Re: Remove ambiguity?

On 12/03/2026 19:28, metachromatic wrote:
> The C programming language has a far more stringent set of specifications
> than Lilypond, yet the behavior of C compilers when encountering some C
> code remains undetermined and computer-dependent.

Because that's what the language spec says? Because that's what compiler 
writers implement. C is notorious for being a badly specified language 
where implementors take delight in driving a coach and horses through 
the spec.

> This proves true of every
> programming language as far as I know.

So you've clearly never heard of BPF. Or Safe Rust.
> 
> In fact, this will*always* be the case with*any* programming language,
> Lilypond included.

Except BPF and Safe Rust prove you wrong. And the proof is trivial.

If you'd bothered to write *any Turing Complete* language, instead of 
"any language", then I'd agree with you, but as soon as you insist on 
defining any lower order language using a higher order language, then 
the proof is trivial that the language is (or can be) completely and 
unambiguously specified.

(Trivial as "we can do it", not "it's easy".)

So the question is do we want the lilypond language to be unambiguous, 
or do we want it to be Turing Complete. We can choose either, but we 
can't have both (and like safe/unsafe Rust, we have an escape hatch with 
callouts to other languages).

Cheers,
Wol