How should the public service engage with controversial topics online?

In my experience, where possible, Australian public servants avoid controversial topics when consulting with the public.

Controversial topics are messy, unpredictable, raise high emotions and draw out divergent viewpoints - making discussions difficult to manage and control. They also often edge into political matters which are outside the scope of the public service, who strive to remain professionally apolitical in their service to their political masters.

Of course, often active discussion thrives on controversy. Radically differing viewpoints and high emotional engagement leads to energetic and insightful debate. They can soar to great heights - and plummet to unspeakable depths.

On the other hand, discussions on topics where most people agree tend to be largely controllable - but also predictable, boring and repetitive. Why bother repeating a 'me too' point or stating something that seems self-evident?

People rapidly lose interest and drift away when there's no cut and thrust of debate and the conclusions are easily arrived at from the proposition.

For public servants striving to generate online discussion on blogs and forums there's a difficult line to walk between proposing topics that are controversial and those that are safe.

Instinct tends to draw public servants to safe topics, where we can predict the likely responses and avoid the risk of heated and uncivil discussion. It's easier (and more risk-adverse) to manage a discussion when the outcome is obvious, it requires less time, effort and critical judgement - and also requires less Ministerial correspondence, scrutiny from senior management and career risk.

However it is hard to get audiences to engage on many safe topics. The public is uninterested, has already agreed on an outcome or simply doesn't feel entertained and stimulated by many safe discussions. To be frank, they are boring and don't materially add to the policy or operational discussion.

So how can public servants engage with controversy online, without engaging too much?

Fortunately there are a number of models on how to do this. People have been stepping through this minefield for thousands of years in physical discussions and many of the same tools work online.

The first approach is to structure the debate where you cannot structure the content. Find a topic and choose two positions. Form 'teams' to argue each of the positions in sequential order. Have an audience able to make side comments and vote on which team did a better job of building a compelling case.

Those of you familiar with formal debating will recognise this approach. It still allows passionate discussion but within a straight-jacket of format and set positions, which avoids a free-for-all. There is a beginning, a middle and an end - which prevents it dragging for an unknown period and usually there are only two 'sides' - positions - which an audience can take.

A second approach is an expert panel, where each expert provides their own position and the audience can comment or vote on the position they most ascribe to. This is more flexible than a debate, however still largely restricts discussion to positions set by 'authorities'. While it provides greater flexibility for diverse views it can also limit discussion and debate between the distinct expert positions as the experts may not be as willing to debate each other or have their supporters do so.

A third option, which I term rotating perspectives, also supports multiple positions, but each is examined sequentially over time by an audience. This focuses discussion on the pros and cons of each particular position over time and allows the community managers to introduce new perspectives based on the direction of the discussion. While more flexible and responsive to audience feedback than an expert panel, and encouraging online audience participation, this approach can lead to uneven analysis of ideas. Early positions may receive more discussion (based on a big promotional launch) and greater critical thought - as they are visible longer for reflection and responses can be made later in the process. This also risks having members of the audience pre-empting certain positions ahead of time - though this isn't necessarily a bad outcome as it increases the sense of active discussion.

My fourth, and final - for now - option is to provide separate groups for discussion of each different position. These can be linked or merged where positions converge or separated out where a single position diverges into several. Audience members can suggest and create their own positions, which then become new groups for discussion. Towards the end of the discussion many positions may merge towards a common core thread - or they may diverge, identifying the most intractable issues that need resolution. Similar to workshopping, this approach is complex, requiring additional moderation and an appropriate technology platform - such as a Nationbuilder (used for Australia2) or Ideascale which allows ideas to be separately discussed, merged as required and with a degree of automated nouse that can merge similar positions.

There are other approaches as well - breaking down a topic into individual issues and discussing each separately, or having the community rate contributions with the aim of self-moderation (which works quite well in some online communities).

What other approaches can you suggest that would allow the public service to engage with controversial topic online while remaining comfortable about the risks?