Hello Knute and Kieren, > Interested to hear what others think!
If I am given two choices, either Knute’s or Kieren’s solution, I would choose Knute’s (sorry Kieren!)… For me, the barline right after m. 24 (i.e. where the repeat sign occurs) in Kieren’s solution makes me think that the music would go forward—to the next system in this case—and somehow the repeat sign back to m. 9 at the very end of the system seems rather secondary to me, precisely because there’s a barline at m. 24. Also, if somehow the system contained some more measures, that barline after m. 24 starts to look even odder, I would think. I think it is important to note that the Knute’s solution works precisely because there is a good space between d’’4. and the cancellation/change of the key signature+time signature. I almost am positive that I have seen something similar in actual pieces before, though I can’t really recall where. I also will say that contrary to one Facebook commenter (kinda weird I can still view the post even though I deleted my FB account) that the cancelling F and C naturals here is visually very a nice touch, and I would argue they should stay. It may be true that such “naturals as a key signature cancellation” haven’t been used often nowadays, but in this case it makes sense. Just having a flat alone would start to look like a kind of a typo. I don’t remember if Gould had anything to say about a situation like this. I can’t check as I left the book at school. That’s my take on this… Yoshi -- -- -- Yoshiaki Onishi https://github.com/yoshiakionishi/lilypond-snippets
