voxpellikbs: if they're using the HTML5 History API rather than hashes for the paths, then the server could render a scrapable version, but then you also duplicate all of the logic in two places almost nullifying the benefits of a client side framework
