> I'd like to add to this that on a semantic / natural language level, > waterway=riverbank (deliberately ignoring long standing, widespread use and > acceptance) would seem to indicate a riverbank, i.e. the bank of a river, or > in other words, the area along a river, which will occassionally but not > always be flooded. > > I am at this point not proposing to remap all of these (unless there would > be compelling agreement by many mappers), as this is a longstanding, very > widespread tag (293.000 occurences, 31.700 of them relations) with > supposedly uniform usage, but it should be noted that there are issues with > it on a logical level.
My main point is that existing tagging (especially widely used one) should not be changed unless it gives some ontological benefit (new features/properties being added, features split etc.). I think that data consumers want stability more than they want tag names to be defined with "more correct" words (especially then lots of consumers are not from English speaking countries). This is why water proposal was wrong in the first place. I agree that it somehow slipped (I accept my guilt as well because I didn't do anything at the time of proposal). But if after five years mappers did not accept it by mapping, maybe it is not too late to revert the proposal (not tags by automated edits). -- Tomas _______________________________________________ talk mailing list talk@openstreetmap.org https://lists.openstreetmap.org/listinfo/talk