First, because it is a beautiful subject. Second, because it seemed at the time (early 1980s) like a good warm-up to higher dimensional curvature flows.
UPDATE For a non-mathematical audience, see the results and references on the Carpenter's Rule Problem.