Questions and discussion for this lecture live here. Fire away by hitting Reply below 
Why you are not considering rotations at joints in order to plot deformed shape.I think it is not strictly necessary to add more nodes to have a better representation of the deformed shape since we have information of the node rotations but we are ignoring that information, why?. Even with just 3 nodes and 2 elements we could get a better represenation, am I wrong?
You are not wrong. We could make use of the additional rotational information at each node. However, I think the most direct (and straightforward) route to producing a deflected shape diagram is by using the translational information and mapping the nodes to their new locations.