YES Succeeded in reading "/export/starexec/sandbox2/benchmark/theBenchmark.ari". (CONDITIONTYPE ORIENTED) (VAR y x x1) (RULES f(c(x),c(c(y))) -> a(a(x)) | c(f(x,y)) == c(a(b)) f(c(c(c(x))),y) -> a(y) | c(f(c(x),c(c(y)))) == c(a(a(b))) h(b) -> b h(a(a(x))) -> a(b) | h(x) == b ) No "->="-rules. Decomposed conditions and removed infeasible rules if possible. (CONDITIONTYPE ORIENTED) (VAR y x x1) (RULES h(b) -> b h(a(a(x))) -> a(b) | h(x) == b ) (VAR y x1) (CONDITION c(f(c(c(x1)),y)) == c(a(b)), c(f(c(x1),c(c(c(c(y)))))) == c(a(a(b))) ) Optimized the infeasibility problem if possible. (VAR y x1) (CONDITION f(c(c(x1)),y) == a(b), f(c(x1),c(c(c(c(y))))) == a(a(b)) ) This is ultra-RL and deterministic. This is operationally terminating and confluent. (RTG_of_NARROWINGTREE (START Gamma[f(c(c(x1)),y) == a(b) : { e, 1, 1.1, 1.1.1 } & f(c(x1),c(c(c(c(y))))) == a(a(b)) : { e, 1, 1.1, 1.2, 1.2.1, 1.2.1.1, 1.2.1.1.1 }] ) (NONTERMINALS Gamma[f(c(c(x1)),y) == a(b) : { e, 1, 1.1, 1.1.1 } & f(c(x1),c(c(c(c(y))))) == a(a(b)) : { e, 1, 1.1, 1.2, 1.2.1, 1.2.1.1, 1.2.1.1.1 }] ) (RULES Gamma[f(c(c(x1)),y) == a(b) : { e, 1, 1.1, 1.1.1 } & f(c(x1),c(c(c(c(y))))) == a(a(b)) : { e, 1, 1.1, 1.2, 1.2.1, 1.2.1.1, 1.2.1.1.1 }] -> EmptySet ) ) YES