Re: Issue 1912 in lilypond: Horizontal spacing depends on stem 'direction

[lilypond]

Updates:
        Labels: Needs-evidence

Comment #2 on issue 1912 by address@hidden: Horizontal spacing depends on  
stem 'direction
http://code.google.com/p/lilypond/issues/detail?id=1912

Ok, so the question we should be asking is not why Lilypond makes a  
different decision based on minuscule reasons: there will always be a  
borderline where it switches behavior.

The question is whether the borderline is sensible, namely at a place where  
we can say that either choice looks about equally good.  When we can say  
that one of those choices looks much better for an identifiable reason, we  
have to try to make that reason accessible to Lilypond's scoring algorithm.

So I am marking this as Needs-evidence.  If no evidence is delivered that  
one of the choices is significantly worse than the other, rather than just  
different, I'd recommend closing this as Invalid.