Agreed. They should probably have held off till they can demonstrate the set of gates required for universal quantum computation. These are well known and have been the target of experimentalists for decades now. In fact, what has held back a quantum computer from being made thus far is that no single hardware implementation has achieved the required accuracy across the full set of gates. If single qubit operations were all that was needed we'd be done by now :)

The tone of the article also seems to imply that the quantum information community is remiss for assuming that P != NP. Everyone is aware that this is a open question: it's probably the largest open question in all mathematics. So if this new approach does yield a way to find new efficient solutions to NP hard problems, that would be amazing and prize worthy.