Re: circle of fifths

[David Wright]

On Tue 06 Sep 2016 at 19:51:25 (+0200), Thomas Morley wrote:
> 2016-09-06 19:34 GMT+02:00 David Kastrup <address@hidden>:
> > Simon Albrecht <address@hidden> writes:
> >
> >> On 06.09.2016 16:50, Simon Albrecht wrote:
> >>> On 06.09.2016 12:51, Andrew Bernard wrote:
> >>>> Hi Federico,
> >>>>
> >>>> But that tool does not change German variable names into names that
> >>>> are in English, and more meaningful to a speaker of that
> >>>> language. [There is lots of scheme in that circle of fifths code
> >>>> with names that in German to me mean nothing.]
> >>>>
> >>>> So I still think it would be beneficial to have a fully Anglicised
> >>>> version of the code on LSR.
> >>
> >> P.S. I don’t think there should be anything _but_ the anglicised
> >> version in the LSR. It’s a very bad idea to use German variable names
> >> in probably any programming language.
> >
> > At least in mostly functional programming languages.  It might be argued
> > that German is perfectly appropriate for imperative languages.

:) Is that PC?

> Apart from all was said before:
> This snippet was discussed in the german forum and I finally approved
> it, although I was not really satisfied by some parts of it. It's not
> tagged docs.
> 
> I tried to find a balance between getting the snippet improved and not
> scaring away a new contributor.
> Nevertheless I'll drop a note there linking to this thread.
> 
> That said, best would be to find a method to make it work with all
> languages, at least the ones defined in LilyPond.
> Will think about it ...

I'm getting confused. Some of the replies seem to be conflating the
three meanings of "language" that might be meant here.

1) the note names, as in   \key fis \major
2) the text printed on the diagram, as in   \combine \move-markup "es/dis"
3) the variable names in the code, as in    (let* ((mittel (mittel-punkt 
stencil))

1 is always dutch, isn't it. 3 depends on who's writing the code. It
gets a bit inconvenient if it has a lot of non-ascii names in it
(especially when attachments' charsets don't match the email headers).

So 2 is the substantial one, isn't it. Presumably LP and scheme know
what the current language setting is, so they could use a lookup table
based on that. (Presumably there must be such a table buried in LP).
That way, any language would be selectable by running ly over the
snippet's code and running LP on the output.

Cheers,
David.