Working to make government more effective

Comment

Consultation on steroids - or genuine co-creation?

The Government says it wants to harness the power of the crowd, but its attempts so far have sometimes felt more like monster versions of the classic Whitehall consultation exercise. What should the next attempt at crowdsourcing public policy look like?

Asking the general public for their views isn't normally top of the to-do list in Whitehall. Of course it does have to happen, and there is a time-honoured process: green papers, white papers, calls for evidence and 12-week windows to respond.

Times are, however, changing. Back in 2006, the then Labour administration launched the first iteration of the government e-petitions portal. Since coming to power the coalition has gone a step further, inviting the public to participate first in Your Freedom, then the Spending Challenge, and now the Red Tape Challenge. People seem willing to have a go: each of these initiatives has generated tens of thousands of responses.

So far, so good. A crowd is definitely involved. But are we really crowdsourcing?

The classic applications of crowdsourcing are about pulling together distributed knowledge. Sometimes, taking the average of lots of people's (unbiased) estimates is a good way to guess an unknown quantity. Other times, lots of people hold different pieces of a puzzle and, with the right collaboration tools, can assemble something wonderful – think Wikipedia or OpenStreetMap. Other times still, it's worth giving lots of people the same problem to solve and seeing who does best. Foldit is a great example of this, with the spatial problem-solving skills of the best players outperforming supercomputers.

Initiatives like the Spending Challenge are doing something a bit different. Ultimately they are exercises in rounding up large numbers of ideas. Taking advantage of the internet to broaden participation is long overdue, and clearly some of what is generated will be useful (though given the common-sense nature of the ideas being adopted, one might argue this says more about the civil service than it does about the power of crowds). There is no particular reason to think, however, that the quantity or distribution of responses gathered in this way gives us much indication of what the right thing to do is. This isn't surprising – we all know that knotty policy problems rarely have an objectively correct answer.

Beyond gathering ideas, then, how might policymakers put crowdsourcing to better use?

The key will be to deploy crowdsourcing in pursuit of solutions, not consensus. Our elected representatives are here to deal with questions of balance and judgment, and we have mechanisms to hold them to account. The power of the crowd is better used to aggregate, refine and innovate. Here are a few thought starters:

  • Encourage people to share nuggets of information that will help to direct public services and processes more efficiently. Portals like FixMyStreet are leading the way here, providing a light-touch channel for data to flow to local authorities
  • Specify the Government's policy intentions clearly, say for a new tax relief, and put the draft legislation up onto a wiki. Let us edit, revise, revert, build and correct it before it gets anywhere near being printed and taken to the Palace of Westminster
  • Open up (appropriately anonymised) telemetry data for Direct.gov.uk and run a competition to find the best collaborative filtering algorithm for the site. The team at GDS should be allowed to enter but not necessarily expect to win...

Related content