3 ms·
Or for the functionally inclined - S & K combinators: http://en.wikipedia.org/wiki/SKI_combinator_calculus http://en.wikipedia.org/wiki/SKI_combinator_calculus
by arethuza 13y ago
Or for the functionally inclined - S & K combinators:
http://en.wikipedia.org/wiki/SKI_combinator_calculus http://en.wikipedia.org/wiki/SKI_combinator_calculus
I just found this document which describes a Q combinator that can be used to define S & K:
http://www.cs.uu.nl/research/techreps/repo/CS-1989/1989-14.pdf http://www.cs.uu.nl/research/techreps/repo/CS-1989/1989-14.p...
- sp332 13y agoI'm not sure I understand the definition of Q from that paper. Q = \xyz.x right? But that requires assuming lambda calculus, which is small I guess but not what I would call minimal.
- arethuza 13y agoI must admit that I just skimmed that paper - I did see that and thought it looked a bit odd but assumed it was just me!
- walrus 13y agoYou can express it formally in terms of a rewrite rule: If Δ is a derivation ending in an expression of the form α(((Qβ)γ)δ)ι, then Δ followed by the term αβι is a derivation. Writing it as \xyz.x is just a less noisy way of saying that.