Well I think the Point is that a self-made solution won't be directly parse able for the $NSA. As a minimum some type of scraping Spider would be needed that in

Well I think the Point is that a self-made solution won't be directly parse able for the $NSA. As a minimum some type of scraping Spider would be needed that indexes the content. Then it would need classification. If everybody were to do this then the job of automatically indexing social relationships wold be much harder.