There are a lot of issues related to ‘wiretapping the Internet.’ A post from Privacy International, from 2012, nicely details the amount of metadata and data fields linked with just a Facebook message and the challenges in ‘just’ picking out certain fields from large lists.
As the organization notes:
Fundamentally, the whole of the request to the Facebook page must be read, at which point the type of message is known, and only then can the technology pretend it didn’t see the earlier parts. Whether this information is kept is often dismissed as “technical detail”, but in fact it is the fundamental point.
We should be vary of government harvesting large amounts of data and then promising to dispose of it; while such actions could be performed, initially, once the data is potentially accessible the laws to legitimize its capture, retention, storage, and processing will almost certainly follow.