Problem: f(x,y) -> x f(x,y) -> f(x,g(y)) g(x) -> h(x) F(g(x),x) -> F(x,g(x)) F(h(x),x) -> F(x,h(x)) Proof: sorted: (order) 0:f(x,y) -> x f(x,y) -> f(x,g(y)) g(x) -> h(x) 1:g(x) -> h(x) F(g(x),x) -> F(x,g(x)) F(h(x),x) -> F(x,h(x)) ----- sorts [0>3, 0>4, 1>2, 2>4] sort attachment (non-strict) f : 3 x 4 -> 0 g : 4 -> 4 h : 4 -> 4 F : 2 x 4 -> 1 ----- 0:f(x,y) -> x f(x,y) -> f(x,g(y)) g(x) -> h(x) Church Rosser Transformation Processor (critical pair closing system, Thm 2.11): f(x23,x22) -> x23 critical peaks: joinable Matrix Interpretation Processor: dim=1 interpretation: [f](x0, x1) = 4x0 + 2x1 + 4 orientation: f(x23,x22) = 2x22 + 4x23 + 4 >= x23 = x23 problem: Qed 1:g(x) -> h(x) F(g(x),x) -> F(x,g(x)) F(h(x),x) -> F(x,h(x)) Church Rosser Transformation Processor: strict: weak: critical peaks: 1 F(h(x),x) <-0|0[]- F(g(x),x) -1|[]-> F(x,g(x)) Redundant Rules Transformation: g(x) -> h(x) F(h(x),x) -> F(x,h(x)) Qed (SakaiOyamaguchiOgawa14)