About "On certain incomplete statistics" by Lima et al

Lima et al recently claim that ({\em Chaos, Solitons $&$ Fractals,} 2004;19:1005)the entropy for the incomplete statistics based on the normalization$\sum_ip_i^q=1$ should be $S=-\sum_ip_i^{2q-1}\ln_qp_i$ instead of$S=-\sum_ip_i^{q}\ln_qp_i$ initially proposed by Wang. We indicate here that thisconclusion is a result of erroneous use of temperature definition for the incompletestatistics.