 ~nude~" which would be just "Nude~nude~" if it were all in English. >>> > It's a collaboration between the groups "dream", "SweetS", & "PARADISE GO!! > GO!!" with the solo artists "長澤奈央", "星井七瀬", "嘉陽愛子", & "斉藤 > 未知 ". > 23 girls total on the song. Current policy depending on your > interpretation would be either randomly select one artist out of that > bunch and add the rest as feat. Otherwise add "3rd X'mas featuring > dream+長澤奈央+SweetS+星井七瀬+嘉陽愛子+PARADISE GO!! GO!!+斉藤未知" or > "dream+長澤奈央+SweetS+星井七瀬+嘉陽愛子+PARADISE GO!! GO!!+斉藤未知" or > "3rd X'mas" and stick all their names in a feat. > << > > Hmm, forget feat. > If we had a system that could show an arbitrary string as the album > title and still link the album to a lot of different artists, what > would _you_ think is the correct 'artist' this should be attributed > to? And under which artist(s) should the album be listed? > > I suppose it should be listed on the discographies of all these: > "dream", "SweetS", & "PARADISE GO!! GO!!", "長澤奈央", "星井七瀬", "嘉 > 陽愛子", and "斉藤未知". In all cases it should still retain the > correct attributed artist "3rd X'mas featuring dream+長澤奈央+SweetS+星 > 井七瀬+嘉陽愛子 +PARADISE GO!! GO!!+斉藤未知". This year's one was actually less confusing because both songs were by the whole group. Instinctive feeling is that the song artist would be "dream+長澤奈央+SweetS+星井七瀬+嘉陽愛子 +PARADISE GO!! GO!!+斉藤未知" while the album artist would be "3rd X'mas featuring dream+長澤奈央+ SweetS+星井七瀬+嘉陽愛子 +PARADISE GO!! GO!!+斉藤未知" And yes, preferably it'd should up on the artist pages for all involved artists. >>> > Also, it's "3rd X'mas" because this is their third year of doing a > Christmas super collaboration. The previous two years had similar but > smaller (only ~12 girls) ones > > The label Avex is also confused about how to handle it, with this year's > one under "3rd X'mas featuring dream+長澤奈央+SweetS+星井七瀬+嘉陽愛子+ > PARADISE GO!! GO!!+斉藤未知" but last year > http://avexnet.jp/item/dskxx/disc/product/AVCD-30677.html is under > "dream+SweetS+嘉陽愛子" in their artists list and two years ago > http://avexnet.jp/item/dfsxx/disc/product/AVCD-30551.html is under > "dream+フルーツポンチ+SweetS" even though all three have been released > under the "xx X'mas featuring <artists>" formatting. > << > > Does this mean this project has become sort of an institution? But IIUC > it has no 'name' of its own. Still, I have to ask: Would it make sense > to create a collaboration like "xx X'mas" and list all albums there? I > suppose not or only additionally to relating each specific album to all > the artists that took part on in. Correct. Lineup keeps changing each year and the year is incremented with the lineup after it with "featuring" prefixed. The groups dream and SweetS have been in it ever year, while the first year had Fruits Punch with them, second year had Kayoh Aiko with them, and this year was them with the group PARADISE GO!!GO!! plus the solo artists Nagasawa Nao, Hoshii Nanase, Kayoh Aiko, and Saito Michi. First year the format went: track 1 (all), track 2 (dream), track 3 (fruits punch), track 4 (SweetS), track 5 (instrumental of track 1), track 6 (instrumental of track 2), track 7 (instrumental of track 3), and track 8 (instrumental of track 4). Second year and this year both had just 2 songs + 2 instrumentals by the whole grouping. I'm not sure if I see what would be gained by sticking them under an artist like "xx X'mas" _______________________________________________ MusicBrainz-users mailing list MusicBrainz-users@... http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users |
|
|
Re: Next Generation Database SchemaExample 6:
The "one track by another artist single artist album" http://www.universal-music.co.jp/ahhco/disco.html ahhco's first album from the end of August this year includes track 4 that was originally released as feat. her by another artist. It's the same recording and her discography page lists it as feat. her. Song appeared on the H☆G☆S☆P album "HIGH-GRADE-SPECIAL" http://www.jvcmusic.co.jp/-/Discography/A018961/-.html over a year before showing up on her album. This is a simple one: would like to be able to credit the track to H☆G☆S☆P feat. her as it is on both releases without the album being stuck down in the VA section. Don Redman wrote: > Hi everybody, > > Robert and me have met in Hamburg and are trying to do as much > preparing work for the summit as possible. We have gathered all the > excelent ideas for database changes that are flying around on the wiki, > and it looks like we will come up with a pretty cool proposal for the > next generation schema. This should solve the major PsITA like SG5 (the > dreaded FeatruingArtistStyle), the mess around > DontCreateRelationshipClusters, the cludgy special > VariousArtistsAlbums, and even a lot of InterNationalization issues. > > The full proposal is still in our brains, but we will write it up soon. > In order to do that, however, > > WE NEED YOUR HELP ! > > We need two kinds of things to strain-test the schema: > > * Examples of Albums, Artists and Tracks with all their weird facts > and connections that the current database schema cannot represent, but > the new one should. > * Usage cases: What do you want MusicBrainz to be able to do? > > Please give us your examples and usage cases. Reply to this post on > mb-users with one example/usage case per mail. If you have described > such an example in the past discussions, please dig that mail out and > forward it here (e.g Orion had some pretty detailed examples, and the > current thread on MakingMusicBrainzUsefulForCollectors contains some > important usage cases). > > I will reply to this mail and add an example and a usage case myself so > that you get an idea of what we need. > > Rob and me will meet again on tuesday, it would be great if we had a > set of at least a dozen of Artists/Albums/Tracks that we could try to > apply to the new schema. > > Thanks in advance, > > DonRedman > _______________________________________________ MusicBrainz-users mailing list MusicBrainz-users@... http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users |
|
|
Re: Next Generation Database SchemaAs mentioned on IRC inhouseuk donredman: another example of an album that has a different name in one country is this. http://musicbrainz.org/album/3b9cc0b1-1a61-48c5-a98d-da28a29767fd.html it was only released as that name in the US due to confusion regarding a telephone message. The story goes that, someone was asked to contact the band to find out what the album title should be and couldn't get hold of them. It is know as "Electric Light Orchestra" in all other markets donredman inhouseuk: and the track titles are all the same? inhouseuk identical, but we only currently have the rerelease version with extra bonus tracks in the database Shepard then it's just another release of the same album. no problem, as it should have a different EAN donredman but you say there is one with the exact same tracks, right? inhouseuk donredman: yes there is, I'll try and find out the catalog numbers and add the original "other markets" release donredman can you post that to mb-users, please? inhouseuk sure I'm still searching for the catalog numbers at the moment, but the UK release was 1971 with the US a year later. This wikipedia gives a little more information: http://en.wikipedia.org/wiki/Electric_Light_Orchestra_%28album%29 Steve (inhouseuk) _______________________________________________ MusicBrainz-users mailing list MusicBrainz-users@... http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users |
|
|
Re: Next Generation Database SchemaA similar usage case that occured to me as well:
I want to be able to find every language a song has been performed in. The simpler form of this would be entering something such as Artist: "Tata Young" Title: "Sexy, Naughty, Bitchy" and having it return the Thai, English, and Japanese versions of it she's done. The more complex form would be searching on Kim-lian's "Teenage Superstar" http://musicbrainz.org/track/7bf9a0ba-3d71-44bd-83b8-af770d229b46.html and having it tell me that Belinda's "Boba Niña Nice" http://musicbrainz.org/track/0b2ad4ed-5e22-40e1-8dbf-a6ae69a96ece.html is a Spanish cover of it and that 玉置成実's "High School Queen" http://musicbrainz.org/track/bb208297-9ef7-485a-8a82-24648b90efb0.html is a Japanese cover of it. Both would probably need a performance language attribute for tracks, seperate from the current script language one. Don Redman wrote: > On Fri, 25 Nov 2005 00:09:55 +0100, Don Redman wrote: > >> I will reply to this mail and add an example and a usage case myself >> so that you get an idea of what we need. > > > And here is a usage case: > > I want to find out how many different recordings of "Star Sprangled > Banner" by Jimi Hendrix there are. > > I want to do no more than enter > Artist: "Jimi Hendrix" > Title: "Star Sprangled Banner", > hit enter and select with max one klick what I am looking for (I am > looking for recordings, not for tracks on releases). > > Then I want MusicBrainz to give me a list of all recordings sorted by > date. And not the kind of list it currently gives me at > /taglookup.html. This does not tell me at all which of these are the > same and which are different. If I am lucky the tilte of a bootleg > album might tell me the date and location of the recording, but > official albums don't. > > With one more klick I want to go to the details of one specific > recording. I want it to tell me which musicians played on that > performance. > > With one more klick I want to be able to have a listing of all albums, > this recording is on. > > DonRedman > > > _______________________________________________ MusicBrainz-users mailing list MusicBrainz-users@... http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users |
|
|
Re: Next Generation Database SchemaA slightly edited example from Lukas on IRC:
<luks> donredman: if you want more examples, http://musicbrainz.org/track/9be92e03-47d1-4f61-a045-caa5ddcd5f84.html and http://musicbrainz.org/track/1b7be9fe-fbeb-4941-b1f7-7eaf9fa9f99e.html <donredman> luks: oh, that is a good one, this seems to be two localized variants of the same album, right? They even have different album titles <luks> yes, two localized version of the album <donredman> note that the frenc version contains a very fine pun, that is untranslated (or even translatable?) in english <luks> but the titles are not a translations <luks> "Magnetic Fields" != "Les Chants Magnétiques" <luks> or to be more specific "fields" != "chants" :) <donredman> is it not the same 'work', nonetheless? <luks> it's the same work <luks> just different titles <donredman> that one is perfect, thanks! -- Words that are written in CamelCase refer to WikiPages: Visit http://wiki.musicbrainz.org/ the best MusicBrainz documentation around! :-) _______________________________________________ MusicBrainz-users mailing list MusicBrainz-users@... http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users |
|
|
Re: Next Generation Database SchemaHi all:
I was wondering about having multiple titles for songs, based on the preferred language of the user. For example, I can understand Japanese and U.S. English, and I like my U.S. country-western songs with English titles, but I want my Japanese songs with Japanese titles. Sometimes I correct pre-existing song titles from English to Japanese, but I suppose there are some people who just want the English version for title tracks, or the Japanese ones, or other language. Also, while I'm asking for a pony, I would like to be able to save all the CD Ids I've ever created, even for CD where the information has already been entered into MB. I suppose if I'm the 1888th person to enter Michael Jackson's Thriller CD, I wouldn't expect to have the CD IDs record on my client or my personal history, right? Best regards, Clifford Caoile _______________________________________________ MusicBrainz-users mailing list MusicBrainz-users@... http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users |
|
|
Re: Next Generation Database SchemaHere is an other example from yllona I found on the chatlog.
This one is pretty complicated. IIUC this is album title: Three Works for Jazz Soloists and Symphony Orchestra album artist: Don Sebesky Tracklisting: 1. Bird and Bela in BB: 1st Movement 2. Bird and Bela in BB: 2nd & 3rd Movement 3. The Rite of Spring 4. Sebastian's Theme If I read allmusic and amazon correctly, the composer of all songs is Sebsky, but his compositions are _based_ on works of Bartok (1&2), Stravinsky (3), and Bach (4). Note that MB should be able to group 1 and 2 under the work "Bird and Bela in Bb". The following is true for both the album and each individual track: Performer (orchestra): Royal Philharmonic Orchestra Conductor: Harry Rabinowitz Performer (guitar): Joe Beck Performer (trumpet): Jon Faddis Performer: (trombone): Bob Brookmeyer and more? See the (heavily edited) Chatlog for details: <yllona> * yllona is listening to Sebesky/Bartok: Bird And Bela In B flat: 1st movement by Don Sebesky; Royal Philharmonic Orchestra; Harry Rabinowitz, Conductor from Three Works For Jazz Soloist & Symphony Orchestra (1979) (0:41 / 28:19) <Shepard> we have a display name and we have the people who really take part in it <Shepard> no problem with that <yllona> the artist is "Royal Philharmonic Orchestra; Harry Rabinowitz, Conductor" <yllona> it appears on a "don sebesky" album <yllona> and let's not get started on the soloists <yllona> the album is "Three Works For Jazz Soloist & Symphony Orchestra" <Shepard> what does the cover say for the artist? <yllona> http://tinyurl.com/7km3h <Shepard> can't read the artist for the track on it :) <yllona> one sec <Shepard> Don Sebesky is the display name for the album artist <yllona> shep: http://www.amazon.com/gp/product/images/B00000JZ13/ref=dp_product-image-only_0/002-9666643-0652017?%5Fencoding=UTF8&n=507846&s=music <yllona> here's the full amazon record: http://tinyurl.com/7zruj <Shepard> you can't handle this like we do with classical stuff, right? like not file it under the performer <yllona> yeah.. i still have issues w/ composer == artist. because who's the compser in this case? don sebesky or igor stravinsky? or bela bartok? <Shepard> why is sebesky written on top? <yllona> the album was released under his name.. he's a rather famous jazz composer/arranger <yllona> like guunther schuller <ruaok> in our model composer [...] is a type of artist. <Shepard> as i said: you have a display artist for the release - in this case it would be sebesky as the album he clearly filed under his name <Shepard> and then you can link who exactly did what work -- Words that are written in CamelCase refer to WikiPages: Visit http://wiki.musicbrainz.org/ the best MusicBrainz documentation around! :-) _______________________________________________ MusicBrainz-users mailing list MusicBrainz-users@... http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users |
|
|
Re: Next Generation Database SchemaOn Fri, 25 Nov 2005 00:09:55 +0100, Don Redman wrote:
> The full proposal is still in our brains, but we will write it up soon. > In order to do that, however, > > WE NEED YOUR HELP ! > > We need two kinds of things to strain-test the schema: > > * Examples of Albums, Artists and Tracks with all their weird facts > and connections that the current database schema cannot represent, but > the new one should. So far we still need: * one or two simple classical examples. * What about these albums that were released as single disc albums/singles and then in a set? Anyone has an example for this? * for further intensive strain-tests (although the japanese examples are already pretty tough) we'd need examples that span more that one release/track: * one example of a simple remaster with or without additional tracks. * are there albums that contain tracks with identical titles and lenghts but different audio (remastered audio or even more differences?) * What about DJ-Mixes? Those that separate the songs in tracks, and those that have all in one track? * In what aspects can two albums differ only _very_slighlty_: * a rerelease with nothing changed except the EAN code? * a rerelease with different cover but otherwise identical? * just a bonus track added * etc (this stuff is important to figure out on which level which information must be stored). > * Usage cases: What do you want MusicBrainz to be able to do? PLEASE tell us what you use MusicBrainz for. I know that this is evident to you, but it is not evident to me. I have once helped to design an iterface for MB with my usage practices in mind: That clunky add-AdvancedRelationships interface. You can tell that I am not a mass-editor who does more than a few edits a day, can't you? I'd hate if this happened again. To be more specific, we need to know what you do (or want to) _use_ MusicBrainz data for. How do you want your files to be tagged? What kind of information do you want to be able to retrieve? To do what? DonRedman -- Words that are written in CamelCase refer to WikiPages: Visit http://wiki.musicbrainz.org/ the best MusicBrainz documentation around! :-) _______________________________________________ MusicBrainz-users mailing list MusicBrainz-users@... http://lists.musicbrainz.org/mailman/listinfo/musicbrainz-users |
| < Prev | 1 - 2 - 3 | Next > |
| Free embeddable forum powered by Nabble | Forum Help |