Component Fault Trees (CFTs) were invented in 2003 as a compositional extension to fault trees to better reflect the technical architecture of a system in its safety analysis model. Since then, a lot of research has been contributed regarding semantic extensions, evaluation techniques, and tighter linking between system and safety models. This paper addresses three main objectives. First, we summarize the most important contributions and shape a vision of better integrated system modeling and safety analysis. Second, we push forward standardization and sketch a new evaluation scheme for quantitative analysis using mdd. Lastly, an outlook on future improvement ideas is given to make CFTs a viable technique for loosely coupled systems and Cyber-Physical Systems.