implicit-state-passing

This is a follow-on question from an earlier question from Mat's answer Starting with this e([number(0)] , t1 , Uc0 , Uc0, Bc0 , Bc0) --> []. e([number(1)] , t2 , Uc0 , Uc0, Bc0 , Bc0) --> []. e([number(2)] , t3 , Uc0 , Uc0, Bc0 , Bc0) --> []. e([op(neg),[Arg]] , u1(E) , [_|Uc0], Uc1, Bc0 , Bc1) --> [_], e(Arg , E , Uc0, Uc1, Bc0, Bc1). e([op(ln),[Arg]] , u2(E) , [_|Uc0], Uc1, Bc0 , Bc1) --> [_], e(Arg , E , Uc0, Uc1, Bc0, Bc1). e([op(add),[Left,Right]], b1(E0,E1) , Uc0 , Uc2, [_|Bc0], Bc2) --> [_,_], e(Left, E0, Uc0, Uc1, Bc0, Bc1), e(Right, E1, Uc1, Uc2, Bc1, Bc2). e([op(sub),[Left,Right]],

I often end up writing code in Prolog which involves some arithmetic calculation (or state information important throughout the program), by means of first obtaining the value stored in a predicate, then recalculating the value and finally storing the value using retractall and assert because in Prolog we cannot assign values to variable twice using is (thus making almost every variable that needs modification, global). I have come to know that this is not a good practice in Prolog. In this regard I would like to ask: Why is it a bad practice in Prolog (though i myself don't like to go through

This question starts from Mat's answer to Algorithm improvement for enumerating binary trees which has only one input value that determines the number of all nodes for the binary tree, and the need to be able to have two input values with one being the number of unary nodes and the other being the number of binary nodes. While I was able to derive a solution by using listing/1 and threading extra state variables: e(t, B, B, U, U). e(u(E), B0, B1, [_|U0], U1) :- e(E, B0, B1, U0, U1). e(b(E0, E1), [_|B0], B2, U0, U2) :- e(E0, B0, B1, U0, U1), e(E1, B1, B2, U1, U2). e(U,B,Es) :- length(Bs, B),