Points (usually small flats) are used to keep the forces to a smaller area than a large area attachment. The problem is that you don't want to apply any side stresses from the mount to the glass as you don't want to stress the glass from the backside. It doesn't take much stress on glass to force it out of the correct shape. For example, a 1:20 piece of glass can't be stood on end without the glass changing it's shape enough to fail the 1/4 wave spec on a surface. You need to support the glass in a special way. People have even noticed errors in the shape of the glass when the support isn't in the right position on the side of the mirror!
As to flatness of the back, it's nice to have it fairly flat so that when the mirror shifts in the support mounting, the pointing of the scope doesn't change. Other than that, it really doesn't matter much. A bit of work has been done on surface finish but it really ended up not making much of a difference in the thermal response time.