Well, aside from calling my @filenames = sort readdir( DIR ) in one step (since by default sort uses effectively the comparitor you've shown, and that copies the results around one less time) what more do you think could be optimized? Doesn't get much more bare bones than that. If you're not interested in every filename that readdir is going to return you might stick a grep in between there, but there's not really much more to be pared down.