A different view of software complexity based on long-range correlatio
ns is presented in this paper. Three different methods are proposed fi
rst and then used in the analysis of C ++, FORTRAN, and Pascal program
s. All methods being used indicate the presence of long-range correlat
ions in programs, but every method gives different results. The presen
ce of long-range correlations denotes that we can regard the complexit
y of computer programs from a different perspective and use it as an a
dvanced software metric.