With the exception of either investing time into targeted request black-holing or disallowing apps from contacting the outside world entirely (which would produ

With the exception of either investing time into targeted request black-holing or disallowing apps from contacting the outside world entirely (which would produce a pretty crappy AirBnB experience) I don't think this is really reasonable. Most useful apps have some legitimate reasons to talk to servers so blocking access before the app can get it (similar to how browsers block access before the site can get it) seems like the most reasonable approach. And websites don't need to inherently communicate with a remote server - there are a bunch of web tools out there that download a bunch of JS and then essentially run in local mode without ever sending that data home... yes the original stuff is coming from a foreign source but that's the same as Apps - the acquisition method is just different (and a lot more prone to abuse I'll grant you). However, pretty much every useful app you're using is calling home for some moderately legitimate reason - so I don't think it's helpful to differentiate the two classes of executables based on remote asset usage.