Isn't that contradictory? If you start with the assumption that it's evolutionary, how does it make sense to judge whether selection pressure is "too weak" and that things "should die off?" Selection pressure is what it is.
Selection pressure may or may not be in a feedback loop with the evolutionary process, but you can still view it as a separate component. In case of computing, the (broadly understood) market is the selection pressure. As for the notion of what should happen, this comes from humans who are capable of thinking about the evolutionary process and who value some goals over others. In particular, those humans tend to notice that the selection pressures in software industry do not promote good, efficient, and well thought-out solutions.
