What would the difference in development/testing time for an aerodynamic vs a software fix?
I am assuming (with zero knowledge of development processes) it would be significantly longer and more expensive to develop and test a hardware fix than a software fix incorrectly deemed to be of lower criticality?