Re: [GLISS] differentiating pre/post/neutral commands

[Joseph Rushton Wakeling]

On 11/09/12 13:04, David Kastrup wrote:
> Basically every construct that we would be tempted to use <> or s1*0 for
> occasionally is one that is not really attached to a note, but rather to
> a moment in time.  You can put it in parallel music without changing
> results.  Most articulations with a shorthand can be attached to
> individual notes in a chord: those are really intrinsically attached to
> the note before, and it makes sense keeping that even if per-chord
> articulations can be placed into parallel music.  But things like ( ) \(
> \) [ ] \p \< \! \> all happen at a moment in time in a voice.  Why is a
> tempo change a separate event, but a dynamic change isn't?

This is starting to sound a bit like SCORE's approach, where various different 
aspects of the music (notes, phrasing, dynamics, ...) are described by separate, 
parallel sequences: see,
http://www.ccarh.org/courses/253/handout/scoreinput/

Since you mentioned articulations: one of my personal bugbears with Lilypond has 
been the case of an articulation that occurs part-way through a given note. 
Consider e.g. the case of a half-note with a \turn that should occur on the 
second beat of that half-note.  It's something which (unless things have changed 
in more recent versions, which I'm not aware of) is surprisingly difficult to 
achieve _effectively_ in Lilypond, because the presence of the offset 
articulation is not taken into account when calculating horizontal spacing.