Independent Reference Group Meeting Minutes March 2014

Meeting held at 46 Waring Taylor Street, Wellington

10:30am, 28 March 2014

Independent Reference Group Attendees:

Nic McCully; David Stone; Conor Roberts; Justine Cornwall and Steve O’Briem


Mark Harris

Department officials:

Peter Pilley, Mike Osmond; Caroline Sullivan and Kirsty Pleace (minutes)

1. Filter System Update

The March report uses a new way of presenting the information for the DCEFS. The statistics engine has been redesigned and can now produce more meaningful numbers. Multiple accesses within one second are now recorded as 1 access for statistical purposes. During February there were 28,000 unique requests for blocked websites. From 1 of March to 20 March there were 22,000 requests suggesting a likely slight increase by the end of the month. This is just over 1000 requests per day. In the majority of cases the access attempts were from typing in the website address rather than referral connections. It was also noted that there appears to be an increase in the inclusion of “kiwi” as a search option. This may suggest that requesters are looking for local content or an evolution of understanding of the products.

There were 4 appeals since the last report which is a significant drop in number. The Censorship Unit are investigating the possible reasons for the change.

2. System Maintenance

The DCEFS equipment has been replaced allowing the testing of the system upgrade.

3. General

Clarification on effectiveness of the filter:

The DCEFS takes out the feeder or gateway sites which creates a greater impact. The current level of refresh for the blocked website list is consistent. The Censorship Unit are developing a system to centralise the process and ensure that classification remains consistent. Website investigation is going well but the Unit are always looking for ways to improve it. The whole list is checked every month and verification is completed weekly. The upgrades are looking for evolutions in the websites. The Unit also use confiscated computers to help identify new websites through forensic analysis.

The Unit confirmed that people are suggesting websites for investigation through the ECPAT website. The Filter has achieved a lot without advertising.

International findings:

The filter literature research resulted in a number of documents on how other filters are working. The IRG felt it was beneficial and the Censorship Unit indicated it would become a regular practice to aid in keeping up with developments. Some of the information indicates an inconsistent approach internationally. The Unit are trying to improve this through sharing information with Interpol. The statistics are shared internationally and some of the information from other countries highlight issues with collation and core consistency. The new process for measuring hits can be applied over the last 2 years. The data will not necessarily predict something but it can illustrate a trend.

The IRG raised questions relating to the literature:

Q. How do we assess whether there is overblocking?

A. The filter can be honed in to one image which means that if a website has other information that is legal it is not affected in general terms. Only the blocked illegal image is not available.

Q. Has DIA considered the possibility of a wider use for the filter? Eg. pirated material?

A. No. We are contracted to maintain a very limited definition for blocking material. DIA were approached by the recording industry but the answer was no.

University research:

DIA have been approached by Otago University and Massey in Auckland with some potential topics for research.

Meeting concluded at 12:10pm

*You need Adobe Reader installed on your computer to view these files. If you are unable to open the files we recommend you get the latest version of Adobe Reader. You can download and install Acrobat Reader for free from the Adobe website.