Re: preliminary GLISS discussions

[Han-Wen Nienhuys]

On Fri, Aug 31, 2012 at 2:19 PM, Jan Nieuwenhuizen <address@hidden> wrote:
> Wouldn't it be helpful if from the syntax one could tell functions from
> postfix operators simple statements?

Yes, it would if we could start from scratch. Sadly, we are not.

We don't have the distinction, because we did not start from scratch,
and had to move over gradually several built-in constructs to music
functions.

I think many of the newer people on the ilst are missing some
historical context. We used to change lots of syntax details all the
time, and users bitched about it, a lot. I think the past changes were
good choices overall (If you think the current language is a mess, try
to use LilyPond 1.4 for comparison), but they caused considerable
anguish.

>  In most languages function
> invocations are easy to spot.

In many languages, invocations have side-effects. The music functions
were intended to be used without side-effects, so there is a little
difference between a function and a built-in input language construct,
such as \times.

> I think in Perl you can have functions
> look like dead statements, but that's probably just making the argument
> better.

-- 
Han-Wen Nienhuys - address@hidden - http://www.xs4all.nl/~hanwen