The spinning thing is tightening, but in reality plagirism will always exist in one form or another. And obfuscating your sources will always be high on the plagirist agenda.
Given the "one point whatever-it-is-now trillion" URL's on the internet and millions of new ones created every day - coupled with Googles finite resources - we can be pretty sure they do not (and in all likelihood never will in the future) micro analyse every one of these URL's in the way that some paranoid people suggest they can.
They are getting better at the macro approach for sure. They are studying metrics such as sentence length, noun/verb relationships, stop word frequency, average word length. and more.
But all of this stuff is statistics based on counting some pretty generic key performance indicators. The idea of haviong the kind of resource to micro analyse several trillion URLS - their linking relationship with each other (imagine THAT diagram!!) the dating of content to ascertain "who was first", the logical reduction of every article to in a sense "read it with human eyes" and test logic...
Well, that's not going to happen.
Not saying they couldn't do it at some point, but despite their protestations to the contrary Google are NOT in the business of giving people a "good experience on the internet" - they are in the business of "making money".
As soon as the cost of micro analysing every URL on the net outweights the profitability of a marginally better index order - you can bet your sweet ass they will stop. So would you or I if we were in their shoes.
They are a business not a public library or the internet police.
Scritty
