Friday 30 December 2011

How to police the web

Although the police have arrested a few people for online hate crimes related to football, these represent a tiny percentage of the incidents on the web. This has led to fans forming posses (or crowd-sourcing justice) to expose the culprits in other ways. This could increase the number of prosecutions, but has downsides.

Although this is often done in a responsible way, I think that when carried out more widely, any method that doesn’t involve the police will be dangerous. There is always the danger of vigilantes publishing addresses, which is wrong if the person is guilty and the address is correct, and even more wrong if mistakes are made. Similarly there has been the naming and shaming of people who turn out to be children, which is perhaps not desirable. It also likely that this approach would soon lead to a slew of tit-for-tat vexatious allegations which the police would ignore.

So the question is, how could the zeal of the masses be harnessed to allow the police to make much larger inroads into this problem?

I think there there should be a web interface where people can register, and be background-checked by the police. The police can then work out how many allegations they can handle in a week and set a capacity for how many allegations can be made by each user. Initially these would be spread equally over all the registered users, but the police can rate the quality of allegation so that those who do well (in terms of seriousness of allegation and evidence provided) are allowed to make more allegations, and others fewer allegations.

The police would have to provide clear guidelines on what they wanted to be told about, and the kind of evidence that should, and shouldn’t, be gathered. the total capacity for allegations would also be a measurement of how seriously the police are taking the problem.

Ideally, the web interface would all people to enter URLs and then the police application would screenscrape a copy of the pages immediately as evidence so it’s easy for the user, and the user can’t photoshop a screencapture. however, the need to log-in may make this difficult in practice.

It would be important to have some kind of feedback loop to “users” so that they know their allegations are being acted on, and ideally leading to convictions. If it takes 6 months for anything beyond a nominal notification to make it back to the user, they will think they are wasting their time.

No comments:

Post a Comment