Their own index? What does that mean? Is the idea that they're trying to figure out if search engine X is actually forwarding requests to search engine Y? Usi

Their own index? What does that mean? Is the idea that they're trying to figure out if search engine X is actually forwarding requests to search engine Y? Using the user-agent string is the naive, and probably worst, way to figure this out. But of course, it's the easiest conceptually. What they should be doing if they want to see if the indexes are unique is to query the index itself ie: send in query strings and see what comes out, then compare them. But then they'd have to figure out a way to compare the returned results and determine if they're "different enough." Using their methodology all they have to do is count.