After much delay the Federal Government has released the results of Enex TestLab’s test pilot into mandatory ISP-level content filtering, finding that a technically competent user could circumvent filtering technology based on ACMA’s blacklist.
According to the report, initially all filters had issues with loading the ACMA blacklist indicating a need for routine checking to ensure the blacklist is filtered correctly with each update.
On the up side, testing also revealed that ISPs filtering only the ACMA blacklist during the trial had no noticeable performance degradation that could be attributed to the filter itself.
Of the ISPs filtering only the ACMA blacklist, two ISPs used Deep Packet Inspection (DPI) pass-through monitoring technology and the third proxy filtering technology.
In addition to filtering the ACMA blacklist, six ISPs filtered other categories of content using DPI pass-through monitoring technology and pass-by hybrid technologies.
In blocking additional categories of content all six ISPs achieved 78 percent to 84 percent accuracy when assessed against the test list of URLs compiled by Enex TestLab.
“These results represent an improved level of performance when compared to previous Enex testing, and suggest commercially available filtering products are increasingly effective at including additional categories of content on their filtering lists,” the report reads. “One hundred per cent accuracy using these commercial lists is unlikely to be achieved as the content on different commercial lists varies and there is a high rate at which new content is created on the internet.
“Testing was also undertaken against a list of content, prepared by Enex, considered to be innocuous and which should not be blocked by a filter. All participants experienced some level of over-blocking in this test. All filters blocked less than 3.4 percent of such content.”
On the performance front, the report found that the majority of these filtering technologies, when correctly installed, enable the filtering of additional content with minimal or no performance impact.
“One technology, however, displayed a noticeable performance impact,” the report reads. “This finding was similar to levels recorded by Enex in previous trials.”
On circumvention, the report found that filtering of additional categories of content enabled ISPs to implement measures which made some common circumvention techniques difficult.
“As a general rule, there appears to be a relationship between measures to counter deliberate circumvention and impact on internet performance -- i.e. stronger circumvention prevention measures can result in greater degradation of internet performance,” the report reads.
End user feedback
Discussing customer feedback on the pilot, the report finds that a small number of customers indicated they had experienced some over-blocking and/or under-blocking of content during the pilot.
“These events were considered relatively minor and occurred only once or twice,” the report reads. “A small number of customers also reported slower network speeds as a result of the service which filtered additional categories of content.
“Overall the service offered by the ISPs was considered effective by customers, with around two-thirds of customers participating in the survey indicating that they would either probably or definitely continue using additional content filtering services.
Customers also expressed the view that it was important for there to be mechanisms for self-management of the filter settings and improved visibility of the filter in action.
Telstra’s Informal Trial
The report also details Telstra’s informal participation in ISP level filtering in which it undertook its own testing of ISP filtering of a blacklist of up to 10,000 URLs using a ‘domain name server plus proxy server’ filtering solution.
No customers were involved in the Telstra trial and testing was conducted using Telstra’s test environment, which is a replication of its network and used by Telstra for testing its products prior to release, according to the report.
:Telstra found that its filtering solution was 100 percent accurate at blocking a blacklist of 10,000 URLs. Telstra also found there was no discernible performance degradation,” the report reads. “Telstra did not test circumvention, because it considers that filtering can be circumvented by a technically competent user.”
According to the report, 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,” the report reads. “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.”