Ofcom report shows consumers don’t want filters

The ISPs told Ofcom how many customers have been offered the filters, and how many have set them up. The percentage of new customers who set up the filters are a small fraction:

• Virgin Media 4%
• BT 5%
• Sky 8%
• TalkTalk 36%

TalkTalk’s 36% take up seems somewhat surprising. TalkTalk is the company that has been providing filters for longest, since 2011, so it could provide a glimpse of what filters will look like once the initial drive stabilises. The report shows that 64% of their customers have made an “active choice” to switch off the filter. Only 20% of their customers made the “active choice” to turn on the filters when the service was first introduced. TalkTalk now reports 36% uptake of filters, after they pre-ticked the box in the welcome screen. Some 16% seem to have simply clicked continue with defaults. Howeve we would like TalkTalk to clarify if this percentage are using filters to prevent access to content other than viruses protection. Without a breakdown between these figures the figure is meaningless. Virgin has a similar system with their ChildSafe and VirusSafe options, and should also provide this data for the avoidance of doubt. 

Given TalkTalk’s aggressive promotion of filters, these figures would hint at the maximum penetration to be expected if customers are given a choice. Even so, their defaults work against genuine user choice and should be opposed.

Because Ofcom didn’t ask the companies to report properly on the demographics of the take up, we don’t know whether customers taking up the filters had households with children. Ofcom needs to get back to the companies and give the public more detailed information once the uptake of filters stabilises.

The figures in the report suggest that marketing and interface design may make some difference, but there’s no evidence that filters are a compelling issue for consumers. So perhaps ISPs should spend less money promoting filters in order to keep No 10 happy, also saving as less hardware is being deployed, and offer instead a cheaper service. This could increase internet access among the poorest households, a more laudable policy goal in an information economy.

Ofcom explains away the low adoption rates with all sort of excuses. The report claims that much higher numbers of existing customers are taking up the filters after installation, but doesn’t provide any proper figures. It is to be expected that larger numbers of existing customers, in comparison to new customers, will take up filters. Simply because overall there are lots more existing customers. What we need to know is the current proportion of filtered internet connections of all customers – new and old – that have been offered filters, and what are their demographics.

Perhaps the most telling aspect of this issue is that the extremely low uptake figures do not appear anywhere in the three-page Executive Summary of Ofcom’s report.

Maybe the newly formed Department of Dirty has had a word – or perhaps Ofcom’s officials think it is a career limiting move to draw attention to the public’s lack of interest in filtering, a policy that David Cameron has been personally driving forward.

According to their own website, Ofcom’s principal duty is to further the interests of citizens and of consumers . It is hard to see how this squares up with the tacit support – evidenced in this biased report – for one sided government policies.

The organisations promoting internet filters are ideologically motivated and will not give up in the face of such disappointing responses to voluntary choices. Next they will be asking for more draconian measures. But these extremely low rates clearly reflect social attitudes against information controls. They should make the coalition government consider what kind of democratic mandate they truly have to promote mass filtering.

If you are worried that a website may be filtered by ISPs you can check it with our website www.blocked.org.uk —or join us to campaign for an open Internet, free from censorship and arbitrary blocking.