Re: Substitute for s1*0

[Trevor Daniels]

Nicolas Sceaux wrote Monday, May 07, 2012 8:32 PM

> Le 7 mai 2012 à 13:58, David Kastrup a écrit :
>
> > \relative c' {
> > e2\p\< d\> s1*0\!
> > } \addlyrics { Oh no }
> >
> > \relative c' {
> > e2\p\< d\> <>\!
> > } \addlyrics { Oh yes }
>
> I think that closes the s1*0 vs. <> debate.
> Because of its unexpected side effects, the s1*0 idiom must be banished.

Yes, I now agree.  We can't continue to advocate s1*0
in the docs now we are aware of these pitfalls.

> Now that this is settled, this leaves the question of using <> (which
> already exists and works) or another new construct, eg. n or z or \null,
> as a replacement for the broken s1*0 idiom.  Please let me quote (or
> paraphrase) Montesquieu:
>
>  It is sometimes necessary to change certain laws, but the case is rare,
>  and should be undertaken with trembling hands.
>
> This perfectly applies to parser changes in general, and to this case
> in particular: n or z or \null is not at all a necessary change.
> <> already exists, works on any LilyPond version, and has understandable
> semantics once explicited in the doc.  Is there still a debate at this
> point?

Not from me: I expressed my opposition to a parser change already.

So at least three of us are now totally in agreement.  We explain
the semantics of <> and change all examples to use it.

Trevor