Difference between revisions of "History talk:CSG Standard/JS Bach"

From MusicBrainz Wiki
(Moved from main page)
Line 1: Line 1:
A little update: I personally won't do too much work on this list before we get all CSG- & worklist-issues sorted out. -- [[User:symphonick|symphonick]] 10:48, 17 April 2008 (UTC)
==Box Sets==
==Box Sets==

Revision as of 09:11, 8 December 2009

A little update: I personally won't do too much work on this list before we get all CSG- & worklist-issues sorted out. -- symphonick 10:48, 17 April 2008 (UTC)

Box Sets

  • Sacred Cantatas: 60 CDs in 4 volumes, conducted by Nikolaus Harnoncourt and Gustav Leonhardt.
    • 60 CDs in 4 volumes, released by Teldec in 1994.
  • Bach 2000: The Complete Bach Edition: 153 CDs in 12 volumes, released by Teldec in 1999.
    • Included a rerelease of all 4 volumes of the 1994 Sacred Cantatas Teldec set.
    • A Bach 2000 (sampler) was also released.
  • Bach 2000 Light: 93 CDs, released in 8 volumes, released by Teldec in 1999.
    • It is identical to the 12 volume Teldec set, but does not include the 4 volumes of cantatas.
  • Bach Edition: 155 CDs, released by Brilliant Classics around 2000 or 2001.
  • Bach Edition: 155 CDs, rereleased with the same title by Brilliant Classics in 2006.
    • Nine CDs were replaced with newer recordings - make sure you are dealing with the correct disc!
  • The Complete Works of Johann Sebastian Bach: 172 CDs, released by Hänssler Classics in 2001.
  • The Harmonia Mundi Bach Edition: 69 CDs, released by Harmonia Mundi around 1999 or 2000.
    • A 2 CD The Harmonia Mundi Bach Edition (sampler) was also released.

Work-specific discussion

18: The chorale is listed as "Ich bitt, o Gott, aus Herzensgrund", but the sources I can find show it as "Ich bitt, o Herr, aus Herzensgrund": http://www.cs.ualberta.ca/~wfb/cantatas/18.html, http://www.bach-cantatas.com/Scores/BWV018-V&P.pdf. Why the difference? -- AndrewConkling 23:05, 17 March 2008 (UTC)

  • Don't know. I have used W. Bishop's site as source/verification for the vocal works I've entered, so I'd prefer "Ich bitt, o Herr" (unless it can be proved that WB got it wrong, of course) -- symphonick 22:14, 19 March 2008 (UTC)
    • My fault, I corrected it -- pradig 05:48, 25 March 2008 (UTC)

182: Can anyone explain the "Concerto" stuff? http://www.cs.ualberta.ca/~wfb/cantatas/182.html just has "Sonata" -- symphonick 11:54, 28 February 2008 (UTC)

Motets 225-230: Someone added mvt numbers to the motets? Since W. Bishop's site doesn't have them, can you explain? I'm not against the quotes, just reluctant to add nos. if Bach didn't write them. http://www.cs.ualberta.ca/~wfb/cantatas/225.html http://www.cs.ualberta.ca/~wfb/cantatas/226.html http://www.cs.ualberta.ca/~wfb/cantatas/227.html http://en.wikipedia.org/wiki/BWV_227 (wikipedia talks about the parts as "movements" on this) http://www.cs.ualberta.ca/~wfb/cantatas/228.html http://www.cs.ualberta.ca/~wfb/cantatas/229.html -- symphonick 22:50, 03 February 2008 (UTC)

  • Are they movements (BWV 227: I. etc)or collection numbers (BWV 227 No. , etc)? I'm not familiar with Bach's motets, but most motets I've ever sung were in collections - though I'm sure there's some out there, I've never seen a a multi-movement motet... -- BrianSchweitzer 03:42, 04 February 2008 (UTC)
    • Had a quick look in a score, & it's neither movements nor collection numbers. Just sections where it's convenient to split a long track on the CD (the chorale-sections in 227 is marked "Corale" in the Bärenreiter score, but no numbers anywhere). -- symphonick 18:32, 24 February 2008 (UTC)
      • How should we proceed here? I entered the mvt numbers, so I'd like to find a solution. You're right, the numbers are not from Bach (as most or even all mvt numbers in the cantatas are probably not from him), but how should we express that the sub-parts of a motet like 227 are performed in a specific sequence? It's definitely not a collection, so collection numbering style would not be right. -- pradig 06:07, 25 March 2008 (UTC)
        • If I'm right, the proposed changes in classical (see RFC: Work lists...) will make it possible for you to enter the tracknames like the tracklist on your release & then link the tracks to the work. IMHO we shouldn't put in stuff in the work lists that the composer didn't write, if it's not necessary (& what you'd expect to see) for identification: Cat. no., voice range for arias, solo instrument for concertos (can't come up with an example with subparts right now...) I really wouldn't mind losing the "mvt numbers" in the cantatas since I never heard anyone use them: "I'm going to sing the 3rd mvt from Bach's cantata no. 33..." For the time being, I suppose we could have duplicate entries in the wiki (at least for 227) until we decide what to do (& I can't figure out if it's I, II, III or i, ii, iii or a, b ,c :( ) -- symphonick 23:40, 27 March 2008 (UTC)

232 Missa h-Moll: The numbering looks odd? The release I have & http://www.cs.ualberta.ca/~wfb/cantatas/232.html don't use a b c sub-parts. We could use "containers" like K626 -> Kyrie: I. Coro "Kyrie eleison", Kyrie: II. Aria ...etc. "Hohe Messe" isn't mentioned anywhere on Bishop's page either. -- symphonick 22:14, 19 March 2008 (UTC)

  • I agree that this looks odd. We could follow either Walter's numbering with containers (which btw seems to follow the "BC" numbering we once had a discussion on, see http://de.wikipedia.org/wiki/H-Moll-Messe#.C3.9Cbersicht), or use the common numbering from the score (also in http://de.wikipedia.org/wiki/H-Moll-Messe#.C3.9Cbersicht). I personally would prefer Walter's numbering, as the score number restart from 1 with each major section and even doesn't have a number for the Sanctus. For the name, "Hohe Messe" is (according to wikipedia) a name that was given to it by music lovers in the 19th century. The most common name is still "h-Moll-Messe" (or "h-moll-Messe" according to CSG). I would vote for the common name. -- pradig 06:24, 25 March 2008 (UTC)

248 Weihnachts-Oratorium: Probably has to be listed twice, numbered 1-64 and per cantata (248I - 248VI). It's usually performed/recorded I-III / IV-VI (& IIRC the Bärenreiter score uses this numbering) but it's written as 6 separate cantatas (I've seen a TV-version that way. No CD's what I can remember) Note that after No. 35 comes No. 24 da capo - maybe this is bad for scripts? -- symphonick 23:50, 01 February 2008 (UTC)

988 Goldberg: I'm not 100% sure about what the abbreviation "clav." stands for in Italian. -- symphonick 22:57, 01 February 2008 (UTC)

Keys still missing (since I can't find them) for: 724, 734 -- BrianSchweitzer 21:59, 11 February 2008 (UTC)

  • That's NP since they shouldn't have a key IMO. I noticed that you added keys for the organ chorales. Why? I have never seen those refered to by key before. -- symphonick 18:32, 24 February 2008 (UTC)

General discussion

Personally, I'd prefer to see English style CSG rather than German style CSG used (seems more common that classical is released in English rather than German style/keys), but that's just my own preference. However, I think some consistency in language and CSG language style is a must. IIRC, per CSG, German keys are just "A-Dur", not "in A-Dur" (or "ein A-Dur"). Also, there's already here a mix of languages for voice part titles - I see at least three already mixed - Alto/Alt, Coro/Chorus/Chori, etc. - some consistency there would be nice. Finally, for the voice parts, afaik, it's not specified as to order within CSG, but I would suggest the OperaStyle version is more readable and consistent, so "II. Recitativo "Sein Wort der Wahrheit stehet fest (Basso)" instead of "II. Recitativo (Basso) "Sein Wort der Wahrheit stehet fest". I'm glad to see this major project get started up though. :) -- BrianSchweitzer 13:06, 02 January 2008 (UTC)

Oh yes, on the formatting change... Though I see why it was done, and it works for the larger works, I think it breaks on the (much more numerous) smaller works. While the other format (columns, like the Mozart listing) may be a bit more redundant on a huge work like Matthäus-Passion, it has 2 benefits that this format loses. In columns form, it's consistent for every work, whether there's 1 or 99 movements. That means that if someone were to write a javascript to do work lookups, it's easy to simply scrape the proper row(s) from the appropriate column. Also, if someone wants to export this list to a spreadsheet, it's already column-formatted. In this current form, it is more readable, but for more machine-based (instead of human-based) use, the "meaning" of each row has been lost. I'd really suggest the format be switched back to the columns-based form. -- BrianSchweitzer 13:12, 02 January 2008 (UTC)

  • Me & pradig have discussed some of the issues you pointed out. We agreed to use OperaStyle for roles (eg. Judas, Pontifex) so I guess having vocalists after the quote would be more consistent, as you say. I put them before the quote because that's how I've always done it :) perhaps we can set a standard here... I think we agree on using Italian for vocalists (looking at pradig's last updates) as the titles are usually in Italian. But if we put vocalists after the quote, perhaps we should use German instead? There's also an inconsistency in roles (Pontifex=DE) vs. vocalists (Tenore=IT) that will be more apparent if we us OperaStyle consistently. Regarding title language, the releases I own have German titles in most cases, perhaps that's different outside Europe. I'd like to stick with German (& Italian), rather than adding a 3rd language. We removed the columns for (human) readability. I just wondered if we should put them back to be able to add another catalogue (BC: http://de.wikipedia.org/wiki/Bach-Compendium). Do you mean the whole page must be one long table? Would headings inside the table be OK (like in the examples at the top of the page)? -- symphonick 13:58, 02 January 2008 (UTC)
    • Sure, you can do headings within the table - see how I do Notes in the Mozart list. I'd suggest, though, that it be kept to a single table, with the headings just rows within that table (Makes it much easier for GreaseMonkey/UserJS/etc to parse elements to the one table rather than trying to script to first find the correct table before doing even trying to find the right row in that table.) If I were to do the Mozart list again, and one thing you might want to consider too, it's much easier to script to the right movement if you also have a movement column, instead of somehow trying to ID the movement from the text, so:
99 1 Kantate, BWV 99 "Was Gott tut, das ist wohlgetan": I. Coro "Was Gott tut, das ist wohlgetan"
99 2 Kantate, BWV 99 "Was Gott tut, das ist wohlgetan": II. Recitativo (Basso) "Sein Wort der Wahrheit stehet fest"
  • I could live with a table like that. It really makes sense to have it parsable for any scripting language one might use. My only concern is that if we put everything in one single table we would end up with a 6000+ lines beast (rough estimate: 1000 number in BWV with 6 tracks each). But as we still don't know whether we might finish the whole work any time soon, we can postpone the decision about that until the table really get too big. -- pradig 00:22, 03 January 2008 (UTC)
    • A beast like this one? (Last I checked it was around 3500 lines or so, and still some 2 dozen or so works not listed). Yes it's a big table, but it's a reference page - I think making it easily scriptable and still readable is a bigger concern than it taking perhaps 8 seconds (on my old P3 machine here) to load and render. -- BrianSchweitzer 21:14, 03 January 2008 (UTC)
  • On the Bach-Compendium catalog, I have to admit I've not heard of it or seen it before. There's no equiv page in the english wiki, but I'm guessing it's something akin to the NME Mozart catalog. My only fear in having additional catalogs used would be if you tried to support multiple catalogs in the tracktext. K1/K6 made sense to list in columns, as each appears frequently without the other on liners, but I avoided listing NME/K2/K3 as columns, as they almost never are on any liner. I've only ever seen BWV numbers for JS Bach, but if you've found that BC numbers appear frequently enough to be worth it, then sure, I'd not see any reason not to list a column for those, to make referencing from BC liners to the correct BWV tracktitles easy. On the other hand, though, I think the point is that we do want the tracktitles to become standardized, so I couldn't see any good reason to have, say, a Matthäus-Passion listing for each BWV and BC, depending on which was used on a liner. (The exception, I think - and something I've thought about doing for the K6 deest Mozart works - would be if the BC has cat #s for any BWV deest works that are worth listing... In such a case, I'd tend to think any solid catalog # is better than yet another a deest, so long as it is indicated as BC (or whatever BC uses) and not BWV in the tracktitle. A similar case to this is Chopin, where editors have been using the B catalog for all the posthum. works that were un-ID in the standard Op. catalog.) -- BrianSchweitzer 18:13, 02 January 2008 (UTC)
    • I have never heard of the Bach-Compendium myself, too. Neither do any of the 100+ release of Bach's composition I own have a single reference to it. And all the major sources on the Web use BWV. So I would propose we consider it esoteric and go forward without it on this page. We don't want to do major research work but just create some sort of standard on track titles. -- pradig 00:22, 03 January 2008 (UTC)
      • Fine with me (I'll admit to having only seen "BC" used a small number of times, for instance on the Naxos webpages IIRC) -- symphonick 12:08, 03 January 2008 (UTC)
        • If either of you knows anything about Chopin, I've got a similar question about his cat #'s over on that wikipage - about to start a listing for him, trying to make a similar decision there. -- BrianSchweitzer 21:14, 03 January 2008 (UTC)

Oops, checking & CSG does define vocalists before the quote... (last example on page) -- symphonick 14:04, 02 January 2008 (UTC)

  • Yep, just saw it too - leaving a new subsection note in CSGD, as I don't think it makes as much sense as the OperaStyle form. -- BrianSchweitzer 18:13, 02 January 2008 (UTC)
    • I think we should put both roles and vocalists at the end of the of the track title using a much Italian as we can. There are not that many words that would be put here (the vocalists SATB plus coro and the roles you find in the oratorios). We could even create an "authentic list" of what is allowed. I agree with symphonick's point that using the English CSG would add another language you do not find in any of Bach's works. -- pradig 00:22, 03 January 2008 (UTC)
      • About language again - maybe we should use what Bach wrote for Worktype too, instead of translating to German? For example, I noticed that he wrote "Moteto" in (facsimile copies of) the autographs to the motets. -- symphonick 18:32, 24 February 2008 (UTC)

Another thing on track numbering. I know there is a discussion on using roman vs. arabic numerals going on. For the big compositions like Matthäus-Passion it really looks more than funny to use roman style (I had look up the logic for numbers > 38 the first time I did it). And all of the Bach scores I own use arabic numerals. Could we set a standard that for Bach's work arabic numerals will be used exclusively? -- pradig 00:22, 03 January 2008 (UTC)

  • IMO a universal solution is needed for this. I don't think we can have an exception just for Bach, & I'm not in favor of Jim's suggestion of a "breakpoint" at 7 mvts either. If we should have exceptions maybe we can consider using Arabic for large-scale vocal works; oriatorios, passions, possibly song-cycles (aren't most cantatas < 10 tracks?) -- symphonick 12:08, 03 January 2008 (UTC)
    • Personally, I don't find the roman numbers all that offensive - in fact, I kind of like them. I really wouldn't want any kind of breakpoint; either one or the other. Of course, that means then we'd have the (to my eyes) uglyness of ": 1. Allegro", ": 6. Minuetto", etc, I'd much rather take the odd IC or LVII if it means avoiding that. On the opera end, where it's much more common, it also means instead of "No. 28 Act VI, Scene XXIII" we'd be in the territory of "No. 28 Act 4, Scene 23", which looks to my eye slightly less "nice", and opera has always used roman in the libretto text. -- BrianSchweitzer 21:14, 03 January 2008 (UTC)
      • To clarify myself; If we should change anything, we need a solution that can be used everywhere. Personally I can live with 4-5 of these works having some readability issues, rather than using "2. Adagio" on the triosonatas & such. -- symphonick 23:00, 06 January 2008 (UTC)
        • Not that it changes my mind about still liking the roman/arabic mix the way it is now, but leivhe and I have run into a real life worst case where this readability does get nasty: "...: II. Minuetto I / III. Minuetto II / II. Minuetto I" (for pretty much all minuet 2nd movements after minuets standardized as the 2nd movement of a concerto, and before the "Minuet I - II" form evolved into the "Minuet & Trio" form. -- BrianSchweitzer 18:00, 26 January 2008 (UTC)

On the Matthäus-Passion, I notice within the dozen or so wikiedit notifications for this page from the last 24 hours that someone changed it to voiceparts, not roles, then it got changed back. Just wanted to say, from my point of view, if there is a useable role, such that OperaStyle can be used in liu of just role and CSG voice-part style, I'd much prefer the roles... -- BrianSchweitzer 18:34, 06 January 2008 (UTC)

  • I suggest we use '(vocal type) "Song" (Character Name)'. This is how I assumed we were doing it but it doesn't appear to be that way in OperaTrackStyle. -AaronCooper
    • Makes sense, and it fits in the formula for CSG and OperaStyle combined I threw out in IRC the other day...
      ...: {movement number(toRoman)}. {Act #(toRoman){, Scene #(toRoman)}.} {No. #(toArabic)} {work type {& repeat all from the colon as needed}{.tempo1 {- tempoN}}{ & repeat all from the colon as needed} {(voice part(s))} {"libretto or movement or submovement title"} {(Character name)} {[movement catalog number]} {/} {&} {repeat all from the colon as needed}
      -- BrianSchweitzer 16:24, 25 January 2008 (UTC)
      • So: ...: [movement]. [overall_multimovement_type]: [performance_type] ([vocal_part(s)]) "name_of_the_song" [(character1, character2, ...)] ? That would give us: ...: * IIIa. "Dies irae" * IIIa. Sequenz: "Dies irae" * IIIa. Sequenz: Aria "Dies irae" * IIIa. Sequenz: Aria (Bass) "Dies irae" * IIIa. Sequenz: Aria (Bass) "Dies irae" (Foobello) * IIIa. Sequenz: Aria (Bass) "Dies irae" (Foobello, John, Genié)

The one other (really rare) funky thing I've run into, just to cover all bases, is the multi-composer collaboratively composed opera, where some parts are separately cataloged within the different composers' catalogs (but not the work entire, so the cat # applies only to the one movement, not the work entire). For these, it doesn't seem to make sense to mix cat #s (esp since not all movements will even have them, and since they don't apply to the work, but only the movement). I'd suggest:

  • IIIa. Sequenz: Aria (Bass) "Dies irae" (Foobello, John, Genié) [K. 999]

with the individually cataloged movement bracketed after the character name area. -- BrianSchweitzer 20:21, 15 January 2008 (UTC)

  • I was wondering about the vocal types (tenor, soprano, etc.) Have we decided anywhere on capitalization and spelling? I had been entering them lowercased and characters (Brunnhilde, Evangelist, etc.)uppercased. Regarding spelling, I've seen "tenor"/"tenore" and "bass"/"basso". Have we decided which to use? -AaronCooper
    • For this particular page I think we agreed on Italian (basso - tenore). I'd like a decision on caps, though. I think I have been using caps because the example in CSG (but that example is in English...) -- symphonick 20:12, 25 January 2008 (UTC)
      • Hmm, I just redid all the lieder and canons (but not the arias) on the Mozart page over the past few days. Only 23 even specify a voice part, but I used English caps for the voice parts in there. The first one was the one I least knew how to list; "Flute (or Clarinet)" is clean, but that same idea when already inside () gets ugly: (Tenor (or Soprano)) (Soprano) (Tenor) (Soprano, Alto, Tenor, Bass) (Soprano, Tenor, Bass) (2 Sopranos, Bass) (2 Sopranos) (Soprano, Tenor, Bass) (Soprano, 2 Tenors, Bass) (5 Sopranos) (3 Sopranos) -- BrianSchweitzer 18:00, 26 January 2008 (UTC)

cooperaa: I reverted your last - not sure what you changed, but in the process, we accidentally lost half the page. Stumbled on to this site, it may be helpful: http://gwdu64.gwdg.de/pls/bach/weeng$.startup -- BrianSchweitzer 01:20, 05 February 2008 (UTC)

Quick question... I just pulled this page to snag a quick non-Mozart example and realized... Where is the instrumentation and key/scale for all these? Also, it looks like Common Name has been used where Work Type ought to be ("Matthäus-Passion", etc)? -- BrianSchweitzer 06:06, 05 February 2008 (UTC)

  • What keys are you missing? I can't remember ever having seen or heard anything about keys in cantatas (or operas)? You want full instrumentation for the Brandenburg Concertos? On the passions, do you think we should have 'Passion, BWV 245 "Matthäus": I. Coro' etc? -- symphonick 08:54, 05 February 2008 (UTC)
    • Well, since cantatas, passions, and masses are covered by CSG, not opera style, I'm just wondering why the instrumentation, key, scale, and vocal parts sections are omitted on all the works so far listed... (By the way, is it Matthäus-Passion or Matthäuspassion ? I see the latter more frequently than the former on the German sites.) ...and man, as much as I hate doing the vocal works, I so do not envy you working on al those cantatatatatatas... :D -- BrianSchweitzer 09:17, 05 February 2008 (UTC)
      • Does CSG require keys on everything? Song-cycles, "The Lark Ascending" etc? If it does, I think we should modify it - who wants to know the key for [first mvt of] "The Messiah"? Read more on MP vs M-P here: http://musicbrainz.org/show/edit/?editid=8016854 (I also thought "Matthäuspassion" was more common) -- symphonick 09:37, 05 February 2008 (UTC)

Another good database -- BrianSchweitzer 21:59, 11 February 2008 (UTC)

Why do we have the whole catalog in German, while we use the Latin "Missa" instead of "Messe"? -- PiotrPKarwasz 19:45, 07 March 2008 (UTC)

  • I don't think we have a consensus regarding the issue of languages in work lists yet. Personally, I'd like to keep as much as possible of what the composer himself wrote. If Bach wrote "Missa", I don't see a need for a translation (in this context). We don't translate "allegro" to "schnell"... Of course, this strategy won't always work; Bach never used the term "Cantata"? and such... & if there's no autograph we'd have to decide what to use next; first edition, catalouge language or composer's language (at the time of writing?)... And there's the issue of what language to use for keys (when those are needed) I don't mind the titles being a little "multilingual" :) but others may disagree... -- symphonick 16:30, 18 March 2008 (UTC)

I noticed that recently there's been a lot of usage of "chorale" instead of "choral", which is on W. Bishops's pages: http://www.cs.ualberta.ca/~wfb/cantatas/55.html (I have no problem if "chorale" is the authentic version for those though). -- symphonick 13:53, 24 March 2008 (UTC)

I added a list of the various JS Bach box sets to the artist annotation, but I'm including it here. The artist page is so huge noone ever loads it to see the annotation anyhow, but also, it maybe can help us make sense of the various sets - I see messy bits and pieces of many of them scattered amidst Bach's listing. The titles I used are as printed on the box (so "Harmonia Mundi" is actually part of the set's title, etc.) I don't know which sets anyone's working on - personally I have the Bach 2000 Complete one that I need to work on. But my sense is, given the sheer number of sets, and the ease by which they could be mixed up, we probably ought to not abbreviate the set titles. On the BC box, while I can find references to there being those 9 different discs, I can't actually find anything saying which they are... so lots of fun potential fix screwed up data between them, since they're the same CDs, same works/movements/titles, same tracklists, same composer, and same label, just different performers and release dates. I'm not sure if the cat # was the same or not between the 2 different releases of the BC box sets.  :( -- BrianSchweitzer 00:32, 30 April 2008 (UTC)

German question: Just noticed this on a release named from here, on 244 | 10 | Matthäus-Passion, BWV 244: Teil I, VI. Aria (Alto) "Buss und Reu". Shouldn't the ss be ß? Don't want to change it preemptively, in case there was a different rule in German of the time... -- BrianSchweitzer 17:10, 20 December 2008 (UTC)