The authors argue that, at present, it may be best to pursue goals to generate tools, studies, metrics, and programs in major deficiencies, rather than call for specific targets, such as a 10% cut in yearly releases.