Lukas-Fabian Moser <[email protected]> writes:
>> { \startMeasureCount s1*2\< s\> s\< s\> <>\! }
>
> Somehow I was never happy with this use of spacer rests for dynamic
> parts (even though it is recommended in the documentation).
>
> My issue is that the "natural" way of reading/writing/speaking would be:
>
> "Start a crescendo, then wait two bars; start a decrescendo; then wait
> another bar and terminate it."
>
> This can be expressed literally by:
>
> <>\< s1*2 <>\> s1 <>\!
>
> In contrast, the method from the documentation
>
> s1*2\< s1\> <>\!
>
> translates as:
>
> "Before skipping two bars, start a crescendo; then, before skipping
> another bar, start a decrescendo" etc. Mental zigzag.
>
> I have a branch on my computer where I propose recommending the use of
> empty chords systematically in the documentation, but I never got
> around to finishing it yet.
Just as a case in point: here's some I entered few days ago:
<< \repeat percent 32
<< { sn2.:32 | } \\ { bd4 r r } >>
{ <>\p \skip 2.*16 <>\mf \skip 2.*4 |
<>\cresc \skip 2.*3 |
<>\f \skip 2.*9 |
}
>>
...
<>\<2.:32 |
2.:32 <>\! | }
Of course there is an additional reason not to use post-events on
tremoli here: I don't fancy what the \articulate script does with that
combination.
At any rate, <>\! is a rather frequent necessity to keep stuff where it
belongs (the closing brace is from a \repeat volta for example).
So I find myself littering a lot of my input with <> in the process of
structuring it in a way I find natural to work with.
--
David Kastrup
