Re: Accidentals: Unwanted naturals

[David Raleigh Arnold]

On Friday 28 August 2009, David Rogers wrote:
> On Fri, Aug 28, 2009 at 10:58, Kieren
> MacMillan<address@hidden> wrote:
> > Hi David (et al),
> >
> > Just to be absolutely clear, the fallacy in your argument lies in 
the
> > following statement:
> >
> >> It's necessary to consider the sound of the music,
> >> *and not the conventional rules of printed scores*
> >> when doing Lilypond pitch input.
> >
> >
> > Quite the contrary, the "conventional rules of printed scores" DO 
consider
> > (incorporate) "the sound of the music" — that's why the Western 
notation
> > system works as well as it does (despite some flaws/shortcomings, 
and
> > countless attempts to replace it with a "superior" alternative).
> >
> > Let's start by considering the CRoPS with respect to a simple 
notation
> > example. If the key signature is D major (i.e., two sharps), and the 
pitch
> > class [!!] being displayed is the top line of the treble clef (i.e., 
F),
> > then the CRoPS tells us that the actual pitch that should be 
performed is an
> > F-sharp (i.e., fis'').
> >
> > Now, let's "do Lilypond pitch input" for this same example. You want
> > Lilypond to output an F-sharp at the top of the treble clef, and 
display the
> > result "in D major" (i.e., with a D major key signature).
> >
> > Step 1 is to define/list the pitch(es) you want engraved:
> >   theMusic = { fis'' }
> >
> > Step 2 is to build the score, with clef and key signature:
> >    \score { \new Staff << \key d \major \clef treble \theMusic >> }
> >
> > Doing the same thing *without* the pitch alteration (sharp) in 
theMusic
> > definition exposes the fundamental problem with 
a "follow-the-key-signature"
> > approach.
> 
> 
> I know that. I think Lilypond is operating correctly here, that this
> part of the code should be kept as is with nothing added, and that
> those users who wish it operated differently are making a mistake, for
> exactly the reasons you've just pointed out.
> 
> HOWEVER, I think it's necessary to explain this issue *in their terms*
> in the documentation, so that they can stop being confused by a
> perfectly good (but logically backwards *to them*) implementation,
> letting them get on with their work.

I wrote a sed script that does the job some time ago.  To implement
\followKeySignature would be much easier than the opponents imagine.
I do not find it confusing to any degree, and it saves typing, and
it makes sense to me because it makes the input more similar
to the notation.  Most would agree on the desirability of that,
or does anyone think that it would be good to get rid of barchecks?
But in this instance, the majority of coders line up in opposition.
You have shouted down the users, but convinced none.  Why?  Because
you are wrong.  Regards, daveA

-- 
For beginners: very easy guitar music, solos, duets, exercises. Early
intermediate guitar solos. One best scale set for all guitarists.
http://www.openguitar.com/scalescomparison.html ::: plus new and
better chord and arpeggio exercises.  http://www.openguitar.com