Journeyman

• One type of test was a simple filtering out of the ACMA blacklist. Access to blacklisted sites was blocked 100% of the time. This does not mean that it was successful. The report notes that “A technically competent user could, if they wished, circumvent the filtering technology.” This is equivalent to saying “100% of people trying to access the building by the front door were stopped, but they could have gone through the back door.”
• The second type of test was to filter “additional categories of content” – including Deep Packet Inspection. Again, 100% of ACMA was blocked and non-blacklist unseemly content was blocked 78-84% of the time. In other words, at least 16% of the time the bad stuff got through. If success is measure by giving parents a measure of security, that is not a success.
• Moreover, the additional filtering overblocked, that is the filter blocked sites that did not have unseemly content, 3.4% of the time. That’s reasonably decent, technically, but small comfort to the 3.4% of online businesses who lose their customers with no idea as to why.
• Circumvention is inherent and trivial. To extend the metaphor, it’s not actually like there’s a backdoor, but rather there’s locks on the front door, but no walls to the building. “Telstra found its filtering solution was not effective in the case of non-web based protocols such as instant messaging, peer-to-peer or chat rooms. Enex confirms that this is also the case for all filters presented in the pilot. Telstra reported that heavy traffic sites could overload its trial filtering solution if included in the filtering blacklist. This is also the case for all filters presented in the pilot.”
• The performance degradation methodology is questionable. “During the pilot and testing in a live ISP environment it was not technically possible to introduce artificial loads across all the participating ISPs and the filtering technologies.” From what I read they were able to say that for one person the filtering load is negligible – but what happens when the entire pipe for everyone is filtered?
• At least the bleeding obvious is recognised: “Participant5 successfully blocked [the highest number of] circumvention attempts resulting in the highest result in the pilot of 94.5 percent. It is noted during the pilot, however, that noticeable performance degradation was observed for the filtered service which was utilising a pass-through technology.” In other words, if you want fast internet, the filter is useless – if you want a moderately effective filter you will have slow internet.

My conclusion – an internet filter is a useless, ineffective, waste of money that if it delivers any sense of security to the end users, it will deliver a false one.

It is certainly an absolute delusion on Senator Conroy’s part to state that “The report into the pilot trial of ISP-level filtering demonstrates that blocking RC-rated material can be done with 100% accuracy and negligible impact on internet speed.” The bottom line of the report, really is “When the content was easily blockable, we blocked it easily.”

And that’s for a pretty tight definition of “easily” – as @NewtonMark points out in a tweet – “You don’t need a VPN to bypass. Most of the tested products fail if you put a ? in the URL. #nocleanfeed”

This trial has done nothing to answer the questions that should be answered before more millions of dollars are committed.