The rotation axis of the planet Uranus is tilted by 98° compared to its plane of orbit. This means that the north pole of Uranus is "under" the ecliptic compared to the north pole of other planets.
Why was the value of 98° chosen, rather than 82° ? How was it decided which pole should be north?
From the Report of the IAU/IAG Working Group on Cartographic Coordinates and Rotational Elements of the Planets and Satellites (2000), section 3:
The rotational pole of a planet or satellite which lies on the north side of the invariable plane will be called north, and northern latitudes will be designated as positive.
The planetographic longitude of the central meridian, as observed from a direction fixed with respect to an inertial system, will increase with time. The range of longitudes shall extend from 0° to 360°.
Thus, west longitudes (i.e., longitudes measured positively to the west) will be used when the rotation is prograde and east longitudes (i.e., longitudes measured positively to the east) when the rotation is retrograde.
The "invariable plane" in question is the invariable plane of the Solar System.
By contrast, the usual convention for quoting obliquity is to use the right-hand rule (fingers curl in direction of rotation, thumb points north). This is inconsistent with the definition used to determine the north pole but people seem to prefer having the rotation period always be a positive number.
