> we also do poorly at guessing the direction or shape of progress I think this is especially true over longer time frames, because in our hubris, we extrapola

> we also do poorly at guessing the direction or shape of progress I think this is especially true over longer time frames, because in our hubris, we extrapolate from what we know today, and cannot (by definition) extrapolate from things we haven't discovered yet. In addition, we tend to filter what we know by what is important and what is not important, and we are not good at understanding that sometimes, a breakthrough doesn't just introduce something new, it changes the importance of something else.