Re: [GLISS] why the hell all this fuss

[Werner LEMBERG]

>> At that time, I really, really, really cursed lilypond and its
>> frequent syntax changes.
> 
> I think that's Graham's point: syntax changes are bad, so if we have
> to make them (and apparently we still have to), let's do it once and
> for all.  Or at most 1-2 times per decade.

I think that Reinhold is a bit unfair due to his frustration: He has
*certainly* used unstable development versions so that he can use his
own code improvements.  It *must* be allowed by developers to change
the syntax while development still happens.

His very point is that deprecated syntax must either cause a warning
or an error *by running LilyPond itself*.  I fully second that, and it
would be a valuable task to check that for the transition from version
2.14 to 2.16.

> Example: hairpins.  There is no convenient way of specifying
> hairpins that don't align with the notes (you have to use spacer
> rests, which is bad for a number of reasons).  We need to have a
> convenient way.  Adding this will be a syntax change, so let's do it
> now instead of later.

IMHO the `convenient way' is to define a macro; I don't see a need for
a syntax change, since you always need rhythmical arguments to specify
the start and end.

> Another example: vertical hairpins attached to arpeggios (Elaine
> Gould shows them).

Interesting.  I've never seen that before, I believe.  Can you give a
link to an image?

> I don't think we have a simple way of extending our syntax to
> express them - some basic design principles would have to be changed
> a bit, i suppose.  So let's change them now.

As far as I can see, what you want is not syntax changes.  You simply
want new, additional commands.  Or rather, an improved set of standard
macros which come with LilyPond.


    Werner