There's a model of computation called 'interaction nets' / 'interaction calculus', which reduces in a more physically-meaningful, local, topologically-smooth way.
I.e. you can see from these animations that LC reductions have some "jumping" parts. And that does reflect LC nature, as a reduction 'updates' many places at once.
IN basically fixes this problem. And this locality can enable parallelism. And there's an easy way to translate LC to IN, as far as I understand.
I'm a noob, but I feel like INs are severely under-rated. I dunno if there's any good interaction net animations. I know only one person who's doing some serious R&D with interaction nets - that's Victor Taelin.
You can enter (λn.n(λc.λa.λb.cb(λf.λx.f(afx)))Fn0)7 to compute the function Col' from [1] to 7, resulting in (3*7+1)/2 = 11. Unfortunately, this visualization is much less insightful than showing the 7 successive succ&swap operations:
I.e. you can see from these animations that LC reductions have some "jumping" parts. And that does reflect LC nature, as a reduction 'updates' many places at once.
IN basically fixes this problem. And this locality can enable parallelism. And there's an easy way to translate LC to IN, as far as I understand.
I'm a noob, but I feel like INs are severely under-rated. I dunno if there's any good interaction net animations. I know only one person who's doing some serious R&D with interaction nets - that's Victor Taelin.
While easy, it sadly doesn't preserve semantics. Specifically, when you duplicate a term that ends up duplicating itself, results will diverge.
There exist more involved semantics preserving translations, using so-called croissants and brackets, or with the recent rephrased approach of [1].
[1] https://arxiv.org/abs/2505.20314