Separation of features by their importance is a core technique in cartography.
But what happens if features of the same feature class (e.g. train stations)
have a varying importance? A manual cartographer can choose the important features
based on other knowledge. Howerver, rule based cartography which is dominating in
the OpenStreetMap environment cannot work this way.
A computer programme rendering a map needs selection rules. Someone has to
implement them.

This talk presents some examples showing how to add lacking relevance information
if the importance of features within one feature class ranges span between extreme
values. It will show how train stations and airports can ranked using other OSM data
and external datasets.