Here, I think, are some of the reasons: He writes well, far beyond the extent that good writing can be expected from a Harvard student. His writing is at once

Here, I think, are some of the reasons: He writes well, far beyond the extent that good writing can be expected from a Harvard student. His writing is at once concise, interesting (at least to this audience), and convincing. Y Combinator seems pretty successful, and PG is good at it. (I'm not going to give reasons for the success of YC here.) His books On Lisp and ANSI Common Lisp are arguably among the best books about Lisp. He is making his own dialect of Lisp (see http://arclanguage.org/ http://arclanguage.org/). I've looked it over, and it looks like a good language. Probably not all the hype cracks it up to be, but then, nothing ever is. Sure, it has its bugs and a few other problems, but it shows a lot of potential. There are probably others, but this is all I'm going to write right now.