Re: missing lilypond-tex-metrics.tex

[Graham Percival]

On 31-May-05, at 7:53 PM, Aaron Mehl wrote:
> > I think we use "should be" and "must" fairly
> > interchangeably.
>   As a techwriter/user this confuses me. should be is
> a much weaker expression, like it doesn't have to but
> if you want.. must says that it is a requirement to
> have on of these.

I really don't think it makes a big difference.  If you want to
submit a patch that changes all "should be"s to "must"s,
I won't complain.

> > > then it says:
> > > "A .ly file contains any number of toplevel
> > > expressions, where a toplevel expression is one of
> > the
> > > following"
>
> You mean to say that if I write a lilypond file
> with a header and a'4 it will parse? I am trying here
> to clarify what the intent of the documentation is
> here, not be picky per se.

No, because:
a'4
is not a top-level expression.  This
{ a'4 }
is an expression; if it occurs at the top-level then it'd be
a top-level expression.

> A user doesn't want to know what he may or should do,
> but what he needs to get this up and running. What I

That's what the tutorial is for.  I still don't understand what
the difficulty is.

> mean is beating around the bush doesn't help. If this
> is a page detailing lilypond file structure it should
> do that specifically not with shoulds and mights. If
> there are different possiblilies they need to be laid
> out.

...  so we should just change all "shoulds" to "must" on that
doc page?  I really don't think that makes a big difference.  I only
see two "should"s, anyway.

I'm not adverse to replacing that section of the docs.  A few
people have asked for more formal definitions of lilypond input
syntax.  If you wrote such a page, it could help some new users.

- Graham