Saturday, 4 July 2015

15 minutes of fame is not enough...

I was on Sky News again this morning - I am getting quite good at getting to Millbank studio and back reasonably efficiently now - leaving Bracknell at 9:02 by train and getting back in Bracknell by taxi at 11:59 for my grandson's birthday party.

A woman from IWF was meant to be debating with me, but she insisted on doing her bit first rather than a discussion/debate. I think I managed to get some of the key points over, and as usual I was not going up against the IWF and I agreed with a lot of what she was saying.

The problem is really that you cannot properly discuss and debate these issues in a 5 minute slot - and to be fair a "15 minutes of fame" slot would probably not be a lot more help.

The IWF (Internet Watch Foundation) work to remove child abuse images from the Internet. Much of this is acting on reports and tracking down the hosting company to get the content removed. They are rightly proud at how well this works in the UK, but when the material is hosted outside the UK they have to work with other organisations in other countries. Sadly they are not as efficient or as effective. Obviously I agree that such material should be tracked down and removed (though I don't agree on illegality of cartoons which has led to some silly cases, but that is another debate and I may yet be convinced otherwise).

The point where it gets contentious is where the IWF have a block list and ISPs use this (pay for it) to block access to some web site URLs (specific image files on web sites). It is not that contentious really with the IWF - they have pretty much always said that this block list is to stop people accidentally accessing material which would be illegal to possess. They don't claim that this block list is a tool to stop child abuse or creation or distribution of child abuse images. It stands to reason that blocking one specific unencrypted protocol for access to material which is illegal is not going to help much. Anyone wanting to access such material can no doubt find it by many other means and protocols over the Internet, many of which cannot be tracked or effectively specifically blocked. If the process of removing content is fast and effective then the block list is pointless even to stop accidental access.

The woman from the IWF did not raise the blocking issue specifically, but the interviewer did ask what ISPs can do, and whether we should be doing more. This is where the debate really starts to unravel and go in slightly different directions.

My view is that ISPs as communications providers should not have to concern themselves with what is communicated. This is a long standing principle (mere conduit) where we are not liable for what is communicated. It is not just some handy cop-out but a key factor in ensuring we even have an Internet. If there was liability for content, or a requirement to monitor police communications, it is hard to see how even a phone network, let alone the Internet, would have been commercial viable.

The problem with this view is that it sounds uncaring, understandably. But I also feel that any attempt to force monitoring, policing and filtering on ISPs is the thin end of the wedge. It opens up the possibility of extending beyond the original remit, and we have already seen this happen where copyright related court orders have been effective on ISPs that have the IWF blocking in place. Ultimately this has also led to the filtering and blocking offered by so many ISPs now at the request of the government. If there are not a ISPs like us standing up for the right to work as mere conduit we will wake up and find ourselves in a police state with approved media and content only.

However, there is another argument against filtering this sort of content - and I made that point in the interview - that the technology is changing and encryption is becoming the normal way we communicate on the Internet. This means that it is difficult or impossible to tell what someone is doing when they communicate. Identifying the "server" is not good enough, you need to tell a lot more about what is communicated (e.g. specific web pages on web sites) otherwise you cause all sorts of collateral damage. This is even more the case as servers end up behind NAT and mapping gateways and people make use of content delivery networks, and so on.

This leads the the counter argument (which is where we simply don't have time on such a short slot on TV) that we need to be able to see through encryption. This is where we start on the current government madness of trying to ban [strong] encryption. Well, I have lots to say on that, and much I have already said, but that is for another time.

Maybe some day I'll get a chance to be in a longer debate on the matter.


  1. You are right about waking up in a police state. On a totally different subject this government, for the first time moved from everything is authorised unless banned to everything is illegal unless authorised paradigm with the law regarding "potentially psychoactive" substances. While I'm not denying an issue with so called legal highs, the new law will ban everything by default (including traditional Chinese and Ayurveda medicines) and it madness even bans studying to prove that a substance is or isn't psycho active. So everything is banned and we can't prove it isn't dangerous and been authorised.
    I suspect Theresa May to entertain the same idea for the internet.

  2. As someone who works in the content filtering industry (we do filtering/auditing systems for schools, which I think is completely different to the prospect of filtering everyone's residential connection by default), I'm afraid I regard the IWF as a bunch of crooks.

    The IWF claim to be funded by donations, but this is a lie - in order to access their block lists you _have_ to make a "donation" of several tens of thousands of pounds. Needless to say, that's out of the reach of small companies. If they were actually interested in stopping people accessing illegal material, they wouldn't be preventing companies from using the IWF block list by forcing everyone to "donate" such a huge amount. I understand the IWF block lists run to a couple of hundred URIs, whereas we buy in third party categorisation databases of a few million URIs (in addition to our own content analysis systems) and it costs us an order of magnitude less than the IWF blocklist would.

  3. As said those who actively want to view CP will do so regardless of block lists ect so they are quite pointless really, what will camerloon want to do next inspect all snail mail , just in case it contains CP or is a terrorist plot,?

  4. Have you put the footage up on youtube yet?