Peterson argues that the idea that anything published ought to be understood as intended for public distribution is an antiquated understanding from the era when publishing was expensive and required a lot of effort. The opposite is true today, it's free and easy to publish - so information at different levels of appropriateness for public eyes is being published. Why not support that?
Instead of what Facebook is doing, Peterson says that a more appropriate understanding of privacy today is based on context. We expect our communication to go on in an appropriate context (no drinking in church or praying in the bar) and we expect to understand how our communication will be distributed.
If a college friend took photos of you drinking in a bar and showed them off to people in church, you might feel your privacy has been violated in both appropriateness and distribution. The bar is a public place, though, and not completely secret. Thus the need for a more sophisticated understanding of privacy that is more than mere secrecy.
We spend a lot of time discussing patron privacy in the library profession. But, like Facebook, we presume that privacy is the same as secrecy. Privacy has the additional component of respect. Patrons discuss their problems with us at the reference desk, with no closed doors to insure secrecy. Yet the expectation of privacy remains. When people use a library, they have an underlying expectation that everyone, not only the library staff, will respect their privacy. This means it is not enough for us to keep records secret. We must also remind people to be considerate and model that behavior ourselves.