I would be surprised to see them suffering for optimising the algorithm to try and trace the original source of content though. With the sort of sites we're tal

I would be surprised to see them suffering for optimising the algorithm to try and trace the original source of content though. With the sort of sites we're talking about, by definition that original source will be more up-to-date content which the user could reasonably expect to be presented preferentially. At which point - you're welcome to make your business from rehosting public content that originates elsewhere, but if you expect us to primarily direct people to your copy rather than the original then I think you're onto a loser. Let's put it another way. I could conceivably build a valuable service by taking content from StackOverflow and Wikipedia (or wherever) and automatically linking the two to enable people to get some more context around some questions, maybe reformatting pages to enable side-by-side content or something similar. It wouldn't be a trivial service but it wouldn't be impossible in the least and it could plausibly add value. As such it wouldn't be unreasonable to preferentially direct users in some cases to that source rather than the original, as algorithmically optimised content - the preferential ranking would be a result of the value of the linking algorithm. Without this though, by what measure am I conceivably superior to the original source by having an out-of-date copy with fewer legitimate inbound links and more irrelevant content (adverts)? Being a monopoly isn't illegal. Using your monopoly power in a way that might harm another company isn't illegal. What's illegal is doing that unreasonably and capriciously, as Microsoft were with Netscape in that trial, or DR and Lotus in previous trials.