That was excellent, thanks for the link. Despite the fact that I've read all of Aaronson's blog posts (and that I got a CS degree before becoming a code monkey), I learned several interesting tidbits. One example: MoR!Harry's trick is capable of solving all problems in PSPACE. (This was proven in 2009.)
Scott Aaronson has published a preliminary version of his long essay titled 'Why Philosophers Should Care About Computational Complexity'. His announcement blog post has some interesting comments, and he welcomes suggestions there. I am not sure I like the organization of the paper. (I know most of the CS stuff discussed, so it is hard for me to decide how readable it is for people who don't.) But it is full of interesting ideas, and some of these are new even for those of us who follow Scott's writings.