Interview with Sarah Robertson, director of communications for the Internet Watch Foundation

Wednesday 08 April 2009   |  5 Comments  |  

The following is the full transcript from Broadband Genie editor Chris Marling's interview with Sarah Robertson (pictured), director of communications for the Internet Watch Foundation (IWF). We spoke in depth about all of the issues brought up in comments by our site visitors the last time we ran a story involving the organisation. This included its role and remit, accusations of secrecy and censorship, the governments 100 per cent take-up target for its blocking initiative, how the blocking initiative works, and how effective it is. There is also a full explanation of the unfortunate Wikipedia incident that generated so much bad press for the IWF in December. 

What is the IWF?

"The IWF is an organisation with charitable status, founded in 1996 by the internet industry, which wanted to do its best to address the fact it could inadvertently be hosting illegal content. It was founded to be a free reporting mechanism for the public and a national 'notice and take down' service for host ISPs - and to do that freely, regardless of membership. It is a self-regulatory body founded and funded by the industry and led by the industry itself, under the direction of an independent board.

"One of the founders of the IWF was Peter Dawe, Cambridge internet entrepreneur and founder of Pipex. He had a building just outside of Cambridge, which is still our home.

"We have an independent chair and board which includes three industry members: each of the 90 member companies elect a representative to a funding council, which in turn directs its policies and work programmes and has overall responsibility for issues such as ensuring budget. That funding council elects three of its members to also stand on the board.

"The IWF's charitable status ensures a more robust governance structure, as we have to adhere to both company and charity law: we have to be accountable and transparent in how we behave and report our activities. We're wholly independent of government, although benefit from it supporting the UK self-regulatory approach: certainly, when it comes to the blocking initiative, that independence is crucial.

"An independent board is no small thing; there is a board of 10 trustees, three board committees and a board sub committee. Because we're dealing with illegal images, we have a memorandum of understanding with the Crown Prosecution Service and the Association of Chief Police Officers (ACPO), giving us special dispensation in the Sexual Offences Act (which amended the Child Protection Act) so we can do our work. We're not a legal body, as such, but we have the authority to do what we do and couldn't do this without support of key partners such as the Child Exploitation and Online Protection Centre (CEOP) and the government."

What is your stance on self-regulation over legislation, and the government's target of getting 100 per cent of UK ISPs signed up to the IWF and using your blocking list?

"Self-regulation can be far more responsive than legislation, especially in an arena such as the internet which is changing all the time. If we had to rely on policy being developed, it would be out of date by the time it was passed.

"Our key industry stakeholders think robust and inclusive consultation procedures mean you can respond immediately. We think the fact nearly 100 per cent of residential broadband connections in the UK are covered, in addition to some business providers and all mobile operators, as well as filtering companies, is extremely impressive. All of this has been achieved on a voluntary basis as a result of the industry’s own commitment to good practice.

"We’re not in the business of naming and shaming, but would certainly encourage ISPs to take our list. However, on a wider scale, it's important they are part of the self-regulatory approach to illegal content through association with the IWF. They can then contribute to showing the way the industry can be responsible for itself, so avoiding potentially more restrictive legislation by finding solutions itself.

"While some (such as the NSPCC) are calling for 100 per cent take-up of the scheme from ISPs, and the government has said it aims for the same 100 per cent figure, we're just proud of our achievements to date. And we're happy to work with the government in trying to work through any fears companies may have in joining, whether they be technical or philosophical.

"The fact is, if this was television, there would not even be a debate, but the internet’s unrestricted and global nature means certain groups and individuals are very protective of its freedom and very concerned about attempts to curtail that freedom. These principles are really important but have to be balanced by the fact that internet services are abused for the distribution of child sexual abuse images. Believe it or not, the IWF considers itself anti-censorship and we certainly wouldn’t want our efforts in disrupting the activities of a criminal few to damage the freedoms of the vast majority. Those advocates of freedoms of expression on the internet are key stakeholders for us and we’ve always looked to engage with them."

How do you respond to accusations the IWF is a secretive organisation?

"We have one of the most comprehensive websites imaginable. We publish annual reports, details of board members and senior staff, board minutes, full lists of members (and who takes the list), full funding details, a monthly e-newsletter and we respond to external enquiries. The whole blocking initiative is described and there’s a huge FAQ section, as we recognise the importance of explaining our role in providing this list, plus a complaints procedure – anything you could want to know about blocking has always been available. Full contact details are also online, including our address. In fact our website is arguably too comprehensive, as information can be hard to come by; a fact we've recognised and why we are in the process or redesigning our site.

"The IWF is subject to regular inspection by independent forensic and technical experts, as well as HR professionals and police officers. And the block list isn’t secret: of course it cannot be published, and the public can’t comment on it, but every URL on it is shared with law enforcement and with hotlines around the world and the process of adding URLs to it is is independently inspected. Misuse of our trusted role in providing a list would quickly come to light.

"While we’ve never been secretive about our initiative, unfortunately probably not that many people knew about it: people would probably have been quite reassured if they had. Certainly our member companies try to publicise it as one of the services they offer.

"But suddenly a lot of people became aware of us due to a censorship debate over the legitimacy of an image [see 'The Wiki Incident' below] which for us was a shame, because bringing down Wikipedia, or picking on record covers, isn’t what we do; what we do is deal with illegal abuse images.

"The website redesign is happening because, at the time of the Wiki incident, valid commentators were posing questions that were in fact already extremely comprehensively answered on the IWF’s website. It was frustrating: in providing too much information people couldn't find the right information, which led to charges of secrecy."

Can you explain the IWF's actual role and remit?

"The list of URLs we provide to our members is just part of our service; it is not part of our core role or remit. We were founded to be a free public report mechanism and a free take-down service for UK ISPs: that is what we do. We also work internationally to share intelligence on child sexual abuse websites and share data with police and other relevant authorities. Regarding our remit, we were formed to deal with child sexual abuse content and criminally obscene content, such as illegal adult pornography. In 2004 incitement to racial hatred content was added. It has not changed since.

"I have read plenty of rhetoric about what we may or may not be doing, but the remit hasn’t crept on behind people’s backs. If and when we are invited to take on any new kind of content – which could happen, as there are lots of types of content worrying the government and the public – we have a very strong consultation process: any change to our remit would see a robust consultation with our board and with our member companies and it would only be taken on if they wanted us to do so. And any change would be made very clear and be well publicised.

"In fact the IWF website’s reporting mechanism [for reporting content] only lets the public report child abuse, obscene, or incitement to racial hatred sites: there is no other button. Anything else should be pointed to sites such as e-victims.org or CEOP.

"Also, crucially, we don’t deal with people, just content. We don’t investigate a person’s crimes, we just try to minimise the availability of illegal online content. It’s quite a specific role. However, if despite the design of the reporting form, someone reports that a child is in danger, for example, or that someone is being groomed, or that they think someone has downloaded illegal content, we will pass that information on to the relevant police agency.

"We welcome many kinds of new members. It may not be the list they benefit from; they are not forced to take it, although most providers do. They may be more interested in the CSR [corporate social responsibility] benefits, or the fact they want to support the national notice and take-down body. But companies don’t have to join us: there is no levy system as there is with other self-regulatory bodies.

"The IWF is also recommended by bodies such as ISPA [the Internet Service Providers Association] and three quarters of its funding comes from the industry [a quarter is provided by the EU’s Safer Internet initiative]. Whilst we could not operate without support from the government, it is important we work in partnership with the people involved in the industry, who know better than anyone how it works."

How do you respond to claims of censorship from the IWF?

"The blocking initiative was industry led. In 2004, our members asked whether we could provide a list so they could offer an extra level of protection to consumers.

"The list was never designed to try to stop offenders having access to this content: we understand those people are often technologically sophisticated and easily able to circumvent network level blocking. The industry simply wanted to try and do something else to protect its customers and it was only ever seen as part of the solution, alongside all of the other strands of our work designed to combat this content.

"It was then designed and developed to be as simple as possible, while being comprehensive. It is URL specific and designed hand-in-hand with the industry. A lot of members now have different post-processing systems, so the list is simply URLs – ISPs then make it work on their particular platforms. Since 2004, many more ISPs now offer this added protection to their customers. In addition, all the mobile operators take it, search providers and filtering companies take it: it is fantastic evidence of a voluntary initiative where an industry is doing its best to demonstrate good practice.

"There are many debates going on in parliament, the media and society in general around harmful and inappropriate internet content: we have no role in that. Who can define what is harmful and inappropriate? For example, there are widely differing views on pornography. We try to remain neutral: we are dealing with widely diverse stakeholders with quite different views. Certainly we are happy to share our experience where useful or relevant.

"ISPs block malware, spyware, viruses, and no one seems to mind. The charge that we’re in some way the only people blocking content is unfounded. ISPs have the right to block whatever they choose, possibly above and beyond the list we provide, under their own terms and conditions.

"Will our remit grow? I don’t know – that would be a matter for our board and member companies. But I think we are successful because our role is quite specific: it concentrates on three very specific illegal areas. We certainly have no plans to increase the block list to other forms of content."

More specifically, some accuse you of being the only body of your kind outside of regimes such as China's?

"Blocking is designed to cause as little collateral damage as possible, hence it being URL specific. And, importantly, the UK approach only involves indecent images of children when it comes to blocking.

"Plenty of other countries, including many across Europe, have seen successes after seeing this voluntary approach being so successful in the UK. This is sometimes government led, or police led, and can involve DNS poisoning or other kinds of more damaging blocking: I would say the UK approach bears up pretty well in comparison.

"So, there are many other blocking initiatives and many other Hotlines – we work with more than 30 around the world as part of the INHOPE [International Association of Internet Hotlines] association – all working to fight child sexual abuse websites."

What can the blocking initiative really do? Is it effective?

"The core of what we do is addressing reports from the public. If they are concerned about content they’ve stumbled across, it is important the public is able to report this somewhere. Wider society wants a safe internet they don’t mind their children using.

"The blocking initiative can disrupt the activities of those supplying and accessing images while reducing the amount of times innocent internet users are exposed. Crucially it reduces the number of times the images themselves are viewed, as this perpetuates the victim’s original exploitation.

"We get around 35,000 individual internet hotline reports per year – around a third are then confirmed as potentially illegal. Last year there were less than 3,000 child sexual abuse websites identified, a tiny proportion of the internet, but they must be tackled. Blocking won’t bring them down but it does help disrupt activities. Of course, the only lasting solution is removal at source – and we’re working at that too.

"However, that is very difficult, as the majority is hosted abroad, so we can’t affect a take-down. However, we are looking at how we could do more in that area. Most of these sites are commercial operations, selling abuse images. They have very complicated business models with lots of servers, payment mechanisms and redirectors: this complex set-up makes them quite resilient to take down. Content moves every few days to different servers in different countries.

"But this is very rarely in the UK; we have a strong reputation for being a very hostile online space for illegal content. It does still happen occasionally, and we still do have a take-down role here, but it’s minimal. The bulk of the content is hosted in the US, where the scale of the internet traffic and services and the number of providers makes it a difficult challenge for our partner bodies. Here in the UK we have great links with the industry, and the industry is smaller, so take-down is an easier process. But don’t forget that regardless of where the content is hosted, the children are being abused, the sites are being operated and the images are being accessed from all around the world.

"We are also contributing to the European Financial Coalition to help track payment mechanisms and therefore trace those running the websites."

Some claim you do not inform websites that they have been blocked. Is this the case?

"Any UK URL is immediately put through the ‘notice and take-down’ procedure: the hosting company is informed and the content is removed, in partnership with the police, which ensures any evidence is secured.

"More than three-quarters of the list is made up of commercial child abuse websites, which may well be hosted by a less than legitimate host company: the IWF would not have a role in alerting a site that was perhaps already under investigation that it knows about them. Furthermore, as no UK-based URLs are put on the list, there may well be jurisdiction issues meaning a UK body cannot approach that non-UK company. Again, this might also undermine a police investigation.

"What we do is pass on details of every URL to the equivalent hotline in that country and/or law enforcement in that country, according to the code of practice between all hotlines and the law enforcement involved. In this way, the hotline and law enforcement agencies can enforce the laws appropriate to the offences within their own boundaries."

Does this make it harder to identify real abusers? Or just drive them underground?

"This is hard to quantify and I don’t think the IWF has the answer. Certainly images are sold or exchanged in other ways, for example via newsgroups or peer-to-peer. While we work to combat such content in newsgroups, we do not have a role dealing with peer-to-peer – that is a matter for our police partner. But the commercial sites often aren’t trying to hide – they’re trying to sell images.

"Whilst there is a very small number of websites, they are very hard to combat. There are still a significant number of not-for-profit sites where paedophiles exchange images, which people won’t stumble upon. The abuser, or someone very close to them, often runs or contributes to these sites and users contribute their own sexually abusive images.

"The key is disruption at many levels (be it through take-down, domain name deregistration, blocking or bringing down payment mechanisms) and data sharing – at home and abroad. It is important to remember the internet didn’t create paedophiles, but the fantastic freedom of information the internet allows for is no reason not to challenge the distribution of sexually abusive images of children.

"What is often overlooked are the sentiments of those abused: for them, the thought of those images being viewed over and over again – often over many years, as the images filter out from private sites to commercial ones – can be almost as bad as the abuse itself."

There are sceptics who suggest self-regulation is simply a smokescreen to avoid legislation from an industry that knows what it is doing is not really effective.

"It’s not a PR stunt. Both the industry and the government want self-regulation to work, and it is working. The IWF can show that 18 per cent of child abuse content was hosted in the UK when it was first formed: that figure has been less than 1 per cent since 2003. The UK is doing a great job, and the data it is sharing all around the world is helping other countries to address the issues too.

"With the founding of coalitions such as the European Financial Coalition, law enforcement agencies – with the help of organisations such as the IWF – are getting to grips with payment mechanisms and chasing the money trail behind the commercial sites. Don’t forget, the number of child sexual abuse websites is decreasing, the situation is improving."

Any closing comments?

"One significant technical problem in five years of providing a list is quite an achievement when dealing with network level blocking, especially as there are 60 companies implementing the list in different ways across different platforms, and that list is updated twice a day.

"We have been working for over 12 years and one of our main jobs is to help foster trust and reassurance in the internet: slowing it down, compromising it or restricting it is not our business. People need to put their faith in the IWF, and we should move on from the censorship debate: I wouldn't want my internet content censored at all, but I would be quite happy for a trusted independent organisation to be able to work in partnership with the internet industry to make a difference in the supply and access of child sexual abuse images.

"The IWF is more than a list, which is a very small part of what we do: we are actually getting on with exactly what people want us to be doing – focusing on getting the content removed and informing the processes which gets those responsible investigated. It’s really not about censorship. And, if blocking is going to happen, who would you rather facilitate it? We’re not secretive and there are no hidden agendas. If you want to know more about our work visit our website (www.iwf.org.uk) and if it’s not mentioned there then we’re not doing it!"

The Wiki incident

"On Friday, December 5, a Wikipedia URL was actioned by the IWF. It was judged to be potentially illegal. The IWF isn’t the law and it doesn’t pretend to be – it acts in accordance with UK legislation and with the support of the police, but doesn’t try to pre-empt a jury by declaring conclusive illegality. What it does is recommend that certain content is blocked through the provision of a voluntary list, requested by the industry.

"It was done with the best intentions, in line with IWF procedures, according to UK legislation, and with subsequent support (through the appeals procedure) of our law enforcement partners. However, what began as a simple blocked URL escalated into much more.

"Many ISPs block by using a two-stage filter. Therefore, most UK internet traffic going to Wikipedia was going through various ISPs’ second stage filter, which meant it appeared to be from the filter’s IP address. Wikipedia has a policy of banning people it doesn't want vandalising its pages and could no longer differentiate who it needed to ban from the UK, so it simply banned all UK editors.

"Some critics focused on the fact we didn’t just block the image and that by blocking the page, legitimate text was also affected. URLs for the images are put onto the lists as well as the URLs for the page containing the image, however adding the image URL to the list doesn’t necessarily mean the embedded image won’t be visible. Also, because the majority of the content on the list is on commercial child sexual abuse sites, they use tactics such as splintering images across a number of different remote servers on different host companies in different countries: there isn't just one image sitting somewhere, so it’s not as simple as just blocking the images with most of the sites we deal with. We're not generally in the business of blocking text, and certainly not legitimate debate, but for these reasons we block the URL of the page depicting the image, therefore other content on that page will be affected.

"There was a history behind this image that meant Wikipedia felt justified in keeping it up, and certainly they refused to remove it. Of course they don't have exactly the same laws or approach in America. Incidentally, I understand that the band involved (the image was a Scorpions record cover) have since come out admitting regret at using the image, and that they were bullied into it.

"Child sex images are graded one to five according to the Sentencing Guidelines Council. They are all illegal. Level one would, for example, depict no sexual activity but erotic posing; these may go on the list as the beginning of a series of illegal images which get increasingly more serious. Level four is penetration; level five would include bestiality or sadism. Level one can be seen as the least severe, as there is no physical sexual exploitation being captured in the shot, although in most cases it is very young children, for example with their legs being held apart.

"Of course there is nothing inherently wrong with naked children or pictures of naked children. What you have to remember is that the bulk of our work focuses on commercial websites which are selling a range of indecent or abusive images under monthly subscriptions to offenders all around the world – and that puts a very different slant on whether there is an abusive aspect to an image’s distribution.

"It was a testing time for us. We managed to convene two board meetings while dealing with all the media and public attention, agreed a unanimous board decision to remove a URL from the list, and turned around an appeals process involving senior law enforcement – all in just two working days. I would say that is a very good example of a responsive organisation, acting swiftly and in the public’s interest.

"The technical issues over whether it could happen again are not just an issue for us as the IWF doesn't block the URLs, we supply a list under contract. However, I understand that a number of ISPs have now made sure Wikipedia has the IP address of the stage two filter – if they use two stage filtering – to ensure a blanket ban wouldn't happen again in the future.

"We really are working hard to avoid the technical issues with the Wikipedia URL happening again. It damages the reputation of what is a great initiative and damages our reputation in being trusted to carry out our work in a reliable way. We want to move on from the Wikipedia incident but also to do everything we can do to make sure there is even less chance of collateral damage to the internet as a result of our work in the future.

"While we didn't enjoy the criticism, we certainly welcomed the review of how we work; we're a learning organisation and are pleased this brought on a review of issues such as the transparency of landing pages for blocked URL requests; the importance of even more independent oversight of the list; as well as the issue of possible additional criteria to be considered when ‘actioning’ a URL. There are a number of other ‘actions’ that can be taken instead of, or in conjunction with, adding a URL to the list. We want to make sure we have the best approach and that our systems are suitably responsive.

"The only intent was to minimise the circulation of an indecent image, which had been considered potentially illegal under UK legislation. With hindsight, it is clear that our actions didn’t achieve our aim to minimise the availability of such content. Unfortunately, what is a very positive initiative was only seen in the context of censorship, which it shouldn’t have been. But because it was an image people considered as borderline, a few felt they could reproduce it."

5 Comments  |