It is a while since I droned on about crime mapping, but came across an interesting blog by a PhD student (who happens to be a retired police officer).

The site has a fairly lengthy post (actually a republishing of a paper) covering some of the problems with mapping, as well as some potential future uses.

I’ll say from the outset that I’m a fan of point mapping. While I accept some of the criticisms made in the paper about point mapping I’m a firm believer that we shouldn’t let perfect be the enemy of good; just because we cannot create an ‘ideal’ map doesn’t mean we shouldn’t try.

But one point with which I wholeheartedly agree is that the current set of maps available do little to offer any reassurance.

The Met’s maps, for example, consistently report most of London having average crime. Currently it suggests 29 out of 32 London boroughs have average crime, with Westminster having above average and Richmond and Bexley below average. Nowhere appears as a high or low crime area. (When you zoom in closer every single Wandsworth ward also has average crime.)

While there is, I know, some statistical justification for that, it immediately seems nonsensical. People ‘think’ some areas are high crime and some areas are low crime. When the official maps so blatantly contradict this belief by suggesting everywhere is ‘average’ it immediately reduces their perceived value and, I would suggest, the degree of trust placed in them.

This is not to say that I believe my maps are faultless, I know they have their flaws. However, I believe that by providing the fairly raw information, like the type and approximate location of a crime, residents can make a better assessment of whether crime is high or low.

In other words it better serves the purpose of either reassuring people crime isn’t that high, or giving the information needed to challenge the police, council and other partners if they believe crime is too high.

But it is incredibly difficult to develop further. I would love, for example, to do punishment maps: detailing crimes that had seen someone caught and prosecuted. Again, this serves the purpose of reassurance (where people are caught and punished) and allowing people to hold the criminal justice system to account (where no-one is caught and punished).

But if the information on sentencing is publicly available I can’t find it. Indeed I find it slightly worrying that South Western Magistrates are busy just a few hundred metres from me but that it’s almost impossible for me to find out what they have done.

The Saferview paper also raises the idea of publishing the routes taken of police cars and beats. A fascinating idea, since it would be an easy way of refuting, or proving, the ‘police never come down my street’ argument. Although the data would need significant cleaning, since timed GPS data would give away specific addresses the police visit, it could potentially be an incredibly reassuring dataset to publish. (Or perhaps not, if it turns out that the dots never really move from the stations.)

But it occurred to me when thinking about these is that the key principle is, really, about freedom of information. It’s essentially saying that information is not ‘owned’ by the police, the council, the government, or anyone else except as a custodian on behalf of the public and, therefore, the public have the right to see it since it reflects what is happening to the public and what is done on behalf of the public.

The key is not whether there is any use for the public in having access to the information, that is irrelevant, it’s whether there is any compelling reason the public shouldn’t have access. What astounds me is not that these ideas are starting to come out, it’s that we are only just starting to talk about these things.