Without comprehensively putting extreme restrictions on what people can put online, there is only ever going to be an exponentially growing trove of data to mine that reveals implications far removed from whatever whoever made it available intended to communicate.
While it may be possible to significantly cut back the amount of surveillance done on communications intended to be private, I think people need to accept that the automation of surveillance and of inferring likely connections from it are developments which irreversibly change how society functions going forward.
What's more, technology has a very persistent tendency to become accessible to more people over time. I have yet to see any reasons this wouldn't be true specifically in the case of automated surveillance. Remember the story about someone trying facial recognition to tie pictures of strangers on the subway to VKontakte profiles? That's just a small taste of the implications.
I think it's realistically possible to push back a large portion of the surveillance that invades private communications, likely by technological means more often than legal ones. And I think the article highlights that possibility, especially at a time when distrust in government is reaching a peak in America and elsewhere. That said, I'm waiting for the Pandora's box of wide access to powerful surveillance tools using publicly available information to be seen for what it is.