December 19, 2011 | Kayahan Cantekin

How does mobile Internet filtering work?

All the major UK mobile operators have Internet filtering schemes that block certain content from users. These filters are designed to protect children from accessing adult material. The filters are turned on by default when anybody signs up to a mobile contract. Age verification, normally via a credit card, is required to turn them off. We've heard a lot of anecdotal evidence of mistakes, over-blocking and the difficulty of pointing out when things go wrong (for example, see James Firth's blog on Vodafone's blocking of underwear sites).

Mobile Internet access is becoming more important as a means of getting online. According to Ofcom, 28% of UK adults said they accessed the internet on their mobile in the first three months of 2011, and mobile data use increased forty-fold between 2007 and 2010. We've started to look more closely at how this blocking works.

It's clear that mobile operators could be much clearer about this. They tend to be pretty opaque as to exactly how their filtering works, and how they decide which Web pages are inappropriate for under 18s.

For example, Orange says that it is the Independent Mobile Classification Body (IMCB) that decides what is adult content or not (see here). However this is not true. The IMCB does provide a framework for determining content that is inappropriate for children and teenagers. But content from the Internet is out of IMCB’s remit, as stated in its Classification Framework (see here). We've been in touch with IMCB about this, and are awaiting a reply.

Mobile operators all declare that they are acting according to a ‘code of conduct’ set by the Mobile Broadband Group. But this code does not provide for any kind of criteria for determining or defining blockable content. It simply points at the IMCB framework.

It is most likely that security contractors and their algorithms employed by mobile operators, such as US companies like Blue Coat, decide what we are able to access. How the policies of these companies fit with the frameworks of the IMCB and the Mobile Broadband Group is another question we are looking to answer. 

As well as a lack of transparency, overblocking and clumsy customer support, there is also plenty of controversy surrounding the methods these security companies use, for example about how much user data they keep hold of and the consent issues raised by the retention of this data. There are also questions about some of these companies seemingly selling their filtering technology to oppressive regimes such as Syria and Burma.

Transparency regarding how mobile operators decide what counts as 'blockable' content is increasingly important. Customers should be able to ascertain how and why content is blocked, and have easier ways to point out when things are going wrong. We'll be developing more work on this, including tools to help you point out when mobile operators are blocking sites, soon. Please let us know if you're interested in helping out.