Filter put to test by Optus and iiNet

The federal government has asked a number of telcos that are currently not keen on the clean-feed internet scheme to spend the holiday season trialing the new technology.

The mandatory imposing of a nationwide internet content filter at ISP level is funded as part of the $125.8 million cyber safety plan initiative by the government, and will see the Australian Communications and Media Authority (ACMA) operating two separate blacklists to block the Web pages detailed within them.

Despite objections to the content filtering plans, Stephen Dalby, the chief regulatory officer for iiNet has said that it will be taking part in these trials. Dalby said “We are very keen to participate in it, but we do not support the notion of ISP-level filtering. The best way to get the facts is to work on the inside. If the technology delivers unacceptable false positives or slows down speeds, we will gladly publish the details.”

Despite considering it a ‘dumb idea’ Dalby did say that it was important to take part in the trials although he said that the content filtering technology would not help to reduce the amount of child porn out there.

In a bid to gauge how effective the content filters would be Maha Krishnapillai, a spokesman for Optus, said that is would also be taking part in the pilot. He said “We want customers to be very comfortable with using the Internet and we do not want to have clumsy technology in place. Optus is committed to work with government. There are real challenges about how it will be done.”

Concerns have been raised by privacy activists and communication technologists who have been angered by the scheme saying that it merely gives the government the power to potentially censor material on the Web like political protest or information regarding euthanasia and have questioned both the accuracy and effectiveness of such a filter.

Tests carried out last July with the filtering technology simulating a load of 30 users met with similar criticism. The incorrect blocking of as many as 10 000 web sites in 1 million would occur even with an accuracy rating of 94 percent, which many considered to be the most accurate filter.