Problem AProVE 07 thiemann08

Tool CaT

Execution TimeUnknown
Answer
MAYBE
InputAProVE 07 thiemann08

stdout:

MAYBE

Problem:
 empty(nil()) -> true()
 empty(cons(x,l)) -> false()
 head(cons(x,l)) -> x
 tail(nil()) -> nil()
 tail(cons(x,l)) -> l
 rev(nil()) -> nil()
 rev(cons(x,l)) -> cons(rev1(x,l),rev2(x,l))
 last(x,l) -> if(empty(l),x,l)
 if(true(),x,l) -> x
 if(false(),x,l) -> last(head(l),tail(l))
 rev2(x,nil()) -> nil()
 rev2(x,cons(y,l)) -> rev(cons(x,rev2(y,l)))

Proof:
 Open

Tool IRC1

Execution TimeUnknown
Answer
MAYBE
InputAProVE 07 thiemann08

stdout:

MAYBE

Tool IRC2

Execution TimeUnknown
Answer
MAYBE
InputAProVE 07 thiemann08

stdout:

MAYBE

'Fastest (timeout of 60.0 seconds)'
-----------------------------------
Answer:           MAYBE
Input Problem:    innermost runtime-complexity with respect to
  Rules:
    {  empty(nil()) -> true()
     , empty(cons(x, l)) -> false()
     , head(cons(x, l)) -> x
     , tail(nil()) -> nil()
     , tail(cons(x, l)) -> l
     , rev(nil()) -> nil()
     , rev(cons(x, l)) -> cons(rev1(x, l), rev2(x, l))
     , last(x, l) -> if(empty(l), x, l)
     , if(true(), x, l) -> x
     , if(false(), x, l) -> last(head(l), tail(l))
     , rev2(x, nil()) -> nil()
     , rev2(x, cons(y, l)) -> rev(cons(x, rev2(y, l)))}

Proof Output:    
  None of the processors succeeded.
  
  Details of failed attempt(s):
  -----------------------------
    1) 'wdg' failed due to the following reason:
         Transformation Details:
         -----------------------
           We have computed the following set of weak (innermost) dependency pairs:
           
             {  1: empty^#(nil()) -> c_0()
              , 2: empty^#(cons(x, l)) -> c_1()
              , 3: head^#(cons(x, l)) -> c_2()
              , 4: tail^#(nil()) -> c_3()
              , 5: tail^#(cons(x, l)) -> c_4()
              , 6: rev^#(nil()) -> c_5()
              , 7: rev^#(cons(x, l)) -> c_6(rev2^#(x, l))
              , 8: last^#(x, l) -> c_7(if^#(empty(l), x, l))
              , 9: if^#(true(), x, l) -> c_8()
              , 10: if^#(false(), x, l) -> c_9(last^#(head(l), tail(l)))
              , 11: rev2^#(x, nil()) -> c_10()
              , 12: rev2^#(x, cons(y, l)) -> c_11(rev^#(cons(x, rev2(y, l))))}
           
           Following Dependency Graph (modulo SCCs) was computed. (Answers to
           subproofs are indicated to the right.)
           
             ->{8,10}                                                    [       MAYBE        ]
                |
                `->{9}                                                   [         NA         ]
             
             ->{7,12}                                                    [     inherited      ]
                |
                `->{11}                                                  [         NA         ]
             
             ->{6}                                                       [    YES(?,O(1))     ]
             
             ->{5}                                                       [    YES(?,O(1))     ]
             
             ->{4}                                                       [    YES(?,O(1))     ]
             
             ->{3}                                                       [    YES(?,O(1))     ]
             
             ->{2}                                                       [    YES(?,O(1))     ]
             
             ->{1}                                                       [    YES(?,O(1))     ]
             
           
         
         Sub-problems:
         -------------
           * Path {1}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(tail^#) = {},
                 Uargs(rev^#) = {}, Uargs(c_6) = {}, Uargs(rev2^#) = {},
                 Uargs(last^#) = {}, Uargs(c_7) = {}, Uargs(if^#) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                nil() = [0]
                        [0]
                        [0]
                true() = [0]
                         [0]
                         [0]
                cons(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                false() = [0]
                          [0]
                          [0]
                head(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                tail(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                rev(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev1(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                rev2(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                last(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                if(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                empty^#(x1) = [0 0 0] x1 + [0]
                              [0 0 0]      [0]
                              [0 0 0]      [0]
                c_0() = [0]
                        [0]
                        [0]
                c_1() = [0]
                        [0]
                        [0]
                head^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_2() = [0]
                        [0]
                        [0]
                tail^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_3() = [0]
                        [0]
                        [0]
                c_4() = [0]
                        [0]
                        [0]
                rev^#(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                c_5() = [0]
                        [0]
                        [0]
                c_6(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev2^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                last^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                c_7(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                if^#(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                c_8() = [0]
                        [0]
                        [0]
                c_9(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                c_10() = [0]
                         [0]
                         [0]
                c_11(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 3'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    innermost DP runtime-complexity with respect to
               Strict Rules: {empty^#(nil()) -> c_0()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(empty^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                nil() = [2]
                        [2]
                        [2]
                empty^#(x1) = [0 2 0] x1 + [7]
                              [2 2 0]      [3]
                              [2 2 2]      [3]
                c_0() = [0]
                        [1]
                        [1]
           
           * Path {2}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(tail^#) = {},
                 Uargs(rev^#) = {}, Uargs(c_6) = {}, Uargs(rev2^#) = {},
                 Uargs(last^#) = {}, Uargs(c_7) = {}, Uargs(if^#) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                nil() = [0]
                        [0]
                        [0]
                true() = [0]
                         [0]
                         [0]
                cons(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                false() = [0]
                          [0]
                          [0]
                head(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                tail(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                rev(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev1(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                rev2(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                last(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                if(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                empty^#(x1) = [0 0 0] x1 + [0]
                              [0 0 0]      [0]
                              [0 0 0]      [0]
                c_0() = [0]
                        [0]
                        [0]
                c_1() = [0]
                        [0]
                        [0]
                head^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_2() = [0]
                        [0]
                        [0]
                tail^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_3() = [0]
                        [0]
                        [0]
                c_4() = [0]
                        [0]
                        [0]
                rev^#(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                c_5() = [0]
                        [0]
                        [0]
                c_6(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev2^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                last^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                c_7(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                if^#(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                c_8() = [0]
                        [0]
                        [0]
                c_9(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                c_10() = [0]
                         [0]
                         [0]
                c_11(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 3'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    innermost DP runtime-complexity with respect to
               Strict Rules: {empty^#(cons(x, l)) -> c_1()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(cons) = {}, Uargs(empty^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                cons(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [2]
                               [0 0 0]      [0 0 0]      [2]
                               [0 0 0]      [0 0 0]      [2]
                empty^#(x1) = [0 2 0] x1 + [7]
                              [2 2 0]      [3]
                              [2 2 2]      [3]
                c_1() = [0]
                        [1]
                        [1]
           
           * Path {3}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(tail^#) = {},
                 Uargs(rev^#) = {}, Uargs(c_6) = {}, Uargs(rev2^#) = {},
                 Uargs(last^#) = {}, Uargs(c_7) = {}, Uargs(if^#) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                nil() = [0]
                        [0]
                        [0]
                true() = [0]
                         [0]
                         [0]
                cons(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                false() = [0]
                          [0]
                          [0]
                head(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                tail(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                rev(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev1(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                rev2(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                last(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                if(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                empty^#(x1) = [0 0 0] x1 + [0]
                              [0 0 0]      [0]
                              [0 0 0]      [0]
                c_0() = [0]
                        [0]
                        [0]
                c_1() = [0]
                        [0]
                        [0]
                head^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_2() = [0]
                        [0]
                        [0]
                tail^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_3() = [0]
                        [0]
                        [0]
                c_4() = [0]
                        [0]
                        [0]
                rev^#(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                c_5() = [0]
                        [0]
                        [0]
                c_6(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev2^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                last^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                c_7(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                if^#(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                c_8() = [0]
                        [0]
                        [0]
                c_9(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                c_10() = [0]
                         [0]
                         [0]
                c_11(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 3'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    innermost DP runtime-complexity with respect to
               Strict Rules: {head^#(cons(x, l)) -> c_2()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(cons) = {}, Uargs(head^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                cons(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [2]
                               [0 0 0]      [0 0 0]      [2]
                               [0 0 0]      [0 0 0]      [2]
                head^#(x1) = [0 2 0] x1 + [7]
                             [2 2 0]      [3]
                             [2 2 2]      [3]
                c_2() = [0]
                        [1]
                        [1]
           
           * Path {4}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(tail^#) = {},
                 Uargs(rev^#) = {}, Uargs(c_6) = {}, Uargs(rev2^#) = {},
                 Uargs(last^#) = {}, Uargs(c_7) = {}, Uargs(if^#) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                nil() = [0]
                        [0]
                        [0]
                true() = [0]
                         [0]
                         [0]
                cons(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                false() = [0]
                          [0]
                          [0]
                head(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                tail(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                rev(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev1(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                rev2(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                last(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                if(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                empty^#(x1) = [0 0 0] x1 + [0]
                              [0 0 0]      [0]
                              [0 0 0]      [0]
                c_0() = [0]
                        [0]
                        [0]
                c_1() = [0]
                        [0]
                        [0]
                head^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_2() = [0]
                        [0]
                        [0]
                tail^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_3() = [0]
                        [0]
                        [0]
                c_4() = [0]
                        [0]
                        [0]
                rev^#(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                c_5() = [0]
                        [0]
                        [0]
                c_6(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev2^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                last^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                c_7(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                if^#(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                c_8() = [0]
                        [0]
                        [0]
                c_9(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                c_10() = [0]
                         [0]
                         [0]
                c_11(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 3'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    innermost DP runtime-complexity with respect to
               Strict Rules: {tail^#(nil()) -> c_3()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(tail^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                nil() = [2]
                        [2]
                        [2]
                tail^#(x1) = [0 2 0] x1 + [7]
                             [2 2 0]      [3]
                             [2 2 2]      [3]
                c_3() = [0]
                        [1]
                        [1]
           
           * Path {5}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(tail^#) = {},
                 Uargs(rev^#) = {}, Uargs(c_6) = {}, Uargs(rev2^#) = {},
                 Uargs(last^#) = {}, Uargs(c_7) = {}, Uargs(if^#) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                nil() = [0]
                        [0]
                        [0]
                true() = [0]
                         [0]
                         [0]
                cons(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                false() = [0]
                          [0]
                          [0]
                head(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                tail(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                rev(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev1(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                rev2(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                last(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                if(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                empty^#(x1) = [0 0 0] x1 + [0]
                              [0 0 0]      [0]
                              [0 0 0]      [0]
                c_0() = [0]
                        [0]
                        [0]
                c_1() = [0]
                        [0]
                        [0]
                head^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_2() = [0]
                        [0]
                        [0]
                tail^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_3() = [0]
                        [0]
                        [0]
                c_4() = [0]
                        [0]
                        [0]
                rev^#(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                c_5() = [0]
                        [0]
                        [0]
                c_6(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev2^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                last^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                c_7(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                if^#(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                c_8() = [0]
                        [0]
                        [0]
                c_9(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                c_10() = [0]
                         [0]
                         [0]
                c_11(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 3'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    innermost DP runtime-complexity with respect to
               Strict Rules: {tail^#(cons(x, l)) -> c_4()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(cons) = {}, Uargs(tail^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                cons(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [2]
                               [0 0 0]      [0 0 0]      [2]
                               [0 0 0]      [0 0 0]      [2]
                tail^#(x1) = [0 2 0] x1 + [7]
                             [2 2 0]      [3]
                             [2 2 2]      [3]
                c_4() = [0]
                        [1]
                        [1]
           
           * Path {6}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(tail^#) = {},
                 Uargs(rev^#) = {}, Uargs(c_6) = {}, Uargs(rev2^#) = {},
                 Uargs(last^#) = {}, Uargs(c_7) = {}, Uargs(if^#) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                nil() = [0]
                        [0]
                        [0]
                true() = [0]
                         [0]
                         [0]
                cons(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                false() = [0]
                          [0]
                          [0]
                head(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                tail(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                rev(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev1(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                rev2(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                last(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                if(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                empty^#(x1) = [0 0 0] x1 + [0]
                              [0 0 0]      [0]
                              [0 0 0]      [0]
                c_0() = [0]
                        [0]
                        [0]
                c_1() = [0]
                        [0]
                        [0]
                head^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_2() = [0]
                        [0]
                        [0]
                tail^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_3() = [0]
                        [0]
                        [0]
                c_4() = [0]
                        [0]
                        [0]
                rev^#(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                c_5() = [0]
                        [0]
                        [0]
                c_6(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev2^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                last^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                c_7(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                if^#(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                c_8() = [0]
                        [0]
                        [0]
                c_9(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                c_10() = [0]
                         [0]
                         [0]
                c_11(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 3'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    innermost DP runtime-complexity with respect to
               Strict Rules: {rev^#(nil()) -> c_5()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(rev^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                nil() = [2]
                        [2]
                        [2]
                rev^#(x1) = [0 2 0] x1 + [7]
                            [2 2 0]      [3]
                            [2 2 2]      [3]
                c_5() = [0]
                        [1]
                        [1]
           
           * Path {7,12}: inherited
             ----------------------
             
             This path is subsumed by the proof of path {7,12}->{11}.
           
           * Path {7,12}->{11}: NA
             ---------------------
             
             The usable rules for this path are:
             
               {  rev2(x, nil()) -> nil()
                , rev2(x, cons(y, l)) -> rev(cons(x, rev2(y, l)))
                , rev(nil()) -> nil()
                , rev(cons(x, l)) -> cons(rev1(x, l), rev2(x, l))}
             
             The weight gap principle does not apply:
               The input cannot be shown compatible
             Complexity induced by the adequate RMI: MAYBE
             
             We have not generated a proof for the resulting sub-problem.
           
           * Path {8,10}: MAYBE
             ------------------
             
             The usable rules for this path are:
             
               {  empty(nil()) -> true()
                , empty(cons(x, l)) -> false()
                , head(cons(x, l)) -> x
                , tail(nil()) -> nil()
                , tail(cons(x, l)) -> l}
             
             The weight gap principle does not apply:
               The input cannot be shown compatible
             Complexity induced by the adequate RMI: MAYBE
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 3'
             --------------------------------------
             Answer:           MAYBE
             Input Problem:    innermost runtime-complexity with respect to
               Rules:
                 {  last^#(x, l) -> c_7(if^#(empty(l), x, l))
                  , if^#(false(), x, l) -> c_9(last^#(head(l), tail(l)))
                  , empty(nil()) -> true()
                  , empty(cons(x, l)) -> false()
                  , head(cons(x, l)) -> x
                  , tail(nil()) -> nil()
                  , tail(cons(x, l)) -> l}
             
             Proof Output:    
               The input cannot be shown compatible
           
           * Path {8,10}->{9}: NA
             --------------------
             
             The usable rules for this path are:
             
               {  empty(nil()) -> true()
                , empty(cons(x, l)) -> false()
                , head(cons(x, l)) -> x
                , tail(nil()) -> nil()
                , tail(cons(x, l)) -> l}
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(tail^#) = {},
                 Uargs(rev^#) = {}, Uargs(c_6) = {}, Uargs(rev2^#) = {},
                 Uargs(last^#) = {1, 2}, Uargs(c_7) = {1}, Uargs(if^#) = {1},
                 Uargs(c_9) = {1}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0 0] x1 + [2]
                            [0 0 0]      [3]
                            [0 0 0]      [3]
                nil() = [2]
                        [0]
                        [0]
                true() = [1]
                         [1]
                         [1]
                cons(x1, x2) = [1 1 0] x1 + [1 3 1] x2 + [2]
                               [0 1 0]      [0 1 0]      [0]
                               [0 0 1]      [0 0 0]      [0]
                false() = [1]
                          [1]
                          [1]
                head(x1) = [3 3 3] x1 + [3]
                           [3 3 3]      [3]
                           [0 0 1]      [3]
                tail(x1) = [2 0 0] x1 + [0]
                           [0 2 0]      [0]
                           [1 0 0]      [0]
                rev(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev1(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                rev2(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                last(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                if(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                empty^#(x1) = [0 0 0] x1 + [0]
                              [0 0 0]      [0]
                              [0 0 0]      [0]
                c_0() = [0]
                        [0]
                        [0]
                c_1() = [0]
                        [0]
                        [0]
                head^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_2() = [0]
                        [0]
                        [0]
                tail^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_3() = [0]
                        [0]
                        [0]
                c_4() = [0]
                        [0]
                        [0]
                rev^#(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                c_5() = [0]
                        [0]
                        [0]
                c_6(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev2^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                last^#(x1, x2) = [3 0 0] x1 + [3 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                c_7(x1) = [1 0 0] x1 + [0]
                          [0 1 0]      [0]
                          [0 0 1]      [0]
                if^#(x1, x2, x3) = [3 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                c_8() = [0]
                        [0]
                        [0]
                c_9(x1) = [1 0 0] x1 + [0]
                          [0 1 0]      [0]
                          [0 0 1]      [0]
                c_10() = [0]
                         [0]
                         [0]
                c_11(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
             Complexity induced by the adequate RMI: YES(?,O(n^3))
             
             We have not generated a proof for the resulting sub-problem.
    
    2) 'wdg' failed due to the following reason:
         Transformation Details:
         -----------------------
           We have computed the following set of weak (innermost) dependency pairs:
           
             {  1: empty^#(nil()) -> c_0()
              , 2: empty^#(cons(x, l)) -> c_1()
              , 3: head^#(cons(x, l)) -> c_2()
              , 4: tail^#(nil()) -> c_3()
              , 5: tail^#(cons(x, l)) -> c_4()
              , 6: rev^#(nil()) -> c_5()
              , 7: rev^#(cons(x, l)) -> c_6(rev2^#(x, l))
              , 8: last^#(x, l) -> c_7(if^#(empty(l), x, l))
              , 9: if^#(true(), x, l) -> c_8()
              , 10: if^#(false(), x, l) -> c_9(last^#(head(l), tail(l)))
              , 11: rev2^#(x, nil()) -> c_10()
              , 12: rev2^#(x, cons(y, l)) -> c_11(rev^#(cons(x, rev2(y, l))))}
           
           Following Dependency Graph (modulo SCCs) was computed. (Answers to
           subproofs are indicated to the right.)
           
             ->{8,10}                                                    [       MAYBE        ]
                |
                `->{9}                                                   [         NA         ]
             
             ->{7,12}                                                    [     inherited      ]
                |
                `->{11}                                                  [         NA         ]
             
             ->{6}                                                       [    YES(?,O(1))     ]
             
             ->{5}                                                       [    YES(?,O(1))     ]
             
             ->{4}                                                       [    YES(?,O(1))     ]
             
             ->{3}                                                       [    YES(?,O(1))     ]
             
             ->{2}                                                       [    YES(?,O(1))     ]
             
             ->{1}                                                       [    YES(?,O(1))     ]
             
           
         
         Sub-problems:
         -------------
           * Path {1}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(tail^#) = {},
                 Uargs(rev^#) = {}, Uargs(c_6) = {}, Uargs(rev2^#) = {},
                 Uargs(last^#) = {}, Uargs(c_7) = {}, Uargs(if^#) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                nil() = [0]
                        [0]
                true() = [0]
                         [0]
                cons(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                false() = [0]
                          [0]
                head(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                tail(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                rev(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev1(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                rev2(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                last(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                if(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                 [0 0]      [0 0]      [0 0]      [0]
                empty^#(x1) = [0 0] x1 + [0]
                              [0 0]      [0]
                c_0() = [0]
                        [0]
                c_1() = [0]
                        [0]
                head^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_2() = [0]
                        [0]
                tail^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_3() = [0]
                        [0]
                c_4() = [0]
                        [0]
                rev^#(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                c_5() = [0]
                        [0]
                c_6(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev2^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                last^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                c_7(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                if^#(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                   [0 0]      [0 0]      [0 0]      [0]
                c_8() = [0]
                        [0]
                c_9(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                c_10() = [0]
                         [0]
                c_11(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 2'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    innermost DP runtime-complexity with respect to
               Strict Rules: {empty^#(nil()) -> c_0()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(empty^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                nil() = [2]
                        [2]
                empty^#(x1) = [2 0] x1 + [7]
                              [2 2]      [7]
                c_0() = [0]
                        [1]
           
           * Path {2}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(tail^#) = {},
                 Uargs(rev^#) = {}, Uargs(c_6) = {}, Uargs(rev2^#) = {},
                 Uargs(last^#) = {}, Uargs(c_7) = {}, Uargs(if^#) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                nil() = [0]
                        [0]
                true() = [0]
                         [0]
                cons(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                false() = [0]
                          [0]
                head(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                tail(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                rev(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev1(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                rev2(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                last(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                if(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                 [0 0]      [0 0]      [0 0]      [0]
                empty^#(x1) = [0 0] x1 + [0]
                              [0 0]      [0]
                c_0() = [0]
                        [0]
                c_1() = [0]
                        [0]
                head^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_2() = [0]
                        [0]
                tail^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_3() = [0]
                        [0]
                c_4() = [0]
                        [0]
                rev^#(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                c_5() = [0]
                        [0]
                c_6(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev2^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                last^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                c_7(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                if^#(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                   [0 0]      [0 0]      [0 0]      [0]
                c_8() = [0]
                        [0]
                c_9(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                c_10() = [0]
                         [0]
                c_11(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 2'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    innermost DP runtime-complexity with respect to
               Strict Rules: {empty^#(cons(x, l)) -> c_1()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(cons) = {}, Uargs(empty^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                cons(x1, x2) = [0 0] x1 + [0 0] x2 + [2]
                               [0 0]      [0 0]      [2]
                empty^#(x1) = [2 0] x1 + [7]
                              [2 2]      [7]
                c_1() = [0]
                        [1]
           
           * Path {3}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(tail^#) = {},
                 Uargs(rev^#) = {}, Uargs(c_6) = {}, Uargs(rev2^#) = {},
                 Uargs(last^#) = {}, Uargs(c_7) = {}, Uargs(if^#) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                nil() = [0]
                        [0]
                true() = [0]
                         [0]
                cons(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                false() = [0]
                          [0]
                head(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                tail(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                rev(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev1(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                rev2(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                last(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                if(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                 [0 0]      [0 0]      [0 0]      [0]
                empty^#(x1) = [0 0] x1 + [0]
                              [0 0]      [0]
                c_0() = [0]
                        [0]
                c_1() = [0]
                        [0]
                head^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_2() = [0]
                        [0]
                tail^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_3() = [0]
                        [0]
                c_4() = [0]
                        [0]
                rev^#(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                c_5() = [0]
                        [0]
                c_6(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev2^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                last^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                c_7(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                if^#(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                   [0 0]      [0 0]      [0 0]      [0]
                c_8() = [0]
                        [0]
                c_9(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                c_10() = [0]
                         [0]
                c_11(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 2'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    innermost DP runtime-complexity with respect to
               Strict Rules: {head^#(cons(x, l)) -> c_2()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(cons) = {}, Uargs(head^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                cons(x1, x2) = [0 0] x1 + [0 0] x2 + [2]
                               [0 0]      [0 0]      [2]
                head^#(x1) = [2 0] x1 + [7]
                             [2 2]      [7]
                c_2() = [0]
                        [1]
           
           * Path {4}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(tail^#) = {},
                 Uargs(rev^#) = {}, Uargs(c_6) = {}, Uargs(rev2^#) = {},
                 Uargs(last^#) = {}, Uargs(c_7) = {}, Uargs(if^#) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                nil() = [0]
                        [0]
                true() = [0]
                         [0]
                cons(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                false() = [0]
                          [0]
                head(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                tail(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                rev(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev1(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                rev2(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                last(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                if(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                 [0 0]      [0 0]      [0 0]      [0]
                empty^#(x1) = [0 0] x1 + [0]
                              [0 0]      [0]
                c_0() = [0]
                        [0]
                c_1() = [0]
                        [0]
                head^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_2() = [0]
                        [0]
                tail^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_3() = [0]
                        [0]
                c_4() = [0]
                        [0]
                rev^#(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                c_5() = [0]
                        [0]
                c_6(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev2^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                last^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                c_7(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                if^#(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                   [0 0]      [0 0]      [0 0]      [0]
                c_8() = [0]
                        [0]
                c_9(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                c_10() = [0]
                         [0]
                c_11(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 2'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    innermost DP runtime-complexity with respect to
               Strict Rules: {tail^#(nil()) -> c_3()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(tail^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                nil() = [2]
                        [2]
                tail^#(x1) = [2 0] x1 + [7]
                             [2 2]      [7]
                c_3() = [0]
                        [1]
           
           * Path {5}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(tail^#) = {},
                 Uargs(rev^#) = {}, Uargs(c_6) = {}, Uargs(rev2^#) = {},
                 Uargs(last^#) = {}, Uargs(c_7) = {}, Uargs(if^#) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                nil() = [0]
                        [0]
                true() = [0]
                         [0]
                cons(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                false() = [0]
                          [0]
                head(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                tail(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                rev(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev1(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                rev2(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                last(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                if(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                 [0 0]      [0 0]      [0 0]      [0]
                empty^#(x1) = [0 0] x1 + [0]
                              [0 0]      [0]
                c_0() = [0]
                        [0]
                c_1() = [0]
                        [0]
                head^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_2() = [0]
                        [0]
                tail^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_3() = [0]
                        [0]
                c_4() = [0]
                        [0]
                rev^#(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                c_5() = [0]
                        [0]
                c_6(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev2^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                last^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                c_7(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                if^#(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                   [0 0]      [0 0]      [0 0]      [0]
                c_8() = [0]
                        [0]
                c_9(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                c_10() = [0]
                         [0]
                c_11(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 2'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    innermost DP runtime-complexity with respect to
               Strict Rules: {tail^#(cons(x, l)) -> c_4()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(cons) = {}, Uargs(tail^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                cons(x1, x2) = [0 0] x1 + [0 0] x2 + [2]
                               [0 0]      [0 0]      [2]
                tail^#(x1) = [2 0] x1 + [7]
                             [2 2]      [7]
                c_4() = [0]
                        [1]
           
           * Path {6}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(tail^#) = {},
                 Uargs(rev^#) = {}, Uargs(c_6) = {}, Uargs(rev2^#) = {},
                 Uargs(last^#) = {}, Uargs(c_7) = {}, Uargs(if^#) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                nil() = [0]
                        [0]
                true() = [0]
                         [0]
                cons(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                false() = [0]
                          [0]
                head(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                tail(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                rev(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev1(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                rev2(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                last(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                if(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                 [0 0]      [0 0]      [0 0]      [0]
                empty^#(x1) = [0 0] x1 + [0]
                              [0 0]      [0]
                c_0() = [0]
                        [0]
                c_1() = [0]
                        [0]
                head^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_2() = [0]
                        [0]
                tail^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_3() = [0]
                        [0]
                c_4() = [0]
                        [0]
                rev^#(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                c_5() = [0]
                        [0]
                c_6(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev2^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                last^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                c_7(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                if^#(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                   [0 0]      [0 0]      [0 0]      [0]
                c_8() = [0]
                        [0]
                c_9(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                c_10() = [0]
                         [0]
                c_11(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 2'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    innermost DP runtime-complexity with respect to
               Strict Rules: {rev^#(nil()) -> c_5()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(rev^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                nil() = [2]
                        [2]
                rev^#(x1) = [2 0] x1 + [7]
                            [2 2]      [7]
                c_5() = [0]
                        [1]
           
           * Path {7,12}: inherited
             ----------------------
             
             This path is subsumed by the proof of path {7,12}->{11}.
           
           * Path {7,12}->{11}: NA
             ---------------------
             
             The usable rules for this path are:
             
               {  rev2(x, nil()) -> nil()
                , rev2(x, cons(y, l)) -> rev(cons(x, rev2(y, l)))
                , rev(nil()) -> nil()
                , rev(cons(x, l)) -> cons(rev1(x, l), rev2(x, l))}
             
             The weight gap principle does not apply:
               The input cannot be shown compatible
             Complexity induced by the adequate RMI: MAYBE
             
             We have not generated a proof for the resulting sub-problem.
           
           * Path {8,10}: MAYBE
             ------------------
             
             The usable rules for this path are:
             
               {  empty(nil()) -> true()
                , empty(cons(x, l)) -> false()
                , head(cons(x, l)) -> x
                , tail(nil()) -> nil()
                , tail(cons(x, l)) -> l}
             
             The weight gap principle does not apply:
               The input cannot be shown compatible
             Complexity induced by the adequate RMI: MAYBE
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 2'
             --------------------------------------
             Answer:           MAYBE
             Input Problem:    innermost runtime-complexity with respect to
               Rules:
                 {  last^#(x, l) -> c_7(if^#(empty(l), x, l))
                  , if^#(false(), x, l) -> c_9(last^#(head(l), tail(l)))
                  , empty(nil()) -> true()
                  , empty(cons(x, l)) -> false()
                  , head(cons(x, l)) -> x
                  , tail(nil()) -> nil()
                  , tail(cons(x, l)) -> l}
             
             Proof Output:    
               The input cannot be shown compatible
           
           * Path {8,10}->{9}: NA
             --------------------
             
             The usable rules for this path are:
             
               {  empty(nil()) -> true()
                , empty(cons(x, l)) -> false()
                , head(cons(x, l)) -> x
                , tail(nil()) -> nil()
                , tail(cons(x, l)) -> l}
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(tail^#) = {},
                 Uargs(rev^#) = {}, Uargs(c_6) = {}, Uargs(rev2^#) = {},
                 Uargs(last^#) = {1, 2}, Uargs(c_7) = {1}, Uargs(if^#) = {1},
                 Uargs(c_9) = {1}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0] x1 + [2]
                            [3 3]      [3]
                nil() = [2]
                        [0]
                true() = [1]
                         [1]
                cons(x1, x2) = [1 1] x1 + [1 3] x2 + [1]
                               [0 1]      [0 1]      [0]
                false() = [1]
                          [0]
                head(x1) = [3 3] x1 + [3]
                           [3 3]      [3]
                tail(x1) = [2 0] x1 + [0]
                           [0 2]      [0]
                rev(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev1(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                rev2(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                last(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                if(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                 [0 0]      [0 0]      [0 0]      [0]
                empty^#(x1) = [0 0] x1 + [0]
                              [0 0]      [0]
                c_0() = [0]
                        [0]
                c_1() = [0]
                        [0]
                head^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_2() = [0]
                        [0]
                tail^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_3() = [0]
                        [0]
                c_4() = [0]
                        [0]
                rev^#(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                c_5() = [0]
                        [0]
                c_6(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev2^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                last^#(x1, x2) = [3 0] x1 + [3 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                c_7(x1) = [1 0] x1 + [0]
                          [0 1]      [0]
                if^#(x1, x2, x3) = [3 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                   [0 0]      [0 0]      [0 0]      [0]
                c_8() = [0]
                        [0]
                c_9(x1) = [1 0] x1 + [0]
                          [0 1]      [0]
                c_10() = [0]
                         [0]
                c_11(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
             Complexity induced by the adequate RMI: YES(?,O(n^2))
             
             We have not generated a proof for the resulting sub-problem.
    
    3) 'wdg' failed due to the following reason:
         Transformation Details:
         -----------------------
           We have computed the following set of weak (innermost) dependency pairs:
           
             {  1: empty^#(nil()) -> c_0()
              , 2: empty^#(cons(x, l)) -> c_1()
              , 3: head^#(cons(x, l)) -> c_2()
              , 4: tail^#(nil()) -> c_3()
              , 5: tail^#(cons(x, l)) -> c_4()
              , 6: rev^#(nil()) -> c_5()
              , 7: rev^#(cons(x, l)) -> c_6(rev2^#(x, l))
              , 8: last^#(x, l) -> c_7(if^#(empty(l), x, l))
              , 9: if^#(true(), x, l) -> c_8()
              , 10: if^#(false(), x, l) -> c_9(last^#(head(l), tail(l)))
              , 11: rev2^#(x, nil()) -> c_10()
              , 12: rev2^#(x, cons(y, l)) -> c_11(rev^#(cons(x, rev2(y, l))))}
           
           Following Dependency Graph (modulo SCCs) was computed. (Answers to
           subproofs are indicated to the right.)
           
             ->{8,10}                                                    [       MAYBE        ]
                |
                `->{9}                                                   [         NA         ]
             
             ->{7,12}                                                    [     inherited      ]
                |
                `->{11}                                                  [         NA         ]
             
             ->{6}                                                       [    YES(?,O(1))     ]
             
             ->{5}                                                       [    YES(?,O(1))     ]
             
             ->{4}                                                       [    YES(?,O(1))     ]
             
             ->{3}                                                       [    YES(?,O(1))     ]
             
             ->{2}                                                       [    YES(?,O(1))     ]
             
             ->{1}                                                       [    YES(?,O(1))     ]
             
           
         
         Sub-problems:
         -------------
           * Path {1}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(tail^#) = {},
                 Uargs(rev^#) = {}, Uargs(c_6) = {}, Uargs(rev2^#) = {},
                 Uargs(last^#) = {}, Uargs(c_7) = {}, Uargs(if^#) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0] x1 + [0]
                nil() = [0]
                true() = [0]
                cons(x1, x2) = [0] x1 + [0] x2 + [0]
                false() = [0]
                head(x1) = [0] x1 + [0]
                tail(x1) = [0] x1 + [0]
                rev(x1) = [0] x1 + [0]
                rev1(x1, x2) = [0] x1 + [0] x2 + [0]
                rev2(x1, x2) = [0] x1 + [0] x2 + [0]
                last(x1, x2) = [0] x1 + [0] x2 + [0]
                if(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                empty^#(x1) = [0] x1 + [0]
                c_0() = [0]
                c_1() = [0]
                head^#(x1) = [0] x1 + [0]
                c_2() = [0]
                tail^#(x1) = [0] x1 + [0]
                c_3() = [0]
                c_4() = [0]
                rev^#(x1) = [0] x1 + [0]
                c_5() = [0]
                c_6(x1) = [0] x1 + [0]
                rev2^#(x1, x2) = [0] x1 + [0] x2 + [0]
                last^#(x1, x2) = [0] x1 + [0] x2 + [0]
                c_7(x1) = [0] x1 + [0]
                if^#(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                c_8() = [0]
                c_9(x1) = [0] x1 + [0]
                c_10() = [0]
                c_11(x1) = [0] x1 + [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 1'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    innermost DP runtime-complexity with respect to
               Strict Rules: {empty^#(nil()) -> c_0()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(empty^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                nil() = [7]
                empty^#(x1) = [1] x1 + [7]
                c_0() = [1]
           
           * Path {2}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(tail^#) = {},
                 Uargs(rev^#) = {}, Uargs(c_6) = {}, Uargs(rev2^#) = {},
                 Uargs(last^#) = {}, Uargs(c_7) = {}, Uargs(if^#) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0] x1 + [0]
                nil() = [0]
                true() = [0]
                cons(x1, x2) = [0] x1 + [0] x2 + [0]
                false() = [0]
                head(x1) = [0] x1 + [0]
                tail(x1) = [0] x1 + [0]
                rev(x1) = [0] x1 + [0]
                rev1(x1, x2) = [0] x1 + [0] x2 + [0]
                rev2(x1, x2) = [0] x1 + [0] x2 + [0]
                last(x1, x2) = [0] x1 + [0] x2 + [0]
                if(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                empty^#(x1) = [0] x1 + [0]
                c_0() = [0]
                c_1() = [0]
                head^#(x1) = [0] x1 + [0]
                c_2() = [0]
                tail^#(x1) = [0] x1 + [0]
                c_3() = [0]
                c_4() = [0]
                rev^#(x1) = [0] x1 + [0]
                c_5() = [0]
                c_6(x1) = [0] x1 + [0]
                rev2^#(x1, x2) = [0] x1 + [0] x2 + [0]
                last^#(x1, x2) = [0] x1 + [0] x2 + [0]
                c_7(x1) = [0] x1 + [0]
                if^#(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                c_8() = [0]
                c_9(x1) = [0] x1 + [0]
                c_10() = [0]
                c_11(x1) = [0] x1 + [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 1'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    innermost DP runtime-complexity with respect to
               Strict Rules: {empty^#(cons(x, l)) -> c_1()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(cons) = {}, Uargs(empty^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                cons(x1, x2) = [0] x1 + [0] x2 + [7]
                empty^#(x1) = [1] x1 + [7]
                c_1() = [1]
           
           * Path {3}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(tail^#) = {},
                 Uargs(rev^#) = {}, Uargs(c_6) = {}, Uargs(rev2^#) = {},
                 Uargs(last^#) = {}, Uargs(c_7) = {}, Uargs(if^#) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0] x1 + [0]
                nil() = [0]
                true() = [0]
                cons(x1, x2) = [0] x1 + [0] x2 + [0]
                false() = [0]
                head(x1) = [0] x1 + [0]
                tail(x1) = [0] x1 + [0]
                rev(x1) = [0] x1 + [0]
                rev1(x1, x2) = [0] x1 + [0] x2 + [0]
                rev2(x1, x2) = [0] x1 + [0] x2 + [0]
                last(x1, x2) = [0] x1 + [0] x2 + [0]
                if(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                empty^#(x1) = [0] x1 + [0]
                c_0() = [0]
                c_1() = [0]
                head^#(x1) = [0] x1 + [0]
                c_2() = [0]
                tail^#(x1) = [0] x1 + [0]
                c_3() = [0]
                c_4() = [0]
                rev^#(x1) = [0] x1 + [0]
                c_5() = [0]
                c_6(x1) = [0] x1 + [0]
                rev2^#(x1, x2) = [0] x1 + [0] x2 + [0]
                last^#(x1, x2) = [0] x1 + [0] x2 + [0]
                c_7(x1) = [0] x1 + [0]
                if^#(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                c_8() = [0]
                c_9(x1) = [0] x1 + [0]
                c_10() = [0]
                c_11(x1) = [0] x1 + [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 1'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    innermost DP runtime-complexity with respect to
               Strict Rules: {head^#(cons(x, l)) -> c_2()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(cons) = {}, Uargs(head^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                cons(x1, x2) = [0] x1 + [0] x2 + [7]
                head^#(x1) = [1] x1 + [7]
                c_2() = [1]
           
           * Path {4}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(tail^#) = {},
                 Uargs(rev^#) = {}, Uargs(c_6) = {}, Uargs(rev2^#) = {},
                 Uargs(last^#) = {}, Uargs(c_7) = {}, Uargs(if^#) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0] x1 + [0]
                nil() = [0]
                true() = [0]
                cons(x1, x2) = [0] x1 + [0] x2 + [0]
                false() = [0]
                head(x1) = [0] x1 + [0]
                tail(x1) = [0] x1 + [0]
                rev(x1) = [0] x1 + [0]
                rev1(x1, x2) = [0] x1 + [0] x2 + [0]
                rev2(x1, x2) = [0] x1 + [0] x2 + [0]
                last(x1, x2) = [0] x1 + [0] x2 + [0]
                if(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                empty^#(x1) = [0] x1 + [0]
                c_0() = [0]
                c_1() = [0]
                head^#(x1) = [0] x1 + [0]
                c_2() = [0]
                tail^#(x1) = [0] x1 + [0]
                c_3() = [0]
                c_4() = [0]
                rev^#(x1) = [0] x1 + [0]
                c_5() = [0]
                c_6(x1) = [0] x1 + [0]
                rev2^#(x1, x2) = [0] x1 + [0] x2 + [0]
                last^#(x1, x2) = [0] x1 + [0] x2 + [0]
                c_7(x1) = [0] x1 + [0]
                if^#(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                c_8() = [0]
                c_9(x1) = [0] x1 + [0]
                c_10() = [0]
                c_11(x1) = [0] x1 + [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 1'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    innermost DP runtime-complexity with respect to
               Strict Rules: {tail^#(nil()) -> c_3()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(tail^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                nil() = [7]
                tail^#(x1) = [1] x1 + [7]
                c_3() = [1]
           
           * Path {5}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(tail^#) = {},
                 Uargs(rev^#) = {}, Uargs(c_6) = {}, Uargs(rev2^#) = {},
                 Uargs(last^#) = {}, Uargs(c_7) = {}, Uargs(if^#) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0] x1 + [0]
                nil() = [0]
                true() = [0]
                cons(x1, x2) = [0] x1 + [0] x2 + [0]
                false() = [0]
                head(x1) = [0] x1 + [0]
                tail(x1) = [0] x1 + [0]
                rev(x1) = [0] x1 + [0]
                rev1(x1, x2) = [0] x1 + [0] x2 + [0]
                rev2(x1, x2) = [0] x1 + [0] x2 + [0]
                last(x1, x2) = [0] x1 + [0] x2 + [0]
                if(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                empty^#(x1) = [0] x1 + [0]
                c_0() = [0]
                c_1() = [0]
                head^#(x1) = [0] x1 + [0]
                c_2() = [0]
                tail^#(x1) = [0] x1 + [0]
                c_3() = [0]
                c_4() = [0]
                rev^#(x1) = [0] x1 + [0]
                c_5() = [0]
                c_6(x1) = [0] x1 + [0]
                rev2^#(x1, x2) = [0] x1 + [0] x2 + [0]
                last^#(x1, x2) = [0] x1 + [0] x2 + [0]
                c_7(x1) = [0] x1 + [0]
                if^#(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                c_8() = [0]
                c_9(x1) = [0] x1 + [0]
                c_10() = [0]
                c_11(x1) = [0] x1 + [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 1'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    innermost DP runtime-complexity with respect to
               Strict Rules: {tail^#(cons(x, l)) -> c_4()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(cons) = {}, Uargs(tail^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                cons(x1, x2) = [0] x1 + [0] x2 + [7]
                tail^#(x1) = [1] x1 + [7]
                c_4() = [1]
           
           * Path {6}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(tail^#) = {},
                 Uargs(rev^#) = {}, Uargs(c_6) = {}, Uargs(rev2^#) = {},
                 Uargs(last^#) = {}, Uargs(c_7) = {}, Uargs(if^#) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0] x1 + [0]
                nil() = [0]
                true() = [0]
                cons(x1, x2) = [0] x1 + [0] x2 + [0]
                false() = [0]
                head(x1) = [0] x1 + [0]
                tail(x1) = [0] x1 + [0]
                rev(x1) = [0] x1 + [0]
                rev1(x1, x2) = [0] x1 + [0] x2 + [0]
                rev2(x1, x2) = [0] x1 + [0] x2 + [0]
                last(x1, x2) = [0] x1 + [0] x2 + [0]
                if(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                empty^#(x1) = [0] x1 + [0]
                c_0() = [0]
                c_1() = [0]
                head^#(x1) = [0] x1 + [0]
                c_2() = [0]
                tail^#(x1) = [0] x1 + [0]
                c_3() = [0]
                c_4() = [0]
                rev^#(x1) = [0] x1 + [0]
                c_5() = [0]
                c_6(x1) = [0] x1 + [0]
                rev2^#(x1, x2) = [0] x1 + [0] x2 + [0]
                last^#(x1, x2) = [0] x1 + [0] x2 + [0]
                c_7(x1) = [0] x1 + [0]
                if^#(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                c_8() = [0]
                c_9(x1) = [0] x1 + [0]
                c_10() = [0]
                c_11(x1) = [0] x1 + [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 1'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    innermost DP runtime-complexity with respect to
               Strict Rules: {rev^#(nil()) -> c_5()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(rev^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                nil() = [7]
                rev^#(x1) = [1] x1 + [7]
                c_5() = [1]
           
           * Path {7,12}: inherited
             ----------------------
             
             This path is subsumed by the proof of path {7,12}->{11}.
           
           * Path {7,12}->{11}: NA
             ---------------------
             
             The usable rules for this path are:
             
               {  rev2(x, nil()) -> nil()
                , rev2(x, cons(y, l)) -> rev(cons(x, rev2(y, l)))
                , rev(nil()) -> nil()
                , rev(cons(x, l)) -> cons(rev1(x, l), rev2(x, l))}
             
             The weight gap principle does not apply:
               The input cannot be shown compatible
             Complexity induced by the adequate RMI: MAYBE
             
             We have not generated a proof for the resulting sub-problem.
           
           * Path {8,10}: MAYBE
             ------------------
             
             The usable rules for this path are:
             
               {  empty(nil()) -> true()
                , empty(cons(x, l)) -> false()
                , head(cons(x, l)) -> x
                , tail(nil()) -> nil()
                , tail(cons(x, l)) -> l}
             
             The weight gap principle does not apply:
               The input cannot be shown compatible
             Complexity induced by the adequate RMI: MAYBE
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 1'
             --------------------------------------
             Answer:           MAYBE
             Input Problem:    innermost runtime-complexity with respect to
               Rules:
                 {  last^#(x, l) -> c_7(if^#(empty(l), x, l))
                  , if^#(false(), x, l) -> c_9(last^#(head(l), tail(l)))
                  , empty(nil()) -> true()
                  , empty(cons(x, l)) -> false()
                  , head(cons(x, l)) -> x
                  , tail(nil()) -> nil()
                  , tail(cons(x, l)) -> l}
             
             Proof Output:    
               The input cannot be shown compatible
           
           * Path {8,10}->{9}: NA
             --------------------
             
             The usable rules for this path are:
             
               {  empty(nil()) -> true()
                , empty(cons(x, l)) -> false()
                , head(cons(x, l)) -> x
                , tail(nil()) -> nil()
                , tail(cons(x, l)) -> l}
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(tail^#) = {},
                 Uargs(rev^#) = {}, Uargs(c_6) = {}, Uargs(rev2^#) = {},
                 Uargs(last^#) = {1, 2}, Uargs(c_7) = {1}, Uargs(if^#) = {1},
                 Uargs(c_9) = {1}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [3] x1 + [3]
                nil() = [1]
                true() = [1]
                cons(x1, x2) = [1] x1 + [1] x2 + [1]
                false() = [1]
                head(x1) = [3] x1 + [3]
                tail(x1) = [3] x1 + [3]
                rev(x1) = [0] x1 + [0]
                rev1(x1, x2) = [0] x1 + [0] x2 + [0]
                rev2(x1, x2) = [0] x1 + [0] x2 + [0]
                last(x1, x2) = [0] x1 + [0] x2 + [0]
                if(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                empty^#(x1) = [0] x1 + [0]
                c_0() = [0]
                c_1() = [0]
                head^#(x1) = [0] x1 + [0]
                c_2() = [0]
                tail^#(x1) = [0] x1 + [0]
                c_3() = [0]
                c_4() = [0]
                rev^#(x1) = [0] x1 + [0]
                c_5() = [0]
                c_6(x1) = [0] x1 + [0]
                rev2^#(x1, x2) = [0] x1 + [0] x2 + [0]
                last^#(x1, x2) = [3] x1 + [3] x2 + [0]
                c_7(x1) = [1] x1 + [0]
                if^#(x1, x2, x3) = [3] x1 + [0] x2 + [0] x3 + [0]
                c_8() = [0]
                c_9(x1) = [1] x1 + [0]
                c_10() = [0]
                c_11(x1) = [0] x1 + [0]
             Complexity induced by the adequate RMI: YES(?,O(n^1))
             
             We have not generated a proof for the resulting sub-problem.
    
    4) 'matrix-interpretation of dimension 1' failed due to the following reason:
         The input cannot be shown compatible
    
    5) 'Bounds with perSymbol-enrichment and initial automaton 'match'' failed due to the following reason:
         match-boundness of the problem could not be verified.
    
    6) 'Bounds with minimal-enrichment and initial automaton 'match'' failed due to the following reason:
         match-boundness of the problem could not be verified.
    

Tool RC1

Execution TimeUnknown
Answer
MAYBE
InputAProVE 07 thiemann08

stdout:

MAYBE

Tool RC2

Execution TimeUnknown
Answer
MAYBE
InputAProVE 07 thiemann08

stdout:

MAYBE

'Fastest (timeout of 60.0 seconds)'
-----------------------------------
Answer:           MAYBE
Input Problem:    runtime-complexity with respect to
  Rules:
    {  empty(nil()) -> true()
     , empty(cons(x, l)) -> false()
     , head(cons(x, l)) -> x
     , tail(nil()) -> nil()
     , tail(cons(x, l)) -> l
     , rev(nil()) -> nil()
     , rev(cons(x, l)) -> cons(rev1(x, l), rev2(x, l))
     , last(x, l) -> if(empty(l), x, l)
     , if(true(), x, l) -> x
     , if(false(), x, l) -> last(head(l), tail(l))
     , rev2(x, nil()) -> nil()
     , rev2(x, cons(y, l)) -> rev(cons(x, rev2(y, l)))}

Proof Output:    
  None of the processors succeeded.
  
  Details of failed attempt(s):
  -----------------------------
    1) 'wdg' failed due to the following reason:
         Transformation Details:
         -----------------------
           We have computed the following set of weak (innermost) dependency pairs:
           
             {  1: empty^#(nil()) -> c_0()
              , 2: empty^#(cons(x, l)) -> c_1()
              , 3: head^#(cons(x, l)) -> c_2(x)
              , 4: tail^#(nil()) -> c_3()
              , 5: tail^#(cons(x, l)) -> c_4(l)
              , 6: rev^#(nil()) -> c_5()
              , 7: rev^#(cons(x, l)) -> c_6(x, l, rev2^#(x, l))
              , 8: last^#(x, l) -> c_7(if^#(empty(l), x, l))
              , 9: if^#(true(), x, l) -> c_8(x)
              , 10: if^#(false(), x, l) -> c_9(last^#(head(l), tail(l)))
              , 11: rev2^#(x, nil()) -> c_10()
              , 12: rev2^#(x, cons(y, l)) -> c_11(rev^#(cons(x, rev2(y, l))))}
           
           Following Dependency Graph (modulo SCCs) was computed. (Answers to
           subproofs are indicated to the right.)
           
             ->{8,10}                                                    [       MAYBE        ]
                |
                `->{9}                                                   [         NA         ]
             
             ->{7,12}                                                    [     inherited      ]
                |
                `->{11}                                                  [         NA         ]
             
             ->{6}                                                       [    YES(?,O(1))     ]
             
             ->{5}                                                       [   YES(?,O(n^3))    ]
             
             ->{4}                                                       [    YES(?,O(1))     ]
             
             ->{3}                                                       [   YES(?,O(n^3))    ]
             
             ->{2}                                                       [    YES(?,O(1))     ]
             
             ->{1}                                                       [    YES(?,O(1))     ]
             
           
         
         Sub-problems:
         -------------
           * Path {1}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(c_2) = {},
                 Uargs(tail^#) = {}, Uargs(c_4) = {}, Uargs(rev^#) = {},
                 Uargs(c_6) = {}, Uargs(rev2^#) = {}, Uargs(last^#) = {},
                 Uargs(c_7) = {}, Uargs(if^#) = {}, Uargs(c_8) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                nil() = [0]
                        [0]
                        [0]
                true() = [0]
                         [0]
                         [0]
                cons(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                false() = [0]
                          [0]
                          [0]
                head(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                tail(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                rev(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev1(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                rev2(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                last(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                if(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                empty^#(x1) = [0 0 0] x1 + [0]
                              [0 0 0]      [0]
                              [0 0 0]      [0]
                c_0() = [0]
                        [0]
                        [0]
                c_1() = [0]
                        [0]
                        [0]
                head^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_2(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                tail^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_3() = [0]
                        [0]
                        [0]
                c_4(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev^#(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                c_5() = [0]
                        [0]
                        [0]
                c_6(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                  [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                  [0 0 0]      [0 0 0]      [0 0 0]      [0]
                rev2^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                last^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                c_7(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                if^#(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                c_8(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                c_9(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                c_10() = [0]
                         [0]
                         [0]
                c_11(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 3'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    DP runtime-complexity with respect to
               Strict Rules: {empty^#(nil()) -> c_0()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(empty^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                nil() = [2]
                        [2]
                        [2]
                empty^#(x1) = [0 2 0] x1 + [7]
                              [2 2 0]      [3]
                              [2 2 2]      [3]
                c_0() = [0]
                        [1]
                        [1]
           
           * Path {2}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(c_2) = {},
                 Uargs(tail^#) = {}, Uargs(c_4) = {}, Uargs(rev^#) = {},
                 Uargs(c_6) = {}, Uargs(rev2^#) = {}, Uargs(last^#) = {},
                 Uargs(c_7) = {}, Uargs(if^#) = {}, Uargs(c_8) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                nil() = [0]
                        [0]
                        [0]
                true() = [0]
                         [0]
                         [0]
                cons(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                false() = [0]
                          [0]
                          [0]
                head(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                tail(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                rev(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev1(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                rev2(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                last(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                if(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                empty^#(x1) = [0 0 0] x1 + [0]
                              [0 0 0]      [0]
                              [0 0 0]      [0]
                c_0() = [0]
                        [0]
                        [0]
                c_1() = [0]
                        [0]
                        [0]
                head^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_2(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                tail^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_3() = [0]
                        [0]
                        [0]
                c_4(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev^#(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                c_5() = [0]
                        [0]
                        [0]
                c_6(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                  [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                  [0 0 0]      [0 0 0]      [0 0 0]      [0]
                rev2^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                last^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                c_7(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                if^#(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                c_8(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                c_9(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                c_10() = [0]
                         [0]
                         [0]
                c_11(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 3'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    DP runtime-complexity with respect to
               Strict Rules: {empty^#(cons(x, l)) -> c_1()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(cons) = {}, Uargs(empty^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                cons(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [2]
                               [0 0 0]      [0 0 0]      [2]
                               [0 0 0]      [0 0 0]      [2]
                empty^#(x1) = [0 2 0] x1 + [7]
                              [2 2 0]      [3]
                              [2 2 2]      [3]
                c_1() = [0]
                        [1]
                        [1]
           
           * Path {3}: YES(?,O(n^3))
             -----------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(c_2) = {},
                 Uargs(tail^#) = {}, Uargs(c_4) = {}, Uargs(rev^#) = {},
                 Uargs(c_6) = {}, Uargs(rev2^#) = {}, Uargs(last^#) = {},
                 Uargs(c_7) = {}, Uargs(if^#) = {}, Uargs(c_8) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                nil() = [0]
                        [0]
                        [0]
                true() = [0]
                         [0]
                         [0]
                cons(x1, x2) = [1 3 3] x1 + [0 0 0] x2 + [0]
                               [0 1 1]      [0 0 0]      [0]
                               [0 0 1]      [0 0 0]      [0]
                false() = [0]
                          [0]
                          [0]
                head(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                tail(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                rev(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev1(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                rev2(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                last(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                if(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                empty^#(x1) = [0 0 0] x1 + [0]
                              [0 0 0]      [0]
                              [0 0 0]      [0]
                c_0() = [0]
                        [0]
                        [0]
                c_1() = [0]
                        [0]
                        [0]
                head^#(x1) = [1 3 3] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_2(x1) = [1 0 1] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                tail^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_3() = [0]
                        [0]
                        [0]
                c_4(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev^#(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                c_5() = [0]
                        [0]
                        [0]
                c_6(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                  [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                  [0 0 0]      [0 0 0]      [0 0 0]      [0]
                rev2^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                last^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                c_7(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                if^#(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                c_8(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                c_9(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                c_10() = [0]
                         [0]
                         [0]
                c_11(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 3'
             --------------------------------------
             Answer:           YES(?,O(n^1))
             Input Problem:    DP runtime-complexity with respect to
               Strict Rules: {head^#(cons(x, l)) -> c_2(x)}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(cons) = {}, Uargs(head^#) = {}, Uargs(c_2) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                cons(x1, x2) = [1 2 2] x1 + [0 0 0] x2 + [2]
                               [0 0 2]      [0 0 0]      [2]
                               [0 0 0]      [0 0 0]      [2]
                head^#(x1) = [2 2 2] x1 + [3]
                             [2 2 2]      [3]
                             [2 2 2]      [3]
                c_2(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [1]
                          [0 0 0]      [1]
           
           * Path {4}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(c_2) = {},
                 Uargs(tail^#) = {}, Uargs(c_4) = {}, Uargs(rev^#) = {},
                 Uargs(c_6) = {}, Uargs(rev2^#) = {}, Uargs(last^#) = {},
                 Uargs(c_7) = {}, Uargs(if^#) = {}, Uargs(c_8) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                nil() = [0]
                        [0]
                        [0]
                true() = [0]
                         [0]
                         [0]
                cons(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                false() = [0]
                          [0]
                          [0]
                head(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                tail(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                rev(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev1(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                rev2(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                last(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                if(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                empty^#(x1) = [0 0 0] x1 + [0]
                              [0 0 0]      [0]
                              [0 0 0]      [0]
                c_0() = [0]
                        [0]
                        [0]
                c_1() = [0]
                        [0]
                        [0]
                head^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_2(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                tail^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_3() = [0]
                        [0]
                        [0]
                c_4(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev^#(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                c_5() = [0]
                        [0]
                        [0]
                c_6(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                  [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                  [0 0 0]      [0 0 0]      [0 0 0]      [0]
                rev2^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                last^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                c_7(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                if^#(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                c_8(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                c_9(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                c_10() = [0]
                         [0]
                         [0]
                c_11(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 3'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    DP runtime-complexity with respect to
               Strict Rules: {tail^#(nil()) -> c_3()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(tail^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                nil() = [2]
                        [2]
                        [2]
                tail^#(x1) = [0 2 0] x1 + [7]
                             [2 2 0]      [3]
                             [2 2 2]      [3]
                c_3() = [0]
                        [1]
                        [1]
           
           * Path {5}: YES(?,O(n^3))
             -----------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(c_2) = {},
                 Uargs(tail^#) = {}, Uargs(c_4) = {}, Uargs(rev^#) = {},
                 Uargs(c_6) = {}, Uargs(rev2^#) = {}, Uargs(last^#) = {},
                 Uargs(c_7) = {}, Uargs(if^#) = {}, Uargs(c_8) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                nil() = [0]
                        [0]
                        [0]
                true() = [0]
                         [0]
                         [0]
                cons(x1, x2) = [1 3 3] x1 + [0 0 0] x2 + [0]
                               [0 1 1]      [0 0 0]      [0]
                               [0 0 1]      [0 0 0]      [0]
                false() = [0]
                          [0]
                          [0]
                head(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                tail(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                rev(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev1(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                rev2(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                last(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                if(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                empty^#(x1) = [0 0 0] x1 + [0]
                              [0 0 0]      [0]
                              [0 0 0]      [0]
                c_0() = [0]
                        [0]
                        [0]
                c_1() = [0]
                        [0]
                        [0]
                head^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_2(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                tail^#(x1) = [1 3 3] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_3() = [0]
                        [0]
                        [0]
                c_4(x1) = [1 0 1] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev^#(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                c_5() = [0]
                        [0]
                        [0]
                c_6(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                  [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                  [0 0 0]      [0 0 0]      [0 0 0]      [0]
                rev2^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                last^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                c_7(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                if^#(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                c_8(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                c_9(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                c_10() = [0]
                         [0]
                         [0]
                c_11(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 3'
             --------------------------------------
             Answer:           YES(?,O(n^1))
             Input Problem:    DP runtime-complexity with respect to
               Strict Rules: {tail^#(cons(x, l)) -> c_4(l)}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(cons) = {}, Uargs(tail^#) = {}, Uargs(c_4) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                cons(x1, x2) = [0 0 0] x1 + [1 2 2] x2 + [2]
                               [0 0 0]      [0 0 2]      [2]
                               [0 0 0]      [0 0 0]      [2]
                tail^#(x1) = [2 2 2] x1 + [3]
                             [2 2 2]      [3]
                             [2 2 2]      [3]
                c_4(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [1]
                          [0 0 0]      [1]
           
           * Path {6}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(c_2) = {},
                 Uargs(tail^#) = {}, Uargs(c_4) = {}, Uargs(rev^#) = {},
                 Uargs(c_6) = {}, Uargs(rev2^#) = {}, Uargs(last^#) = {},
                 Uargs(c_7) = {}, Uargs(if^#) = {}, Uargs(c_8) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                nil() = [0]
                        [0]
                        [0]
                true() = [0]
                         [0]
                         [0]
                cons(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                false() = [0]
                          [0]
                          [0]
                head(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                tail(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
                rev(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev1(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                rev2(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                last(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                if(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                empty^#(x1) = [0 0 0] x1 + [0]
                              [0 0 0]      [0]
                              [0 0 0]      [0]
                c_0() = [0]
                        [0]
                        [0]
                c_1() = [0]
                        [0]
                        [0]
                head^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_2(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                tail^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_3() = [0]
                        [0]
                        [0]
                c_4(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev^#(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                c_5() = [0]
                        [0]
                        [0]
                c_6(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                  [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                  [0 0 0]      [0 0 0]      [0 0 0]      [0]
                rev2^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                last^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                c_7(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                if^#(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                c_8(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                c_9(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                c_10() = [0]
                         [0]
                         [0]
                c_11(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 3'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    DP runtime-complexity with respect to
               Strict Rules: {rev^#(nil()) -> c_5()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(rev^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                nil() = [2]
                        [2]
                        [2]
                rev^#(x1) = [0 2 0] x1 + [7]
                            [2 2 0]      [3]
                            [2 2 2]      [3]
                c_5() = [0]
                        [1]
                        [1]
           
           * Path {7,12}: inherited
             ----------------------
             
             This path is subsumed by the proof of path {7,12}->{11}.
           
           * Path {7,12}->{11}: NA
             ---------------------
             
             The usable rules for this path are:
             
               {  rev2(x, nil()) -> nil()
                , rev2(x, cons(y, l)) -> rev(cons(x, rev2(y, l)))
                , rev(nil()) -> nil()
                , rev(cons(x, l)) -> cons(rev1(x, l), rev2(x, l))}
             
             The weight gap principle does not apply:
               The input cannot be shown compatible
             Complexity induced by the adequate RMI: MAYBE
             
             We have not generated a proof for the resulting sub-problem.
           
           * Path {8,10}: MAYBE
             ------------------
             
             The usable rules for this path are:
             
               {  empty(nil()) -> true()
                , empty(cons(x, l)) -> false()
                , head(cons(x, l)) -> x
                , tail(nil()) -> nil()
                , tail(cons(x, l)) -> l}
             
             The weight gap principle does not apply:
               The input cannot be shown compatible
             Complexity induced by the adequate RMI: MAYBE
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 3'
             --------------------------------------
             Answer:           MAYBE
             Input Problem:    runtime-complexity with respect to
               Rules:
                 {  last^#(x, l) -> c_7(if^#(empty(l), x, l))
                  , if^#(false(), x, l) -> c_9(last^#(head(l), tail(l)))
                  , empty(nil()) -> true()
                  , empty(cons(x, l)) -> false()
                  , head(cons(x, l)) -> x
                  , tail(nil()) -> nil()
                  , tail(cons(x, l)) -> l}
             
             Proof Output:    
               The input cannot be shown compatible
           
           * Path {8,10}->{9}: NA
             --------------------
             
             The usable rules for this path are:
             
               {  empty(nil()) -> true()
                , empty(cons(x, l)) -> false()
                , head(cons(x, l)) -> x
                , tail(nil()) -> nil()
                , tail(cons(x, l)) -> l}
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {1}, Uargs(cons) = {}, Uargs(head) = {1},
                 Uargs(tail) = {1}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(c_2) = {},
                 Uargs(tail^#) = {}, Uargs(c_4) = {}, Uargs(rev^#) = {},
                 Uargs(c_6) = {}, Uargs(rev2^#) = {}, Uargs(last^#) = {1, 2},
                 Uargs(c_7) = {1}, Uargs(if^#) = {1, 2, 3}, Uargs(c_8) = {1},
                 Uargs(c_9) = {1}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [2 0 0] x1 + [2]
                            [0 0 0]      [3]
                            [0 0 0]      [3]
                nil() = [0]
                        [2]
                        [0]
                true() = [1]
                         [1]
                         [1]
                cons(x1, x2) = [1 1 0] x1 + [1 3 0] x2 + [0]
                               [0 1 0]      [0 1 0]      [0]
                               [0 0 1]      [0 0 1]      [0]
                false() = [1]
                          [1]
                          [1]
                head(x1) = [2 0 0] x1 + [3]
                           [3 3 3]      [3]
                           [0 0 1]      [3]
                tail(x1) = [2 0 0] x1 + [2]
                           [0 2 0]      [0]
                           [0 0 1]      [0]
                rev(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev1(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                rev2(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                last(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                               [0 0 0]      [0 0 0]      [0]
                               [0 0 0]      [0 0 0]      [0]
                if(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0 0 0]      [0]
                empty^#(x1) = [0 0 0] x1 + [0]
                              [0 0 0]      [0]
                              [0 0 0]      [0]
                c_0() = [0]
                        [0]
                        [0]
                c_1() = [0]
                        [0]
                        [0]
                head^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_2(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                tail^#(x1) = [0 0 0] x1 + [0]
                             [0 0 0]      [0]
                             [0 0 0]      [0]
                c_3() = [0]
                        [0]
                        [0]
                c_4(x1) = [0 0 0] x1 + [0]
                          [0 0 0]      [0]
                          [0 0 0]      [0]
                rev^#(x1) = [0 0 0] x1 + [0]
                            [0 0 0]      [0]
                            [0 0 0]      [0]
                c_5() = [0]
                        [0]
                        [0]
                c_6(x1, x2, x3) = [0 0 0] x1 + [0 0 0] x2 + [0 0 0] x3 + [0]
                                  [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                  [0 0 0]      [0 0 0]      [0 0 0]      [0]
                rev2^#(x1, x2) = [0 0 0] x1 + [0 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                last^#(x1, x2) = [3 0 0] x1 + [3 0 0] x2 + [0]
                                 [0 0 0]      [0 0 0]      [0]
                                 [0 0 0]      [0 0 0]      [0]
                c_7(x1) = [1 0 0] x1 + [0]
                          [0 1 0]      [0]
                          [0 0 1]      [0]
                if^#(x1, x2, x3) = [3 0 0] x1 + [3 3 3] x2 + [3 0 0] x3 + [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                                   [0 0 0]      [0 0 0]      [0 0 0]      [0]
                c_8(x1) = [1 0 0] x1 + [0]
                          [0 1 0]      [0]
                          [0 0 1]      [0]
                c_9(x1) = [1 0 0] x1 + [0]
                          [0 1 0]      [0]
                          [0 0 1]      [0]
                c_10() = [0]
                         [0]
                         [0]
                c_11(x1) = [0 0 0] x1 + [0]
                           [0 0 0]      [0]
                           [0 0 0]      [0]
             Complexity induced by the adequate RMI: YES(?,O(n^3))
             
             We have not generated a proof for the resulting sub-problem.
    
    2) 'wdg' failed due to the following reason:
         Transformation Details:
         -----------------------
           We have computed the following set of weak (innermost) dependency pairs:
           
             {  1: empty^#(nil()) -> c_0()
              , 2: empty^#(cons(x, l)) -> c_1()
              , 3: head^#(cons(x, l)) -> c_2(x)
              , 4: tail^#(nil()) -> c_3()
              , 5: tail^#(cons(x, l)) -> c_4(l)
              , 6: rev^#(nil()) -> c_5()
              , 7: rev^#(cons(x, l)) -> c_6(x, l, rev2^#(x, l))
              , 8: last^#(x, l) -> c_7(if^#(empty(l), x, l))
              , 9: if^#(true(), x, l) -> c_8(x)
              , 10: if^#(false(), x, l) -> c_9(last^#(head(l), tail(l)))
              , 11: rev2^#(x, nil()) -> c_10()
              , 12: rev2^#(x, cons(y, l)) -> c_11(rev^#(cons(x, rev2(y, l))))}
           
           Following Dependency Graph (modulo SCCs) was computed. (Answers to
           subproofs are indicated to the right.)
           
             ->{8,10}                                                    [       MAYBE        ]
                |
                `->{9}                                                   [         NA         ]
             
             ->{7,12}                                                    [     inherited      ]
                |
                `->{11}                                                  [         NA         ]
             
             ->{6}                                                       [    YES(?,O(1))     ]
             
             ->{5}                                                       [   YES(?,O(n^2))    ]
             
             ->{4}                                                       [    YES(?,O(1))     ]
             
             ->{3}                                                       [   YES(?,O(n^2))    ]
             
             ->{2}                                                       [    YES(?,O(1))     ]
             
             ->{1}                                                       [    YES(?,O(1))     ]
             
           
         
         Sub-problems:
         -------------
           * Path {1}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(c_2) = {},
                 Uargs(tail^#) = {}, Uargs(c_4) = {}, Uargs(rev^#) = {},
                 Uargs(c_6) = {}, Uargs(rev2^#) = {}, Uargs(last^#) = {},
                 Uargs(c_7) = {}, Uargs(if^#) = {}, Uargs(c_8) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                nil() = [0]
                        [0]
                true() = [0]
                         [0]
                cons(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                false() = [0]
                          [0]
                head(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                tail(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                rev(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev1(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                rev2(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                last(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                if(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                 [0 0]      [0 0]      [0 0]      [0]
                empty^#(x1) = [0 0] x1 + [0]
                              [0 0]      [0]
                c_0() = [0]
                        [0]
                c_1() = [0]
                        [0]
                head^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_2(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                tail^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_3() = [0]
                        [0]
                c_4(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev^#(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                c_5() = [0]
                        [0]
                c_6(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                  [0 0]      [0 0]      [0 0]      [0]
                rev2^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                last^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                c_7(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                if^#(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                   [0 0]      [0 0]      [0 0]      [0]
                c_8(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                c_9(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                c_10() = [0]
                         [0]
                c_11(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 2'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    DP runtime-complexity with respect to
               Strict Rules: {empty^#(nil()) -> c_0()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(empty^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                nil() = [2]
                        [2]
                empty^#(x1) = [2 0] x1 + [7]
                              [2 2]      [7]
                c_0() = [0]
                        [1]
           
           * Path {2}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(c_2) = {},
                 Uargs(tail^#) = {}, Uargs(c_4) = {}, Uargs(rev^#) = {},
                 Uargs(c_6) = {}, Uargs(rev2^#) = {}, Uargs(last^#) = {},
                 Uargs(c_7) = {}, Uargs(if^#) = {}, Uargs(c_8) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                nil() = [0]
                        [0]
                true() = [0]
                         [0]
                cons(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                false() = [0]
                          [0]
                head(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                tail(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                rev(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev1(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                rev2(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                last(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                if(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                 [0 0]      [0 0]      [0 0]      [0]
                empty^#(x1) = [0 0] x1 + [0]
                              [0 0]      [0]
                c_0() = [0]
                        [0]
                c_1() = [0]
                        [0]
                head^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_2(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                tail^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_3() = [0]
                        [0]
                c_4(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev^#(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                c_5() = [0]
                        [0]
                c_6(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                  [0 0]      [0 0]      [0 0]      [0]
                rev2^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                last^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                c_7(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                if^#(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                   [0 0]      [0 0]      [0 0]      [0]
                c_8(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                c_9(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                c_10() = [0]
                         [0]
                c_11(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 2'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    DP runtime-complexity with respect to
               Strict Rules: {empty^#(cons(x, l)) -> c_1()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(cons) = {}, Uargs(empty^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                cons(x1, x2) = [0 0] x1 + [0 0] x2 + [2]
                               [0 0]      [0 0]      [2]
                empty^#(x1) = [2 0] x1 + [7]
                              [2 2]      [7]
                c_1() = [0]
                        [1]
           
           * Path {3}: YES(?,O(n^2))
             -----------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(c_2) = {},
                 Uargs(tail^#) = {}, Uargs(c_4) = {}, Uargs(rev^#) = {},
                 Uargs(c_6) = {}, Uargs(rev2^#) = {}, Uargs(last^#) = {},
                 Uargs(c_7) = {}, Uargs(if^#) = {}, Uargs(c_8) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                nil() = [0]
                        [0]
                true() = [0]
                         [0]
                cons(x1, x2) = [1 1] x1 + [0 0] x2 + [0]
                               [0 1]      [0 0]      [0]
                false() = [0]
                          [0]
                head(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                tail(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                rev(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev1(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                rev2(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                last(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                if(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                 [0 0]      [0 0]      [0 0]      [0]
                empty^#(x1) = [0 0] x1 + [0]
                              [0 0]      [0]
                c_0() = [0]
                        [0]
                c_1() = [0]
                        [0]
                head^#(x1) = [3 3] x1 + [0]
                             [0 0]      [0]
                c_2(x1) = [1 0] x1 + [0]
                          [0 0]      [0]
                tail^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_3() = [0]
                        [0]
                c_4(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev^#(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                c_5() = [0]
                        [0]
                c_6(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                  [0 0]      [0 0]      [0 0]      [0]
                rev2^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                last^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                c_7(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                if^#(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                   [0 0]      [0 0]      [0 0]      [0]
                c_8(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                c_9(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                c_10() = [0]
                         [0]
                c_11(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 2'
             --------------------------------------
             Answer:           YES(?,O(n^1))
             Input Problem:    DP runtime-complexity with respect to
               Strict Rules: {head^#(cons(x, l)) -> c_2(x)}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(cons) = {}, Uargs(head^#) = {}, Uargs(c_2) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                cons(x1, x2) = [1 2] x1 + [0 0] x2 + [2]
                               [0 0]      [0 0]      [2]
                head^#(x1) = [2 2] x1 + [7]
                             [2 0]      [7]
                c_2(x1) = [0 0] x1 + [0]
                          [0 0]      [1]
           
           * Path {4}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(c_2) = {},
                 Uargs(tail^#) = {}, Uargs(c_4) = {}, Uargs(rev^#) = {},
                 Uargs(c_6) = {}, Uargs(rev2^#) = {}, Uargs(last^#) = {},
                 Uargs(c_7) = {}, Uargs(if^#) = {}, Uargs(c_8) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                nil() = [0]
                        [0]
                true() = [0]
                         [0]
                cons(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                false() = [0]
                          [0]
                head(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                tail(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                rev(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev1(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                rev2(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                last(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                if(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                 [0 0]      [0 0]      [0 0]      [0]
                empty^#(x1) = [0 0] x1 + [0]
                              [0 0]      [0]
                c_0() = [0]
                        [0]
                c_1() = [0]
                        [0]
                head^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_2(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                tail^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_3() = [0]
                        [0]
                c_4(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev^#(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                c_5() = [0]
                        [0]
                c_6(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                  [0 0]      [0 0]      [0 0]      [0]
                rev2^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                last^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                c_7(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                if^#(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                   [0 0]      [0 0]      [0 0]      [0]
                c_8(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                c_9(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                c_10() = [0]
                         [0]
                c_11(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 2'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    DP runtime-complexity with respect to
               Strict Rules: {tail^#(nil()) -> c_3()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(tail^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                nil() = [2]
                        [2]
                tail^#(x1) = [2 0] x1 + [7]
                             [2 2]      [7]
                c_3() = [0]
                        [1]
           
           * Path {5}: YES(?,O(n^2))
             -----------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(c_2) = {},
                 Uargs(tail^#) = {}, Uargs(c_4) = {}, Uargs(rev^#) = {},
                 Uargs(c_6) = {}, Uargs(rev2^#) = {}, Uargs(last^#) = {},
                 Uargs(c_7) = {}, Uargs(if^#) = {}, Uargs(c_8) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                nil() = [0]
                        [0]
                true() = [0]
                         [0]
                cons(x1, x2) = [1 1] x1 + [0 0] x2 + [0]
                               [0 1]      [0 0]      [0]
                false() = [0]
                          [0]
                head(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                tail(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                rev(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev1(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                rev2(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                last(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                if(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                 [0 0]      [0 0]      [0 0]      [0]
                empty^#(x1) = [0 0] x1 + [0]
                              [0 0]      [0]
                c_0() = [0]
                        [0]
                c_1() = [0]
                        [0]
                head^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_2(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                tail^#(x1) = [3 3] x1 + [0]
                             [0 0]      [0]
                c_3() = [0]
                        [0]
                c_4(x1) = [1 0] x1 + [0]
                          [0 0]      [0]
                rev^#(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                c_5() = [0]
                        [0]
                c_6(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                  [0 0]      [0 0]      [0 0]      [0]
                rev2^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                last^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                c_7(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                if^#(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                   [0 0]      [0 0]      [0 0]      [0]
                c_8(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                c_9(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                c_10() = [0]
                         [0]
                c_11(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 2'
             --------------------------------------
             Answer:           YES(?,O(n^1))
             Input Problem:    DP runtime-complexity with respect to
               Strict Rules: {tail^#(cons(x, l)) -> c_4(l)}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(cons) = {}, Uargs(tail^#) = {}, Uargs(c_4) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                cons(x1, x2) = [0 0] x1 + [1 2] x2 + [2]
                               [0 0]      [0 0]      [2]
                tail^#(x1) = [2 2] x1 + [7]
                             [2 0]      [7]
                c_4(x1) = [0 0] x1 + [0]
                          [0 0]      [1]
           
           * Path {6}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(c_2) = {},
                 Uargs(tail^#) = {}, Uargs(c_4) = {}, Uargs(rev^#) = {},
                 Uargs(c_6) = {}, Uargs(rev2^#) = {}, Uargs(last^#) = {},
                 Uargs(c_7) = {}, Uargs(if^#) = {}, Uargs(c_8) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                nil() = [0]
                        [0]
                true() = [0]
                         [0]
                cons(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                false() = [0]
                          [0]
                head(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                tail(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
                rev(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev1(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                rev2(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                last(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                if(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                 [0 0]      [0 0]      [0 0]      [0]
                empty^#(x1) = [0 0] x1 + [0]
                              [0 0]      [0]
                c_0() = [0]
                        [0]
                c_1() = [0]
                        [0]
                head^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_2(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                tail^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_3() = [0]
                        [0]
                c_4(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev^#(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                c_5() = [0]
                        [0]
                c_6(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                  [0 0]      [0 0]      [0 0]      [0]
                rev2^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                last^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                c_7(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                if^#(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                   [0 0]      [0 0]      [0 0]      [0]
                c_8(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                c_9(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                c_10() = [0]
                         [0]
                c_11(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 2'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    DP runtime-complexity with respect to
               Strict Rules: {rev^#(nil()) -> c_5()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(rev^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                nil() = [2]
                        [2]
                rev^#(x1) = [2 0] x1 + [7]
                            [2 2]      [7]
                c_5() = [0]
                        [1]
           
           * Path {7,12}: inherited
             ----------------------
             
             This path is subsumed by the proof of path {7,12}->{11}.
           
           * Path {7,12}->{11}: NA
             ---------------------
             
             The usable rules for this path are:
             
               {  rev2(x, nil()) -> nil()
                , rev2(x, cons(y, l)) -> rev(cons(x, rev2(y, l)))
                , rev(nil()) -> nil()
                , rev(cons(x, l)) -> cons(rev1(x, l), rev2(x, l))}
             
             The weight gap principle does not apply:
               The input cannot be shown compatible
             Complexity induced by the adequate RMI: MAYBE
             
             We have not generated a proof for the resulting sub-problem.
           
           * Path {8,10}: MAYBE
             ------------------
             
             The usable rules for this path are:
             
               {  empty(nil()) -> true()
                , empty(cons(x, l)) -> false()
                , head(cons(x, l)) -> x
                , tail(nil()) -> nil()
                , tail(cons(x, l)) -> l}
             
             The weight gap principle does not apply:
               The input cannot be shown compatible
             Complexity induced by the adequate RMI: MAYBE
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 2'
             --------------------------------------
             Answer:           MAYBE
             Input Problem:    runtime-complexity with respect to
               Rules:
                 {  last^#(x, l) -> c_7(if^#(empty(l), x, l))
                  , if^#(false(), x, l) -> c_9(last^#(head(l), tail(l)))
                  , empty(nil()) -> true()
                  , empty(cons(x, l)) -> false()
                  , head(cons(x, l)) -> x
                  , tail(nil()) -> nil()
                  , tail(cons(x, l)) -> l}
             
             Proof Output:    
               The input cannot be shown compatible
           
           * Path {8,10}->{9}: NA
             --------------------
             
             The usable rules for this path are:
             
               {  empty(nil()) -> true()
                , empty(cons(x, l)) -> false()
                , head(cons(x, l)) -> x
                , tail(nil()) -> nil()
                , tail(cons(x, l)) -> l}
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {1}, Uargs(cons) = {}, Uargs(head) = {1},
                 Uargs(tail) = {1}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(c_2) = {},
                 Uargs(tail^#) = {}, Uargs(c_4) = {}, Uargs(rev^#) = {},
                 Uargs(c_6) = {}, Uargs(rev2^#) = {}, Uargs(last^#) = {1, 2},
                 Uargs(c_7) = {1}, Uargs(if^#) = {1, 2, 3}, Uargs(c_8) = {1},
                 Uargs(c_9) = {1}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [2 0] x1 + [3]
                            [3 3]      [3]
                nil() = [2]
                        [0]
                true() = [0]
                         [1]
                cons(x1, x2) = [1 1] x1 + [1 3] x2 + [2]
                               [0 1]      [0 1]      [0]
                false() = [0]
                          [1]
                head(x1) = [3 3] x1 + [3]
                           [3 3]      [3]
                tail(x1) = [2 0] x1 + [0]
                           [0 2]      [0]
                rev(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev1(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                rev2(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                last(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                               [0 0]      [0 0]      [0]
                if(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                 [0 0]      [0 0]      [0 0]      [0]
                empty^#(x1) = [0 0] x1 + [0]
                              [0 0]      [0]
                c_0() = [0]
                        [0]
                c_1() = [0]
                        [0]
                head^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_2(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                tail^#(x1) = [0 0] x1 + [0]
                             [0 0]      [0]
                c_3() = [0]
                        [0]
                c_4(x1) = [0 0] x1 + [0]
                          [0 0]      [0]
                rev^#(x1) = [0 0] x1 + [0]
                            [0 0]      [0]
                c_5() = [0]
                        [0]
                c_6(x1, x2, x3) = [0 0] x1 + [0 0] x2 + [0 0] x3 + [0]
                                  [0 0]      [0 0]      [0 0]      [0]
                rev2^#(x1, x2) = [0 0] x1 + [0 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                last^#(x1, x2) = [3 0] x1 + [3 0] x2 + [0]
                                 [0 0]      [0 0]      [0]
                c_7(x1) = [1 0] x1 + [0]
                          [0 1]      [0]
                if^#(x1, x2, x3) = [3 0] x1 + [3 3] x2 + [3 0] x3 + [0]
                                   [0 0]      [0 0]      [0 0]      [0]
                c_8(x1) = [1 0] x1 + [0]
                          [0 1]      [0]
                c_9(x1) = [1 0] x1 + [0]
                          [0 1]      [0]
                c_10() = [0]
                         [0]
                c_11(x1) = [0 0] x1 + [0]
                           [0 0]      [0]
             Complexity induced by the adequate RMI: YES(?,O(n^2))
             
             We have not generated a proof for the resulting sub-problem.
    
    3) 'wdg' failed due to the following reason:
         Transformation Details:
         -----------------------
           We have computed the following set of weak (innermost) dependency pairs:
           
             {  1: empty^#(nil()) -> c_0()
              , 2: empty^#(cons(x, l)) -> c_1()
              , 3: head^#(cons(x, l)) -> c_2(x)
              , 4: tail^#(nil()) -> c_3()
              , 5: tail^#(cons(x, l)) -> c_4(l)
              , 6: rev^#(nil()) -> c_5()
              , 7: rev^#(cons(x, l)) -> c_6(x, l, rev2^#(x, l))
              , 8: last^#(x, l) -> c_7(if^#(empty(l), x, l))
              , 9: if^#(true(), x, l) -> c_8(x)
              , 10: if^#(false(), x, l) -> c_9(last^#(head(l), tail(l)))
              , 11: rev2^#(x, nil()) -> c_10()
              , 12: rev2^#(x, cons(y, l)) -> c_11(rev^#(cons(x, rev2(y, l))))}
           
           Following Dependency Graph (modulo SCCs) was computed. (Answers to
           subproofs are indicated to the right.)
           
             ->{8,10}                                                    [       MAYBE        ]
                |
                `->{9}                                                   [         NA         ]
             
             ->{7,12}                                                    [     inherited      ]
                |
                `->{11}                                                  [         NA         ]
             
             ->{6}                                                       [    YES(?,O(1))     ]
             
             ->{5}                                                       [   YES(?,O(n^1))    ]
             
             ->{4}                                                       [    YES(?,O(1))     ]
             
             ->{3}                                                       [   YES(?,O(n^1))    ]
             
             ->{2}                                                       [    YES(?,O(1))     ]
             
             ->{1}                                                       [    YES(?,O(1))     ]
             
           
         
         Sub-problems:
         -------------
           * Path {1}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(c_2) = {},
                 Uargs(tail^#) = {}, Uargs(c_4) = {}, Uargs(rev^#) = {},
                 Uargs(c_6) = {}, Uargs(rev2^#) = {}, Uargs(last^#) = {},
                 Uargs(c_7) = {}, Uargs(if^#) = {}, Uargs(c_8) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0] x1 + [0]
                nil() = [0]
                true() = [0]
                cons(x1, x2) = [0] x1 + [0] x2 + [0]
                false() = [0]
                head(x1) = [0] x1 + [0]
                tail(x1) = [0] x1 + [0]
                rev(x1) = [0] x1 + [0]
                rev1(x1, x2) = [0] x1 + [0] x2 + [0]
                rev2(x1, x2) = [0] x1 + [0] x2 + [0]
                last(x1, x2) = [0] x1 + [0] x2 + [0]
                if(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                empty^#(x1) = [0] x1 + [0]
                c_0() = [0]
                c_1() = [0]
                head^#(x1) = [0] x1 + [0]
                c_2(x1) = [0] x1 + [0]
                tail^#(x1) = [0] x1 + [0]
                c_3() = [0]
                c_4(x1) = [0] x1 + [0]
                rev^#(x1) = [0] x1 + [0]
                c_5() = [0]
                c_6(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                rev2^#(x1, x2) = [0] x1 + [0] x2 + [0]
                last^#(x1, x2) = [0] x1 + [0] x2 + [0]
                c_7(x1) = [0] x1 + [0]
                if^#(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                c_8(x1) = [0] x1 + [0]
                c_9(x1) = [0] x1 + [0]
                c_10() = [0]
                c_11(x1) = [0] x1 + [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 1'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    DP runtime-complexity with respect to
               Strict Rules: {empty^#(nil()) -> c_0()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(empty^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                nil() = [7]
                empty^#(x1) = [1] x1 + [7]
                c_0() = [1]
           
           * Path {2}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(c_2) = {},
                 Uargs(tail^#) = {}, Uargs(c_4) = {}, Uargs(rev^#) = {},
                 Uargs(c_6) = {}, Uargs(rev2^#) = {}, Uargs(last^#) = {},
                 Uargs(c_7) = {}, Uargs(if^#) = {}, Uargs(c_8) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0] x1 + [0]
                nil() = [0]
                true() = [0]
                cons(x1, x2) = [0] x1 + [0] x2 + [0]
                false() = [0]
                head(x1) = [0] x1 + [0]
                tail(x1) = [0] x1 + [0]
                rev(x1) = [0] x1 + [0]
                rev1(x1, x2) = [0] x1 + [0] x2 + [0]
                rev2(x1, x2) = [0] x1 + [0] x2 + [0]
                last(x1, x2) = [0] x1 + [0] x2 + [0]
                if(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                empty^#(x1) = [0] x1 + [0]
                c_0() = [0]
                c_1() = [0]
                head^#(x1) = [0] x1 + [0]
                c_2(x1) = [0] x1 + [0]
                tail^#(x1) = [0] x1 + [0]
                c_3() = [0]
                c_4(x1) = [0] x1 + [0]
                rev^#(x1) = [0] x1 + [0]
                c_5() = [0]
                c_6(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                rev2^#(x1, x2) = [0] x1 + [0] x2 + [0]
                last^#(x1, x2) = [0] x1 + [0] x2 + [0]
                c_7(x1) = [0] x1 + [0]
                if^#(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                c_8(x1) = [0] x1 + [0]
                c_9(x1) = [0] x1 + [0]
                c_10() = [0]
                c_11(x1) = [0] x1 + [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 1'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    DP runtime-complexity with respect to
               Strict Rules: {empty^#(cons(x, l)) -> c_1()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(cons) = {}, Uargs(empty^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                cons(x1, x2) = [0] x1 + [0] x2 + [7]
                empty^#(x1) = [1] x1 + [7]
                c_1() = [1]
           
           * Path {3}: YES(?,O(n^1))
             -----------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(c_2) = {},
                 Uargs(tail^#) = {}, Uargs(c_4) = {}, Uargs(rev^#) = {},
                 Uargs(c_6) = {}, Uargs(rev2^#) = {}, Uargs(last^#) = {},
                 Uargs(c_7) = {}, Uargs(if^#) = {}, Uargs(c_8) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0] x1 + [0]
                nil() = [0]
                true() = [0]
                cons(x1, x2) = [1] x1 + [0] x2 + [0]
                false() = [0]
                head(x1) = [0] x1 + [0]
                tail(x1) = [0] x1 + [0]
                rev(x1) = [0] x1 + [0]
                rev1(x1, x2) = [0] x1 + [0] x2 + [0]
                rev2(x1, x2) = [0] x1 + [0] x2 + [0]
                last(x1, x2) = [0] x1 + [0] x2 + [0]
                if(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                empty^#(x1) = [0] x1 + [0]
                c_0() = [0]
                c_1() = [0]
                head^#(x1) = [3] x1 + [0]
                c_2(x1) = [1] x1 + [0]
                tail^#(x1) = [0] x1 + [0]
                c_3() = [0]
                c_4(x1) = [0] x1 + [0]
                rev^#(x1) = [0] x1 + [0]
                c_5() = [0]
                c_6(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                rev2^#(x1, x2) = [0] x1 + [0] x2 + [0]
                last^#(x1, x2) = [0] x1 + [0] x2 + [0]
                c_7(x1) = [0] x1 + [0]
                if^#(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                c_8(x1) = [0] x1 + [0]
                c_9(x1) = [0] x1 + [0]
                c_10() = [0]
                c_11(x1) = [0] x1 + [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 1'
             --------------------------------------
             Answer:           YES(?,O(n^1))
             Input Problem:    DP runtime-complexity with respect to
               Strict Rules: {head^#(cons(x, l)) -> c_2(x)}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(cons) = {}, Uargs(head^#) = {}, Uargs(c_2) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                cons(x1, x2) = [1] x1 + [0] x2 + [7]
                head^#(x1) = [1] x1 + [7]
                c_2(x1) = [1] x1 + [1]
           
           * Path {4}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(c_2) = {},
                 Uargs(tail^#) = {}, Uargs(c_4) = {}, Uargs(rev^#) = {},
                 Uargs(c_6) = {}, Uargs(rev2^#) = {}, Uargs(last^#) = {},
                 Uargs(c_7) = {}, Uargs(if^#) = {}, Uargs(c_8) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0] x1 + [0]
                nil() = [0]
                true() = [0]
                cons(x1, x2) = [0] x1 + [0] x2 + [0]
                false() = [0]
                head(x1) = [0] x1 + [0]
                tail(x1) = [0] x1 + [0]
                rev(x1) = [0] x1 + [0]
                rev1(x1, x2) = [0] x1 + [0] x2 + [0]
                rev2(x1, x2) = [0] x1 + [0] x2 + [0]
                last(x1, x2) = [0] x1 + [0] x2 + [0]
                if(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                empty^#(x1) = [0] x1 + [0]
                c_0() = [0]
                c_1() = [0]
                head^#(x1) = [0] x1 + [0]
                c_2(x1) = [0] x1 + [0]
                tail^#(x1) = [0] x1 + [0]
                c_3() = [0]
                c_4(x1) = [0] x1 + [0]
                rev^#(x1) = [0] x1 + [0]
                c_5() = [0]
                c_6(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                rev2^#(x1, x2) = [0] x1 + [0] x2 + [0]
                last^#(x1, x2) = [0] x1 + [0] x2 + [0]
                c_7(x1) = [0] x1 + [0]
                if^#(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                c_8(x1) = [0] x1 + [0]
                c_9(x1) = [0] x1 + [0]
                c_10() = [0]
                c_11(x1) = [0] x1 + [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 1'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    DP runtime-complexity with respect to
               Strict Rules: {tail^#(nil()) -> c_3()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(tail^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                nil() = [7]
                tail^#(x1) = [1] x1 + [7]
                c_3() = [1]
           
           * Path {5}: YES(?,O(n^1))
             -----------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(c_2) = {},
                 Uargs(tail^#) = {}, Uargs(c_4) = {}, Uargs(rev^#) = {},
                 Uargs(c_6) = {}, Uargs(rev2^#) = {}, Uargs(last^#) = {},
                 Uargs(c_7) = {}, Uargs(if^#) = {}, Uargs(c_8) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0] x1 + [0]
                nil() = [0]
                true() = [0]
                cons(x1, x2) = [1] x1 + [0] x2 + [0]
                false() = [0]
                head(x1) = [0] x1 + [0]
                tail(x1) = [0] x1 + [0]
                rev(x1) = [0] x1 + [0]
                rev1(x1, x2) = [0] x1 + [0] x2 + [0]
                rev2(x1, x2) = [0] x1 + [0] x2 + [0]
                last(x1, x2) = [0] x1 + [0] x2 + [0]
                if(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                empty^#(x1) = [0] x1 + [0]
                c_0() = [0]
                c_1() = [0]
                head^#(x1) = [0] x1 + [0]
                c_2(x1) = [0] x1 + [0]
                tail^#(x1) = [3] x1 + [0]
                c_3() = [0]
                c_4(x1) = [1] x1 + [0]
                rev^#(x1) = [0] x1 + [0]
                c_5() = [0]
                c_6(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                rev2^#(x1, x2) = [0] x1 + [0] x2 + [0]
                last^#(x1, x2) = [0] x1 + [0] x2 + [0]
                c_7(x1) = [0] x1 + [0]
                if^#(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                c_8(x1) = [0] x1 + [0]
                c_9(x1) = [0] x1 + [0]
                c_10() = [0]
                c_11(x1) = [0] x1 + [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 1'
             --------------------------------------
             Answer:           YES(?,O(n^1))
             Input Problem:    DP runtime-complexity with respect to
               Strict Rules: {tail^#(cons(x, l)) -> c_4(l)}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(cons) = {}, Uargs(tail^#) = {}, Uargs(c_4) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                cons(x1, x2) = [0] x1 + [1] x2 + [7]
                tail^#(x1) = [1] x1 + [7]
                c_4(x1) = [1] x1 + [1]
           
           * Path {6}: YES(?,O(1))
             ---------------------
             
             The usable rules of this path are empty.
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {}, Uargs(cons) = {}, Uargs(head) = {},
                 Uargs(tail) = {}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(c_2) = {},
                 Uargs(tail^#) = {}, Uargs(c_4) = {}, Uargs(rev^#) = {},
                 Uargs(c_6) = {}, Uargs(rev2^#) = {}, Uargs(last^#) = {},
                 Uargs(c_7) = {}, Uargs(if^#) = {}, Uargs(c_8) = {},
                 Uargs(c_9) = {}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [0] x1 + [0]
                nil() = [0]
                true() = [0]
                cons(x1, x2) = [0] x1 + [0] x2 + [0]
                false() = [0]
                head(x1) = [0] x1 + [0]
                tail(x1) = [0] x1 + [0]
                rev(x1) = [0] x1 + [0]
                rev1(x1, x2) = [0] x1 + [0] x2 + [0]
                rev2(x1, x2) = [0] x1 + [0] x2 + [0]
                last(x1, x2) = [0] x1 + [0] x2 + [0]
                if(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                empty^#(x1) = [0] x1 + [0]
                c_0() = [0]
                c_1() = [0]
                head^#(x1) = [0] x1 + [0]
                c_2(x1) = [0] x1 + [0]
                tail^#(x1) = [0] x1 + [0]
                c_3() = [0]
                c_4(x1) = [0] x1 + [0]
                rev^#(x1) = [0] x1 + [0]
                c_5() = [0]
                c_6(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                rev2^#(x1, x2) = [0] x1 + [0] x2 + [0]
                last^#(x1, x2) = [0] x1 + [0] x2 + [0]
                c_7(x1) = [0] x1 + [0]
                if^#(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                c_8(x1) = [0] x1 + [0]
                c_9(x1) = [0] x1 + [0]
                c_10() = [0]
                c_11(x1) = [0] x1 + [0]
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 1'
             --------------------------------------
             Answer:           YES(?,O(1))
             Input Problem:    DP runtime-complexity with respect to
               Strict Rules: {rev^#(nil()) -> c_5()}
               Weak Rules: {}
             
             Proof Output:    
               The following argument positions are usable:
                 Uargs(rev^#) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                nil() = [7]
                rev^#(x1) = [1] x1 + [7]
                c_5() = [1]
           
           * Path {7,12}: inherited
             ----------------------
             
             This path is subsumed by the proof of path {7,12}->{11}.
           
           * Path {7,12}->{11}: NA
             ---------------------
             
             The usable rules for this path are:
             
               {  rev2(x, nil()) -> nil()
                , rev2(x, cons(y, l)) -> rev(cons(x, rev2(y, l)))
                , rev(nil()) -> nil()
                , rev(cons(x, l)) -> cons(rev1(x, l), rev2(x, l))}
             
             The weight gap principle does not apply:
               The input cannot be shown compatible
             Complexity induced by the adequate RMI: MAYBE
             
             We have not generated a proof for the resulting sub-problem.
           
           * Path {8,10}: MAYBE
             ------------------
             
             The usable rules for this path are:
             
               {  empty(nil()) -> true()
                , empty(cons(x, l)) -> false()
                , head(cons(x, l)) -> x
                , tail(nil()) -> nil()
                , tail(cons(x, l)) -> l}
             
             The weight gap principle does not apply:
               The input cannot be shown compatible
             Complexity induced by the adequate RMI: MAYBE
             
             We apply the sub-processor on the resulting sub-problem:
             
             'matrix-interpretation of dimension 1'
             --------------------------------------
             Answer:           MAYBE
             Input Problem:    runtime-complexity with respect to
               Rules:
                 {  last^#(x, l) -> c_7(if^#(empty(l), x, l))
                  , if^#(false(), x, l) -> c_9(last^#(head(l), tail(l)))
                  , empty(nil()) -> true()
                  , empty(cons(x, l)) -> false()
                  , head(cons(x, l)) -> x
                  , tail(nil()) -> nil()
                  , tail(cons(x, l)) -> l}
             
             Proof Output:    
               The input cannot be shown compatible
           
           * Path {8,10}->{9}: NA
             --------------------
             
             The usable rules for this path are:
             
               {  empty(nil()) -> true()
                , empty(cons(x, l)) -> false()
                , head(cons(x, l)) -> x
                , tail(nil()) -> nil()
                , tail(cons(x, l)) -> l}
             
             The weightgap principle applies, using the following adequate RMI:
               The following argument positions are usable:
                 Uargs(empty) = {1}, Uargs(cons) = {}, Uargs(head) = {1},
                 Uargs(tail) = {1}, Uargs(rev) = {}, Uargs(rev1) = {},
                 Uargs(rev2) = {}, Uargs(last) = {}, Uargs(if) = {},
                 Uargs(empty^#) = {}, Uargs(head^#) = {}, Uargs(c_2) = {},
                 Uargs(tail^#) = {}, Uargs(c_4) = {}, Uargs(rev^#) = {},
                 Uargs(c_6) = {}, Uargs(rev2^#) = {}, Uargs(last^#) = {1, 2},
                 Uargs(c_7) = {1}, Uargs(if^#) = {1, 2, 3}, Uargs(c_8) = {1},
                 Uargs(c_9) = {1}, Uargs(c_11) = {}
               We have the following constructor-restricted matrix interpretation:
               Interpretation Functions:
                empty(x1) = [1] x1 + [3]
                nil() = [3]
                true() = [1]
                cons(x1, x2) = [1] x1 + [1] x2 + [3]
                false() = [1]
                head(x1) = [1] x1 + [3]
                tail(x1) = [1] x1 + [3]
                rev(x1) = [0] x1 + [0]
                rev1(x1, x2) = [0] x1 + [0] x2 + [0]
                rev2(x1, x2) = [0] x1 + [0] x2 + [0]
                last(x1, x2) = [0] x1 + [0] x2 + [0]
                if(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                empty^#(x1) = [0] x1 + [0]
                c_0() = [0]
                c_1() = [0]
                head^#(x1) = [0] x1 + [0]
                c_2(x1) = [0] x1 + [0]
                tail^#(x1) = [0] x1 + [0]
                c_3() = [0]
                c_4(x1) = [0] x1 + [0]
                rev^#(x1) = [0] x1 + [0]
                c_5() = [0]
                c_6(x1, x2, x3) = [0] x1 + [0] x2 + [0] x3 + [0]
                rev2^#(x1, x2) = [0] x1 + [0] x2 + [0]
                last^#(x1, x2) = [3] x1 + [3] x2 + [0]
                c_7(x1) = [1] x1 + [0]
                if^#(x1, x2, x3) = [3] x1 + [3] x2 + [3] x3 + [0]
                c_8(x1) = [1] x1 + [0]
                c_9(x1) = [1] x1 + [0]
                c_10() = [0]
                c_11(x1) = [0] x1 + [0]
             Complexity induced by the adequate RMI: YES(?,O(n^1))
             
             We have not generated a proof for the resulting sub-problem.
    
    4) 'matrix-interpretation of dimension 1' failed due to the following reason:
         The input cannot be shown compatible
    
    5) 'Bounds with perSymbol-enrichment and initial automaton 'match'' failed due to the following reason:
         match-boundness of the problem could not be verified.
    
    6) 'Bounds with minimal-enrichment and initial automaton 'match'' failed due to the following reason:
         match-boundness of the problem could not be verified.