Schoenberg's Theory

Actually, while I don't know the full story, Wikipedia says it was written in 1910, which makes it public domain in the US and probably most other countries. Any given modern edition might still be copyright protected, so photocopying and distributing the examples from a printed copy of a more recent printing would still be illegal, but re-entering the musical examples yourself should be OK. Not that I'd recommend taking my word as definitive, but I do think it could still be worth pursuing.

I have begun my part from the last file, posted by Jojo-Schmitz just above, and urge you to to the same. It is, again, imperative we start from an identical file, since parts are to be joined later, and what work you have already done should copy/paste OK.

Some formatting should be standardized:
Example number: Rehearsal Mark [Ctrl + M]
Example subdivision (‘a), b), c), …’): Staff Text [Ctrl + T]

Everything else may be considered cosmetical, thus left at respective discretion. For instance, on the one hand, I have kept the 1- to 4-voice progressions to discrete voices, on the other I have done the Roman numeral notation only for a couple examples where they can be done entirely as lyrics [Ctrl + L]. (I think there is a better way using figured bass, but will not dig deeper for now.)

Figured bass is created using Ctrl+G in 2.0 Beta/Nightly. You'd have to fake it using ordinary text in 1.3. I wouldn't recommend wasting time on that. FWIW, not sure it's been totally decided yet, but I wouldn't advise using 1.3 for this project. 2.0 is the way of the future; if you're creating something for posterity, better to use the newer format, even if that means you have to wait a little while before completing the project.

As for time signatures - not sure what you mean exactly, but if it's just a matter of wanting to put some random number of beats in a measure, you don't need time signatures for that. Right click a measure, Measure Properties, change the Actual Duration. Set it to 50/4 and you can have 25 half notes. Here also, 2.0 will provide easier ways of doing this so you don't have to count the beats first - just enter as separate measures, then select the range and do "Join Measures". But the underlying mechanism is actually the same either way.

If they are not the same pitch, then by definition it's not a tie. In regular music (not a textbook) I'd say it's probably a slur. Here it is more likely just a generic line used to indicate some sort of relationship between the notes that is explained in the text. But a slur (from the Lines) palette is still the best way to enter that, assuming you so want it curved.

Unisons already appear side by side by default in 2.0, another reason to use that over 1.3 :-)

To force notes to a different side of the stem than the default in either version, click and press Shift+X.

You can also manually shift things sideways. In 1.3 Win, you must double-click the element to move, and click elsewhere to release before double-clicking next element. Selecting a note other than the one on top is easier when zooming [Ctrl + ↑/↓], using-choosing voice layer, or playing with wrong pitch.

Sideways big step: [←/→] or [Shift + ←/→]
Sideways small step: [Ctrl + ←/→]

Schönberg differs from recent notational standard in placement of accidentals and cluster notes. With a risk of being mossy, I think it best to quote his original layout (for notes, not accidentals) for better visual accord; especially since stuff like chord analysis may not (initially, at least) be fully covered by us, and it will be possible with a mere few keystrokes to revert to today’s practice.

Just plow through it, yes, the way to go. Notes and numbering are essential, as said before; chord/functional analysis cannot be transcribed in full, should therefore be [considered cosmetical =] left for later, be it ever so big a part to exclude. When one comes upon a new bird, check the corpus for whether it can be transcribed in full without disturbing what has already been set:

#130a: big “parenthesis” whose left part is missing for me, probably still there but gone fishy outside of working area. Maybe jumped at save because it was linked to the same note as the right parenthesis half, or because I twisted it to get it vertical; it is actually 1 of 2 slurs from the lines palette. (The smaller parenthesis framing the top natural in the same bar is of another type, one of >1 actual parantheses, behaves differently and is not affected.) Better to delete the “parenthesis” and keep record of this and similar specimen to be fixed later.

#131c: “×”. Can be done with ‘cross head’ from [Z] Symbols palette, but is not a head (and slightly smaller); therefore I did it in Lyric as an ASCII multiplication sign (pedant-approved size).

#102 bar 9: “!!” done in Lyric as 2 exclamation marks because of “×” above. (There is a Unicode ‘Double exclamation mark’ but one might as well have double-flat “bb” for precedence.)

# 140b: Not there yet.

…

Also, Schönberg at times discards example subdivision or gives Key headers “instead”. (See #134; discarding for no apparent reason, as he has both subdivision and key header in examples both preceding and following close; I am also unsure about his use of Double Bar (and 1.3 difference between dragging and double-clicking it).) He also detours into complex numbering as in #136, prompting numbering format exceptions. Such strains are enough for 1 item (Stave Text) without also venturing it onto functional &/or chord notation the standards of which we know beforehand it cannot but approximate. One could in plain “flat” text notate a 6/4 chord discernable from a 6—4 progression, but in this case should not.

No need to rush things, I am slow too. There are as always other things less important but more urgent, so my part will come along in fairly random bursts.

I like a challenge. If you can post an image of what you want to achieve for those three missing staves I'll give it a go and (if I can do it) I'll post a step-by-step how-to.

Back again, but needing some help. Hours and pages down the drain, I can neither get rid of the corruption mentioned earlier, nor get past it.

See attached file, measure 479 (#141/b, 5th measure; pdf page 212). Deleting the mess, even deleting every corrupt measure and the following blank ones as well as several preceding ones, appending new blanks and renotating them does not help: Notes look fine when doing it, saving also, but reload the file and – voilà, back again! Have done many partly remakes and 3 full ones with different deletions, it repeats exactly and I cannot proceed.

Go to measure 491, select 3rd note (C in voice 1): playback has 3 simultaneous notes, all in voice 1. Select any of the 3, playback has them all. (They are properly notes 5, 8 & 9 in a phrase.)
Curiously, #142 a) looks OK but is not (see playback).
Also, vertically crammed manually adjusted spacing is no problem (see measures 269—84, —331).

Probably, one would nightly get a better look at the file, but the latest version I could run on present computer is 2014-10-22-1409-75d9e76, deleted at update and since then not recovered by me. (Is there a possibility for downloading/reconstructing outdated versions?) I have other computers without restrictions, but until Monday not with enough time.

Joe H: “no way to predict what notes and chords will occur next” in Schönberg’s Theory?! ;D
Well, as with his use of extra-musical symbols, then.

Hopefully, the joint effort will carry friction-less through 2.0 and, once saved there, things too cumbersome at the moment can be postponed until then, then easier done. In any version, certainly trickier luring someone into proof-reading the lot.
All well.

Problem yet unsolved. Have been digging around it a lot; trying to isolate it by re-notating several stretches from different starting points (not exhausted yet), going further with the problematic area skipped in different ways (gives no solution, or no definitive ‘health clearance’). I cannot see what causes the measures to corrupt before they have done so, and trying nightly for a better look at the manifest thing causes crash.

I will, hopefully tonight, post a couple of threads to sort out possibilities with other’s help.

FWIW, there are many more and much worse problems with corruption in 1.3 than in the current 2.0 builds.

I think the way to move forward is to not try to do this as one monolithic file but to break it up. Maybe not one file per example, but group related examples together. Then you can use 2.0's album feature to connect them all.

Did you look at corruption details?

Bar 106 Stave 2 incomplete. Expected: 13/1; Found: 12/2
Bar 267 Stave 2 incomplete. Expected: 12/1; Found: 28/4
Bar 268 Stave 2 incomplete. Expected: 15/1; Found: 12/1


All the corruptions seem to be related to rests being "accidentally" added in stave 2.

Hi again. I noticed your posts earlier but needed time to check things through.

Joe H: The text is available under Monoskop. I had 2 different editons borrowed from my National Library of Music, but only partial scans of otherwise illegible detail. Also, if you go no further, please post your file/-s here, so I know what is left.

And damn. Very informative clues at last.

fabo92: Looking at corruption details caused program crash in earlier nightly, as said before. The 3 inconsistencies you listed comes from 2.0.1 ‘Load error’, which is a new feature. I know not whether there are more ones.

Jojo-Schmitz: I see. That use of voices is of course intentional, also as explained briefly before. It may be considered improper by MuseScore, but is essential to music, being a manifestation of traditional 4-part harmony by voice leading; and with Schönberg’s treatise systematically as pedagogically mapping that very realm, and MuseScore offering means, I thought it worthwhile a reasonable amount of extra work to make it read out. Allocation of the 1×4 voices to 2 hands/staves must not be ‘enharmonically’ misunderstood as 2×2 (= a tabulatory voice 1 & 2 right + 1 & 2 left); it is for practical reasons only that scores are monochrome, but the 1×4 rule is kept proof through either plain typography (prime intervals shown as such with double stems/heads) or implicit adherence to (vis-à-vis knowledge of) voice leading – with few exceptions; Schönberg has them too, as noted earlier. One will certainly find a related argument behind MuseScore developer’s choice of 4 voices and different colours.

That strange enough behaviour is exactly what I tried to probe earlier. In 2.0.1 debugger the program no longer crashes, and from your post I have indication but not understanding. In measure 106, I have used from top to bottom Voice 1, 2 (top stave) & 3,4 (lower stave); which translates to ‘Track’ 0, 1 & 6, 7, and in between them (Track 4) is the accidentally added rest of duration 12/2 corresponding to the Load error message. Similar rests are added for other measures, but for this one duration is incorrect? The 3 listed errors all have ‘long’ time signatures (13/1, 12/1, 15/1), but measure 129 (13/1) has its added rest of correct duration (13/1). If the ‘improper’ use of voices is causing the corruption, then why is only 3 inconsistencies listed? I do not really need an answer to this, but a way to safe 2.0 save.

I find it quite odd though, forcing the use of voices to always start with a nominal 1, the more so since one can manually skip that 1 to start with 2/3/4 and have MuseScore keep its count by those formal voice 1 rests, and one therefore expects not corruption from “accidentally” added rests. In short, ‘Voices’ comes with as tiny as deep a reservation to theory & practice, and one cannot be sure to handle >4 voices per system. I find this misconceptional, but understand it coming from my own assumption, and hope it be a 2.0 non-issue.

Rant aside, I will of course renotate all corrupting passages provided they are all accounted for and cannot be handled otherwise by 2.0.

Me too, so how do you want the remainder? I have done a starting several in file per example, will now tweak things in my old file, that comes from migrating 1.3 to 2.0.1 .

Happily enough, the corruption that has taken so much extra time and effort is not a problem anymore. I will notate as before (manifest voices where applicable) so maybe corruption will reappear. I know not what causes it, but how to detect (2.0.1 reload after save) and fix it (explicit input of ‘missing’ time signature in listed measures, exchange voices forth and back).

I understand and appreciate. You should in that case take Goetschius to the finish before anything more with Schönberg without feeling burdened. I have hands full too, but can take some steps forward in the meantime. It should be rewarding fun, and no hurry.

see https://musescore.org/en/handbook/measure-operations-0#split-join

Rehearsal marks seem the best idea to me if you're determiend to do this all in one score. It still seems it would make more sense to each a separate score.

I don't understand what you mean about blocks changing direction when releasing the shift key. There was no such feature in 1.X. But it is true the the order in which notes are added to chords when adding them one at a type via Shift+letter has changed - they always add bottom to top, instead of in a mostly unpredictable order as they did in 1.X (sometimes up, sometimes down depending on the context).

OK, I will then do as I said, and complement the separate files with an album. That way neither of us nor a relay user will have to do any extra work, except for my part already intended.

select the note and double-click the desired accidental on the palette, this is easier to get right. With drag and drop you need to drop when the target note changes color, sometimes it is difficult to find the spot, esp. if the target note is selected already, in which case it won't change color (in 1.x it turned red, in 2.0 it turns blue, hard to tell the dii to voice 1 color), so you have no clue whether it is the target or not. All that is not an issue using the other method

Or just click the appropriate accidentals on the toolbar.

What goes wrong when you follow those instructions? Works fine for me.

For the measure that has a half note with no half rest, change the measure's actual duration to 2/4 in its Measure Properties.

Good question! That's actually somehting I hadn't thought about when suggesting these be done as separate files to be combined later - it's really best if the files to be combined have the same number of staves. I believe we currently do support it in the development builds if some scores have fewer than the default (the default being whatever the first score has), but there have been bugs, but it's probably not such a bad idea to use two anyhow. That's my opinion at least.

Well, yes, just as with 1.3 - you'd need to have four staves throughout, and then use hide empty staves to hide the unused staves for the rest of the score. If you are doing this in 2.0 as separate scores that then get joined, you will have to make sure the joined score has four staves, and the way to do that is to put four staves in the first score.

I guess it is a multi measure rest

In which case, turn them off by pressing "M".

See https://musescore.org/en/handbook/measure-rests-0#multimeasure-rest.

See the Handbook under "Custom palette"