3 ms·
That's insufficient. If you call enclosed areas (including the one 'at the putside') faces, you sometimes can draw an edge A-B either through face F or through
by Someone 13y ago
That's insufficient. If you call enclosed areas (including the one 'at the putside') faces, you sometimes can draw an edge A-B either through face F or through face G.
For example, in http://en.wikipedia.org/wiki/File:Brussel_Sprouts_Game.png http://en.wikipedia.org/wiki/File:Brussel_Sprouts_Game.png, position 3, you can draw a 'diagonal' vertically through the large enclosed area, but you can also draw a graph edge between the same two vertices 'around the outside'.
Similarly, you can have disconnected parts of the graph, where it is essential to know in what 'face' they lie. For example, starting with 3 points A, B, and C, draw a circular edge from A to itself. There are two ways to do that, one where B and C end up in the same face and one where they do not.
If you store the graph as a pair (V,E), you lose that information.