Failed or troubled modernization efforts, such as the multi-million dollar 1997-2000 ROCC/SOCC failure, are a serious acquisition problem for the Air Force. Using both historical data and a survey of current Air Force software acquisition program key staff, this research examined the Air Forces ability to modernize legacy software systems.The search of historical program data, to identify trends or similarities between known failed software modernization efforts, failed to uncover sufficient data for analysis. This lack of project data indicates a knowledge management issue (i.e. lessons learned are not recorded and stored so that they can be accessed by other programs) in the acquisition community.The Phase II survey gathered data on current software programs and addressed the recommendations of the 2000 Defense Science Board (DSB) Study on Software. The goal was to determine first, had the recommendations been implemented, second, did program characteristics effect implementation, and third, did implementing the recommendations lead to program success. The survey results indicate that most of the recommendations of the DSB are not in practice in the acquisition community. They also indicate that support programs are more likely to have implemented the recommendations than are weapons systems.