Creator of the Map, Ben Charlton, removes it from his site. Here’s his post explaining why:
I have decided to take down the map. Many people have commented that the map does give a false impression of accuracy, despite my making this clear, and I’m tempted to agree. I do not want to single anybody out and by removing the accuracy from the map it is possible that it ends up incorrectly implying a property contains a BNP member. It has been suggested that an inaccurate map that doesn’t make that clear is worse than publishing the list itself, and I think that’s a reasonable comment.
[Updated: Please see further updates below]
As news about the leaking of the membership list of the far-right/racist British National Party broke in the UK, it quickly became a headline story in the mainstream media. But the chatter amongst geeks on Twitter this morning was about the possible “mashup” that could be done of the members’ postcodes with a Google Map. As soon as that idea had propagated, it was bound to happen, since the list has been available on Bitorrent and on Wikileaks for around 24 hrs now. Frankly I thought it would take a little longer, but as I debated the issue on Twitter this morning, someone sent me a link to an already-built map.
And it’s scary.
The map is a sea of red pins, indicating that there are in fact BNP members spread pretty widely across the UK. London is barely visible under the map. Zooming out means that the map can no longer render all the data points adequately. It’s worth remembering that Britain is in fact one of the least racist, most tolerant nations in Europe, but the map looks pretty scary covered in these large red pins.
As various news organisations have been reporting, the membership list for the British National Party has been leaked online. It seems a court injunction kept this quiet for some time, but the list has well and truly escaped – currently hosted on bittorrent and the always excellent wikileaks site.
Curious about the list, I managed to find a copy and after checking it for the obvious odd family members or people you know and think might be members, I thought it’d be cool to make a Google Maps mashup of the data. I extracted the postcodes from the member list and converted them into latitude and longitude co-ordinates, which are then plotted on the map.
The map is a bit slow to render, especially if there are a lot of data points, but it’s interesting to see the distribution of members. As I’m only generating a list of latitude and longitude points, I’m not revealing any of the actual list data. If you want to find out who the hate-filled racist that lives near you is, you’ll need to find and look through the list yourself.
The implications of this action are pretty big. I speculated on Twitter this morning that a mashup which identified the actual locations of BNP members would be highly problematic, and possibly even subject to vigilante attack. However I still believe that a map which showed more general areas, like towns and cities, could actually be helpful to local authorities for creating policies to tackle attitudes towards diversity. If you were a local councillor and had been made aware that there were lots of BNP members in your area, you may be able to do something about the attitudes which lead to support for such a far-right political party.
But now this more accurate map is out of the bag, and others are probably planning similar maps, the implications are far reaching. What no-one would wish is for a witch-hunt to begin. Plotting this map will make the data available to just about anyone. My personal opinion is that in this, quite accurate form, it should be taken down. On the one hand the BNP has a notorious history of violence. On the other, it also has a history of attracting disaffected young people who later regret joining – their lives should not be tarred by their actions as naive youngsters.
UPDATE 1: I emailed Ben Charlton about the map and emailed him some questions to which he has now responded.
He says the map is not as a accurate I feared: “I deliberately only pulled out postcodes from the file, converted those to latitude/longitude, then stripped a chunk of precision off the end. It gives a vague location, probably to street level, but it’s certainly not accurate by any means.”
1. Why create the map?
“I thought what I’ve seen a lot of people say – it would be interesting to visualise the data in a way that makes sense to people. It’s all very well being able to look for people in your home town, but it’s nice to see easily how that compares to membership in other locations.”
2. How accurate are people’s locations? Would someone be able to work out they lived next door to a member.
“The original list is very specific – listing addresses and names. I deliberately only pulled out postcodes from the file, converted those to latitude/longitude, then stripped a chunk of precision off the end. It gives a vague location, probably to street level, but it’s certainly not accurate by any means. Of course, there’s nothing to stop people then looking on the proper list – it’s not hard to find now.”
3. Do you plan to maintain the map?
“It’s based on the leaked list. If there’s a new list leaked I might update it, but for the time being it won’t change.”
4. Are you worried about BNP members identifying you?
“It’s always a concern, but I deliberately kept my app vague so as not to directly identify anyone as it’s not my intention to cause upset. I’d hope I’m not a target for too much abuse – especially compared to what a lot of other sites are saying.”
Update 2: However, an inaccurate map could be more dangerous than an accurate one. As others point out in comments below, the Google pin can, and perhaps should, be changed to something more like a “circle with fading semi-transparent edges” to prevent someone getting the wrong idea about their neighbour.
Update 3: Map creator Ben Charlton has now posted a heat-map of the data here but it largely shows what you might expect – that BNP membership roughly maps to urban centres of population.
Addendum: Others are making interesting use of the data, such as this graph by region.
And at least some can see the funny side now.