Re: Diatonic notation system

[Hans Aberg]

On 10 Dec 2008, at 07:55, Graham Breed wrote:

> > I attach an example file.
>
> I don't see any transpositions.

From what I recall, if one uncomments the key signature, lilypond  
complains that it cannot handle the rational numbers. So one does not  
get as far as doing transpositions. Otherwise, one could have written
  \key b \bayati

> > I think it is the commented out part (long time ago), which for  
> > some reason
> > only works with E24. Therefore, the keys must be written explicitly.
>
> There's a key signature specification which does look like a bug to
> me.  I get the same result in 2.11.65.  Have you reported it as a bug?

I think the reason is that key signature only work in E24. Then on  
top of that, an non-transposable model for Turkish music was added.  
But it isn't really transposed, so it is not a problem.

> Some other things from your comments:
>
> The maqamGlyphs list needed ratios because it's quoted, and so the
> accidental lookup wasn't doing the correct match against the quoted
> pair.  I'm not sure exactly what the right hand side of a variable
> assignment is, but this seems to work:
>
> maqamGlyphs = #(list
>        (cons FLATFLAT "accidentals.flatflat")
>        (cons FLATHALFFLAT "accidentals.mirroredflat.flat")
>        (cons FLAT "accidentals.flat")
>        (cons HALFFLAT "accidentals.mirroredflat")
>        (cons NATURAL "accidentals.natural")
>        (cons HALFSHARP "accidentals.sharp.slashslash.stem")
>        (cons SHARP "accidentals.sharp")
>        (cons SHARPHALFSHARP  
> "accidentals.sharp.slashslash.stemstemstem")
>        (cons SHARPSHARP "accidentals.doublesharp")
>        )
>
> You say:
>
> %{ Causes error in lilypond/current/scm/lily-library.scm:135:5:
>   In procedure ly:book-process in expression (process-procedure  
> book paper ...)
>   Wrong type (expecting exact integer): -5/3
> %}
>
> I don't see this error.  Looks like a bug's been fixed.

It could have been fixed - it was in the beginning of this year.  
Basically, the limitation on key signatures needed to be lifted by  
admitting arbitrary numbers. Must try again some time.

> > Let me know how to do E53.
>
> I'll look into it.

You can have a look at the attachment.

> > I have not asked for changes, have I?
>
> What is the purpose of your being on the development list?

I wanted to drop off the model I gave to some developer, so that they  
could think it was useful to them. The changes run too deep for it  
making sense on any other list.

But you are not developing LilyPond either, right? - so my attempt  
failed.

> > > Then don't use E24 for Arabic music.
> >
> > You already do.
>
> I didn't think I used either Arabic music or E24 but I'll yield to
> your superior insight.

A guy already implemented an "official" E24 version, so that what  
people is gonna use. The accidentals derive from E24 notation, but  
that is jut about it. In the sound examples I measured on  
maqamworld.com, both Persian E53 and E12 is played.

  Hans