I think the first bit, that is also always dropped, is also important: "We should forget about small efficiencies". Removed from that context, it has morphed into an excuse by some people to not pay attention to efficiency at all, even when the potential gains are large for a relatively small amount of work.