Re: Gis major key signature; Lily's key signature algorithm

[Hans Åberg]

> On 7 Feb 2018, at 22:18, Urs Liska <address@hidden> wrote:
> 
>> That double crosses and double ♭s happen frequently if you transcripe music. 
>> in this cases it's better to use the circle of fifth/fourth, however you 
>> might call it.
> 
> Wow, quite a bold statement, given that we have no clue about the historical 
> context of the original poster's question.
> I'd always argue that depending on the style (actually most European music 
> from the 18th until far into the 20th century) E major is worlds apart from 
> Fes major (and with "worlds" I really mean heaven/earth, life/death, 
> dream/reality, whatever you want).

The staff system refers to Pythagorean tuning, and orchestral instruments, 
mainly the strings, adapt the harmony into 5-limit Just Intonation. It makes 
distant keys (with many accidentals) harder to perform, and less harmonically 
focused, which the composer might exploit.

> My favourite example is in Schubert's song Schwangesang D 744 
> (http://imslp.org/wiki/Schwanengesang,_D.744_(Schubert,_Franz) ).
> The song is in a flat major, then turns to the darker mood of the variant a 
> flat minor and its parallel c flat major (both six flats) and then reaches an 
> absolute anticlimax on the word "auflösend" (meaning: life is dissolving) on 
> the minor subdominant: a fes minor seventh chord (=> <fes' asas' ces'' 
> eses''> in LilyPond language)! There's no way this could ever make sense in e 
> minor.
> But what makes even *less* sense is the helpless rendering of the original 
> edition: <fes g ces d> (the d even being "resolved" to des).

This probably happens on piano, too, before the development of effective E12 
tuning methods, which is early 1900s.