The results of ISP-level content filtering tests released today by the federal government have revealed that the products tested could filter websites with illegal content or block entire peer-to-peer networks such as BitTorrent, but could not identify illegal content shared on peer-to-peer (P2P) networks.
The report, released today by the Department of Broadband, Communications and the Digital Economy, showed ISP filtering technologies were improving, however it also highlighted significant holes in current technologies to automatically filter content shared over peer-to-peer networks.
While all six tested products, which were not named, were able to block entire networks of non-Web protocol traffic, such as instant messaging and peer-to-peer networks, none could identify illegal or inappropriate content over those networks.
However, the report showed that new filtering technologies imposed far less network degradation when turned on than under previous tests using older technology.
Under previous tests, network performance degradation was no less than 75 per cent, while this round of tests ranged between two per cent for the best product and 87 per cent for the worst. Products also performed better in accurately blocking blacklisted content.
Current technologies can block entire P2P networks, but not specific content. (Credit: ACMA)
"It is very encouraging to see that the industry has made significant progress with ISP filtering products and we are heartened that many of the products tested are commercially available, with many of them already deployed overseas," Communications Minister Senator Conroy said.
"The next step is to test filter technologies in a real world environment with a number of ISPs and internet users," Senator Conroy said.
A spokesperson for Conroy's office said the department expected a live test to begin before the end of the year. The government will release an expression of interest to ISPs for the tests.
A range of filtering techniques were tested at Telstra's Broadband eLab by testing company Enex Testlabs, which compared Domain Name Service poisoning, packet filtering and analysis-based filtering.
The trials are aimed at determining whether broad-scale ISP level filtering would be feasible for the purpose of boosting online safety laws.