The two examples below illustrate the issue
Query1:
[out:csv(“name”;false)]
[bbox:-33.9480172,151.1107534,-33.8594833,151.2207935];
area[name=“Sydney”];
way(area)[highway][name];
out;
One result (after obtaining lat-lon and distance) is:
Airport Drive, Sydney, 2000, -33.9272726, 151.1694315, 2.637km
Query 2:
[out:csv(“name”;false)]
[bbox:-33.9480172,151.1107534,-33.8594833,151.2207935];
area[name=“Mascot”];
way(area)[highway][name];
out;
One result (after obtaining lat-lon and distance) is:
Airport Drive, Mascot, 2020, -33.9272726, 151.1694315, 2.637km
The two results are essentially identical (eg lat-lons) except for the name of the suburb. The problem appears to be due to there being an Administrative Area called Sydney which embraces all suburbs. Also, within the all-embracing Sydney, there is a relatively small suburb area also with the name Sydney.
If one is looking for streets in suburbs within a largish BBox well inside the larger Sydney (which includes by design, the smaller Sydney), a large number of ‘duplicate’ (spurious) streets are generated with a suburb name of Sydney, it would seem, due to the outer Sydney boundary.
How can such a situation be avoided?