Thursday, 4 November 2010

The Attention Deficit of Crowds

Few of us haven't read James Surowiecki's The Wisdom of Crowds. Its positive slant on wise crowd behaviour is striking, compelling and - if you don't think too hard about it - persuasive. Striking, too, is the total absence of counter evidence; the testimonies of individuals crushed under the heels of history's self-harming rabbles.

If you haven't come across TWOC, the notion is this, concisely expressed in its subtitle: 'The many are smarter than the few.' Obviously, it's not quite that simple. There are conditions, but, once met, crowds - the argument goes - are wiser than you would predict or expect; often wiser than the experts.

They can guess the weight of an ox, for example, without much knowledge of bovine density. While so-called experts, on the other hand, locked in groupthink, will do daft things like invade the Bay of Pigs or precipitate the Wall Street Crash.

It's an idea that matters to us journalists. A lot. One of the transformations we've undergone in the past decade is to realise that 'the people formerly known as the audience' (the crowd) may be wiser, usually, than the few experts (we journalists). They might be closer to, or in, the story.

So we tell ourselves that our jobs now are a conversation with those wiser than us. We work with our former audiences, devise new ways of networking them and the information they have.

One approach to networked journalism that caught on fast was crowdsourcing ... and please don't pretend you don't know what it is.

In theory, crowd-sourced news and investigations are networked journalism perfection. In the same way that in classical economic theory markets are perfect and consumers perfectly rational. That's to say, not at all. Not a bit. Substitute 'herd' for 'crowd' and you start to get the drift.

Here's an example.


During the October, anti-Sarkozy strikes, my local paper in France - Le Journal de Montreuil - launched what was, on the face of it, a fantastic piece of crowd-sourced journalism. But it turned into a journalistic travesty.

Now, Le Journal is a fantastic local paper - it's tragic that more British papers aren't as assiduous, public-minded and community-focused. It's not very 21st C - it's only been online for a year and its photographers still turn their lenses to the audience rather than players at any event.

Google Map of petrol shortages in France.

When petrol deliveries began to be interrupted - on 18 October - Le Journal had a brainwave and set up a simple Google Map to log 'penuries d'essence' - the petrol stations that had run out. The idea was that readers emailed the paper when they found a station that was voided of fuel.

Great idea. Over the next four days, the map sprouted blue flags marking empty stations. It was exactly what real people in real communities wanted to know. Real networked news.

Except two Bad Things happened. Or rather, one Bad Thing happened and another thing didn't. Which was a Bad Thing.

After the first few days, the rush of info slowed - so the map never became complete. Worse than that, as the blue flags sprung up over the map, car owners started to panic. For most of the region, there are only three or four petrol stations within 10km. Once the nearest one was flagged, people got in their cars - with tanks 7/8 full - and headed off to the next nearest without a flag. Just in case.

Of course, the absence of a flag didn't mean the presence of petrol - just that no-one had flagged it. So when the slightly panicking driver turned up at an unflagged station to find it had no petrol ... slight turned into total turned into queues, forecourt fistfights etc, etc.

Wise crowd?

After four days, deliveries restarted. Again, at first the newly replenished stations were flagged ... until there was enough petrol in the system for no-one to care enough about the map to email the paper - though not enough to ensure there was petrol everywhere.

Le Journal's experiment highlighted the problems with crowds and crowd-sourced journalism. First - crowds aren't always as wise in practice as the theory predicts they ought to be or could be. Self-interest is a powerful driver, even within the context of an exercise in public good.

Second - networked journalism needs a much higher level of journalistic 'curation' than its advocates will usually concede. Le Journal's flagging was haphazard and failed to reflect the true situation on the ground, especially when deliveries began once again and the crowd's enthusiasm for the exercise waned.

In the end, because it promised more than 'traditional' journalism but delivered less, it was worse than useless. Worse than a reporter's daily ring-round. And you don't get much more old-media than that.

In one of the villages on the way to one of my local petrol stations, there's an old church. In the 1790s, a 'crowd' of revolutionaries took over the building and renamed it a 'temple to reason and sanity' - words they scratched into the chalk brickwork of the exterior.

'Crowds', 'reason', 'sanity'. I'm not convinced.

No comments: