A Proof for P vs. NP Problem

http://arxiv.org/abs/1005.3010

``This paper proposes a theoretic proof for P vs. NP problem. The central idea of this proof is a recursive definition for Turing machine (shortly TM). By the definition, an infinite sequence of TM is constructed and it is proved that the sequence includes all TM. Based on these TM, the class D that includes all decidable languages is defined. By proving P=D, the result P=NP is proved.''

>>17
'Efficient' is either incorrect or simply meaningless. Do you think that reducing NP to P means eliminating the possibility of prohibitive computation costs? I have some very bad news for you if you do.