Re: For future 2.14 doc discussion -- increasingly realistic musical examples

[Trevor Bača]

On 8/30/07, Graham Percival <address@hidden> wrote:
> Trevor Bača wrote:
> > The second example was infinitely easier: we just needed something to
> > show a text script complicated enough to require a markup (which is
> > almost everything). The indication, is, of course, Debussy's famous
> > marking at the beginning of Des pas sur la neige.
> >
> Let's ignore the \fatText issue for now.  If I understand you correctly,
> the proposal is to replace the current "first example" of text scripts:
>
>     c'4^\markup { bla \bold bla }
>
> with something like this?
>
>     \key f \major
>     \override Staff TimeSignature #'style = #'numbered
>     <<
>        { \times 2/3{d16[ e8 ~ } e] ~ e4 } \\
>        { d2--_\markup{ \italics { \column
>             \line{ Ce rhythme doit avoid la valuer sonore} }
>             \line {d'un fond de paysage triste et glac\'e } }}
>     >>
>
> ?   (that probably won't compile)
>
> In your pngs, you showed the input for the "before" pictures, but not
> for the "after" pictures.  But I really think the manual should print
> the input, and as you can see, creating beautiful musical examples
> increases the complexity _enormously_.  An experienced lilypond user can
> glance at the above code and pick out the normal
> rhythm/note/tie/polyphony notation, and pay attention to the new \markup
> command... but no newbie is going to spend enough time deciphering this
> input.
>
>
> I think I've misunderstood something.  Is this just supposed to whet
> people's appetites, to inspire them to read the rest of the subsections?
>
> ... hmm, I might see a use for that.  We could have one really fancy
> example, *without* input code, in the "root" node of each section.

Hi Graham,

Here's example 1's input:

%%% EX 1 %%%

\new Staff {
        \key b \major
        \time 12/16
        b8 \staccato \p \> [
        fis'8 \staccato
        cis'8 \staccato
        e'8 \staccato
        gis'8 \staccato
        \fatText
        b'8 \! \laissezVibrer ] ^ "laissez vibrer"
}

%%% EX 1 %%%


And here's example 2's input:

\new Staff {
        \key d \minor
        \override Staff.TimeSignature #'style = #'()
        \time 4/4
        <<
        {
                \voiceOne
                \times 2/3 {
                        d'16
                        e'8 ~
                }
                e'8 ~
                e'4
                \times 2/3 {
                        e'16
                        f'8 ~
                }
                f'8 ~
                f'4
        }
                \\
        {
                \voiceTwo
                d'2 \tenuto
                        _ \markup \italic \column {
                                "Ce rhythme doit avoir la valeur sonore"
                                "d'un fond de paysage triste et glacé"
                           }
                d'2 \tenuto
        }
        >>
}

%%% END %%%


Example 2 has more complex input. That is, of course, what it takes to
get much more beautiful output when working with a descriptor language
(rather than a GUI).

Looking through the Finale and Sibelius manuals again, it's easy for
them: to get pictures of real music all they have to do is take a
screenshot (and then write some docs that say "click on this and
double click on that"). For us, on the other hand, to get real music
we have to type some more.

The fear about increased (input) complexity is totally valid. And
still at some point it would be quite lovely for users to see
beautiful music as they leaf through the manual. Looking at Lily's
output is a pleasure.

Also, "whetting" is a good way to look at things. I do actually think
that eventually a majority of the one-measure manual examples should
rise to the level of actual music. But the idea of a single,
worked-out, header example in each section may be a nice compromise,
too.



-- 
Trevor Bača
address@hidden