@tess time running in production is (by itself) not an indicator for quality in software. It can be, but it could also be sign of a local optimum that prevented better things from taking root (switching cost too high). Or simply an area where requirements did not change or the complexity was always or later moved elsewhere. Assuming quality plays a role: which software quality attribute do you think contributed most to your work's longevity?