Free data, yes! But please don’t go mental on maps (Web 2.2)

Last year social media was twitter, this year it is maps; well in my world… and it is a bit annoying.

There is a groundswell of activity to free up the public data, that which we are allowed to know (not the personal stuff) thanks mainly to the Power of Information report and also the general noise: ref Rewired State 🙂 *plug* this is great, but what is not so great is that the easiest thing to do with this data is to shove it on a map, and combine it with other data sets.

Hacking data together and putting it on a map is fun and clever; but not always useful.

Richard Pope sent me a presentation of his this morning, after I yelped on twitter about the number of map requests I had pouring in, not only at work, but it seems to be everywhere I go, people I speak to… everyone’s the cartographer. Here’s his presentation:

My favourite page from his presentation is here:

picture-1Just because you can!!

And look, here’s all the help you will ever need: and here are a load of must sees.

Mind boggling… let’s cut to the chase. There are two kinds of map I am regularly asked for:

User-generated content maps

One where anyone is able to upload data onto a geographical map of an area. Wonderful, but why and what will happen to the information? Who is moderating it? Can I really just stick a pointer in this and say that in this area blue squirrels regularly admonish the badgers for discovering their hoards of cornflakes? If I can’t and I have a text box I can fill in, how long before someone is able to moderate and upload it? And what is the cost? And again in employee hours?

Let’s take an example. I am going to use user-generated content to crowdsource and rate pubs in GU1, (by crowdsourcing, I mean I am not going to provide details of the pubs, but enable anyone who finds their way to my map to identify where a pub is). Next I need to give my trusty crowdsource a text field to fill in the detail of the pub, and in true social media style, enable rating categories for other people to fill in once the pub has been plotted and identified. Handy information for someone new to the area, or passing through. However, unless I personally check that each pub exists, my raw data is potentially flawed; thereby the rating on each establishment is one step further away from flawed: flaw-flawed? Until and unless I personally guarantee the accuracy of dataset 1 my map is firstly annoying and secondly pointless. (Yes this can be applied to any crowdsourced based app, but my topic is maps :)).

Now assign the above example to, say, crime.

User generated content on maps is fabulous within online social communities, where the trust is implicit and anyone caught abusing the technology would be duly dealt with: crowdsourced public shaming takes no prisoners I imagine. It has no place in the public, nor in the private, sector.

I hear a collective gasp as everyone watches Emma’s career prospects disappear down the cartographically defined toilet… I am not saying that there is no explicit trust in the public or private sector, that is not the point. The point is that online social communities operate under very strict trust conditions, and I am assuming that those reading this post have already understood this, that’s Web 2.1.

The psychology around web communities aside, if you (the crowd) were providing data on a public sector map, for example, you would rightfully (possibly) expect something to happen with your bothering to plot your pin/piece of content. Therefore it should never be the case that a map is enabled for user-generated content UNLESS that map is plugged in to the geographically associated people who can respond or deal with the issue that has been raised.

Imagine every pin laid on a public/private sector map to be the same as a request for a duel at dawn. Unless you can guarantee that you will take each challenge and deal with it: don’t do it.

Publicly held data-driven map mash-ups

By this I mean you take a plain map and plug your APIs/other data sources into it. Now these are the rightful domain of public sector social map enthusiasts: taking proven data and plugging it into a geographically represented area is fine. But again, not willy nilly (ref Richard’s map above). These would provide informative pictures of an area: here is a great example of such a thing. A clear key and a text box that filters information, lovely.

The temptation to avoid with this one is simple: TMI (too much information). Web 2.0 technologies do not mean that we ignore the basic rules of web 1.0: who are you doing it for and why? If you are providing the mapping information for allotments in the UK, think before you add another dataset, just because you can include the location of the nearest post office, does not necessarily mean that you should.

Of course, these are far less risky map hacks than the UGC one detailed above; but they are utterly pointless, in equal value, if you are not clear about what information you are providing through the map, why you are choosing to do it using a map and where to stop.

A combination of the two is inevitable and the risk of pointlessness/being sued escalates with each step away from the original question: Why?