Re: Dynamic mark at the end of a bar

[Simon Albrecht]

Am 27.05.2015 um 22:03 schrieb David Sumbler:
> But both of these are "work-arounds" for something which is actually
> fairly normal musical notation.  They are work-arounds because one
> method implies that a note is not really the length that it appears to
> be, and the other method implies that the dynamics actually belong to
> another, silent voice.
No, it doesn’t. They are in separate parallel music expressions, one of 
which contains only skips and dynamics and the other contains everything 
else. But since there is no \new Voice (or \\ as a shorthand) here, both 
are read into one voice and one _stream of events_.
>    Neither of these is true in reality.
>
> Ideally one shouldn't have to use trial and error, varying the length of
> the second invisible rest, to get an acceptable appearance.  We know
> exactly where the final dynamic should go: it should be immediately
> before the bar-line.
>
> Perhaps at some stage the Lilypond developers could look at the
> possibility of having a method of adding a dynamic (or text) for cases
> like this, so that it is automatically right-aligned to a note or rest.
> Additionally, if the note or rest comes immediately after a barline,
> then the dynamic/text should be right-aligned to the barline itself.  I
> would suggest that a suitable input method would be to use a slash (/)
> instead of or as well as the usual backslash (\).
>
> I am not criticising Lilypond, which is magnificent.  But it does seem
> unfortunate that at the moment there is no way of directly inputting a
> perfectly normal, if not particularly common, bit of standard notation.
I think the point here is very fundamental: Lilypond input language is 
designed to represent musical content. So while we are used (as all or 
most people not using Lilypond have been) to just graphically placing a 
dynamic sign using pen and paper or a WYSIWYG notation program, Lilypond 
follows an entirely different approach. Which especially in this case is 
a problem. (And perhaps it will emerge not to be a dilemma).
So, until we’ve found a better solution matching the principles of 
Lilypond _and_ providing the traditional ‘inexact’, intuitive dynamics 
placement, we’ll go with the methods sketched by David.

Regards, Simon