Can the area of graphics and visualization, particularly, that of
modeling, benefit from the qualitative representation of information.
For  example,  can   we   store   a   face   as   a   qualitative
relation/constraints between nose, mouth, chin, etc? Is this what
cartoonists do?