<code>string</code> takes the place of <code>P</code>.
And since <code>string</code> is identical to <code>string</code>,
this unification step succeeds as well.
Unification of the LHS and RHS of the equation is now finished.
Type inference succeeds because there is only one type equation,
no unification step failed, and the map is fully populated.
</p>

<p>
Unification uses a combination of <i>exact</i> and <i>loose</i>