Can a very simple algorithm be efficient for solving the SAT problem