[Cryptography] connectivity without complete loss of privacy

[John Denker]

On 04/04/2016 06:15 PM, dan at geer.org wrote:

> Anyone who voluntarily uses a device whose inherent function requires
> continuous connectivity has no, repeat no, reasonable expectation
> of not being tracked.

That seems like it ought to be a fixable problem.

A lot of people seem to think that tor works, or could be made
to work somehow.  Let's accept that as a working hypothesis, at
least temporarily.

Then it would seem that voice over IP over tor should allow me
to move around and communicate without being tracked.

I'm not saying that's optimal, but it may serve as an existence
proof and a conversation starter.

Does that work?  Could it be made to work?  If not, why not?