Interesting as I've seen a few search engine start ups that seem to scrape other search engine results, depending on your definition of scraping. My definition would be a user agent that doesn't uniquely identify itself that isn't using an authorised API.