Does Review Make Implementing Filter Impossible?

For some time I have believed that a possible function of the so-called National Classification Scheme Review (NCSR) could be to position the government for a soft-landing, get-out-of-jail-free card in relation to their mandatory internet filtering plan. I most recently discussed my feelings on this topic in this article from last week.

Reading some more about the NCSR over the weekend, I found myself asking if the terms of the review, actually words away the possibility of the filter ever being implemented.

Amongst the clauses for the expected framework for the review, here is the one that I found interesting, stating that the resulting framework must be:

“enforceable and [promote] public trust in the regulatory system.”

For me, the interesting word is “enforceable”, and it makes me wonder how exactly they intend to police this particular aspect of the plan.

Filter opponents know only too well that systems such as The Onion Router (TOR) and Virtual Private Networks (VPNs) would make the filter instantly and trivially bypassable, so to deliver on the “enforcable” requirement of the reviewed classification system, they are going to hit a brick wall unless they ban all possible forms of filter avoidance.

There are of course many other methods by which one might avoid the proposed filter.

Banning any of them immediately delivers a massive scope creep on the mandatory filtering plan. Remember – the government insists that all they would do is block access to a defined list of URLs – not blocking particular service types.

Given that such methods employ encryption to hide the content of their associated network traffic, it would be nigh on impossible to block them.

Traffic flows via TOR might be observable by tracking unusual flows of traffic to and from individual hosts, but this does not categorically prove that the traffic is from TOR, or that the encrypted content within can be deemed “inappropriate” or not.

Detecting encrypted VPN would traffic might be achieved similarly, but again does not provide any kind of indication of what the traffic is. Banning VPNs would anger businesses legitimately using the technology for mobile and remote users, and would therefore be a very unpopular response.

So it gets down to this “enforcable” thing – if they can’t prove that traffic they are suspicious of contains what they might deem as “inappropriate material”, how do they on the internet, enforce their classification system that must be by their own definition, enforceable?

Do they get a warrant to search every computer on the internet in Australia, that they detect with an “unusual” traffic flow that might be TOR or VPN traffic, and which might be “inappropriate”?

I don’t think so.

Even if they do finally manage to put the filter in place, their own goal of having a classification system across “a diverse range of forms of information and entertainment content across media platforms” – which includes the internet – then they’ll have been against their own terms of reference, even before they started the review.

Let alone after it.