Re: Proposed new available and recommended behavior of \relative

[Paul Morris]

On Mar 9, 2013, at 4:47 AM, David Kastrup <address@hidden> wrote:

> So that's the next step: opening the door on \relative { } again, or rather a 
> different door with the same door handle.

Interesting discussion.  I like the new/proposed behavior for \relative { ... } 
 (without reference pitch), and would probably use it myself.  

As someone mentioned, it might be helpful to explain things in the docs 
something like the following:  If there is no explicit reference pitch, the 
first note defaults to being relative to f (or else the middle note of the 
current scale if the scale has been changed).  And also say that this happens 
to correspond with seeing the first pitch as written in absolute notation (as a 
quick way of determining the pitch of the first note).  So both ways of looking 
at it work.  (Am I understanding this right?)

That would address concerns about the perception of mixing absolute and 
relative notation inside the { ... }.

OTOH, it is pretty straightforward to say that the first note after \relative 
is interpreted in absolute terms as a starting point / reference pitch for 
subsequent notes, whether it is inside or outside of the { ... }.


> Then there are two more questions:
> 
> a) should the LilyPond codebase walk through that door?

I would say yes.  But seems like there is another question:

a2) should \relative { ... }  be the default/recommended approach as presented 
in the docs?

I'm leaning slightly towards yes on this, but it's probably worth sitting with 
it for a bit.


> b) should convert-ly make user code walk through that door once?

Hmmm...  If people have been using an explicit reference pitch, nothing changes 
in that case.  So it's just if they have *not* been using one (which has been 
deprecated, right?) that convert-ly will need to either 

1) possibly change the octave of the first pitch inside { ... }
2) possibly add an explicit reference pitch.  

I'm not sure which is better, and maybe this is getting ahead of ourselves.


On Mar 9, 2013, at 5:40 AM, David Kastrup <address@hidden> wrote:

> a) stop any further use of the current \relative { ... }
>   That's issue 3231.
> b) Implement new proposed behavior for \relative { ... }.
>   That's the ly/music-functions-init.ly part of issue 3229.
> c) give this proposed behavior equal coverage in the passages talking
>   about absolute and relative pitches in Learning and Notation.
>   Impact: a handful of paragraphs.
> d) wait and see.

Sounds good.  

-Paul