Denis Koroskin wrote: > Safe as in SafeD (i.e. no memory corruption) :) Right. The problems with other definitions of safe is they are too ill-defined.