Michael Lugo (author of the original post that Mark was responding to) here. I had written $1/2 (1-e^{-2})$ in my post, meaning the 0.4323 constant, but with the usual rules of precedence you do in fact get 0.5782; he just propagated my error.