I can’t think of any other situations where key signatures show signposts, apart from the one I’ve already stated.
Don’t they show anywhere there’s a change after the initial key signature as well?
Not as far as I can tell.
(I usually keep signposts turned off so I couldn’t remember.)
OK, I have a score that changes to “Atonal”, which shows as a signpost as expected, and then changes to A major, which also generates a signpost; at another point, the change from Atonal to F major also generates a signpost. In both cases, if I try to delete the tonal key-signature signpost, its key signature is deleted as well.
Now, any time I try to trim the (rather lengthy) score down so I can attach it here to show this phenomenon, the signpost for the tonal key signature is no longer present. So there’s evidently some principle at work here, and that’s what I’m struggling to understand.
I suppose I could take the attitude, “What does it matter, as long as the music is the way I want it?” But I suppose that’s not my nature; I want to understand Why. I guess I’ll continue to try on my own, since producing a manageable file to attach here removes the problem.
That’s odd… switching from atonal to a defined key doesn’t create a signpost. I just checked it again to confirm.
Perhaps there’s something else going on in your file.
Yup, clearly there is. (The very first change from Atonal to a key – Ab major – in my project never had a signpost.)
Pinpointing the “something else” is what’s nagging at me. If I delete measures from elsewhere in the score (as I did, to minimize file size if I was going to attach it here), the signpost disappears of its own accord, so clearly something relevant changed at a distance. That’s why I was wondering if there was a place where the whole philosophy of key signposts was explained at exhaustive length; it’s what I need right now.
Can you post some large screenshots perhaps?
Leo’s explanation a few posts earlier isn’t long, but it’s pretty thorough.
Do you have independent barlines/meters/key signatures anywhere earlier in the score? Was it a musicxml import initially?
Sure, here’s one, if it can be of any use. I know screenshots are generally frowned on around here (with good reason) for diagnostic evidence, but here goes.
I suspect the F Major shows a signpost because, for the fourth, ninth and tenth staves, they constitute a shift to C major (which would otherwise be invisible).
(Sorry, I didn’t see the inquiry before my previous reply till after I’d posted. Yes, I probably buried the information that it was an XML import from Sibelius in the middle of a paragraph on the first page.)
Anyway, pianoleo’s suspicion makes sense to me. If true, it clears up a lot, and I’ll keep an eye out for similar situations in future.
(I did have another signpost later in the piece for a change to A major, where there weren’t any such signature-less staves; but in the course of fiddling around trying to come up with something attachable here, that one has vanished of its own accord. I may never know why!)
Thanks, all.
Out of interest, do these signposts remain or disappear if you switch the view to Concert pitch?
They’re still there when I view in Concert Pitch.
Signposts for global items like time and key signatures always appear on the top staff in the system, even if in the case of a key signature the staff for which they are appearing is actually further down the system. And you cannot select and delete a signpost without deleting the item to which it corresponds because the signpost is the item, simply the item drawn in an alternative way.
I’ve resolved, I think, all the unexpected key signature signposts in this particular score, but here’s an unexpected and undetectable time-signature signpost, that’s still there when I condense it down for attachment here. (Again, this originated as an XML import.)
After one measure, what is the reason for the undeletable “4/4 (q,1+1+1+1)” signpost? I’ve never had a change of meter there. My thanks for any insight.
JK by RRB forum excerpt.dorico.zip (398 KB)
I’m not in front of a computer right now, but what happens if you attempt to delete the barline immediately before the red signpost?
Rinaldo,
This is one of those weird things from XML. We’ve seen this before with other projects. Click on the barline to which the signpost is attached and hit delete. The signpost will disappear but the barline will remain. Hitting delete seems to clear whatever data erroneously cause Dorico to think there was an additional time signature when there wasn’t.
Leo beat me to it lol. I was typing when he responded. I did open your project, fwiw, and it does disappear just as we predicted.
Yes, that takes care of it. (I just tried this myself, and unsurprisingly duplicated your results.) I had tried a number of desperation solutions – make the barline double and then return it to single, add a time signature and then delete it – without result, but would never have thought that “delete the barline” would do anything helpful, when most of the time it would simply delete the barline. I’m still at a loss how I might have guessed this would work, but I’m more than willing to accept it as just another example of “Sometimes XML import leaves weird artifacts,” and to try to remember it for next time.
My sincere thanks, as so often before, to pianoleo and Romanos401.