This was used in old times to distinguish between triplets of
shapefiles. Now we have the philosophy of creating a dedicated
directory for a triplet. So the script `distribution_function.py` just
has to be run in the triplet directory.