This site was developed to support the Government 2.0 Taskforce, which operated from June to December 2009. The Government responded to the Government 2.0 Taskforce's report on 3 May 2010. As such, comments are now closed but you are encouraged to continue the conversation at

Layered Participation

2009 June 25
by Nicholas Gruen


If you click on this link you’ll find an interesting Vox Pop tool for public deliberation on what should be the “Top [policy] priorities for Australia”. Now although public discussion of this kind is by definition useful or at least wholly in the spirit of democracy, there’s a problem. Or rather a number of problems. One is that it’s a kind of Vox Pop approach to policy – which is to say that as people swim through the issues on the list and vote for or against various propositions, their views may be almost completely uninformed. Since we don’t require people to prove their informed to vote, this isn’t a show stopper. But nor are they representative. Not only are they not representative, but the moment such a site began to have any direct significance it would immediately be a potential target for various political campaigns. IIRC when Barack Obama asked citizens to nominate their top priorities in a similar kind of exercise, a group seeking to liberalise marijuana laws got going and registered thousands of votes to push their issue up the agenda.

Anyway Paul Roberts offers this gloss on the new tool (comment 10 in this discussion thread).

After some more thought I feel that I have a germ of an idea. I’d call it ”layered participation”. It just occurred to me this evening so I’m floating this as an idea without much critical analysis. I’ll see what the crowd reaction is.

The citizens priority tool would be broadly in line with as is developed. It would be used by the digital literati for the most part. I would call this the underlying layer. I’m suggesting an additional layer – an overarching one in terms of participation rate – drawing on the contributions from the underlying layer. The “overarching layer” would have an incredibly easy UI and would invite/attract many more citizens to share their view of priorities by polling them. Like, “Here are the top 5 priorities from the underlying layer, what do you think? Make your selection here”.

In summary, the concept is one of layered influence. Those willing and able to identify and describe a priority, and to make a judgment on a mass of other priorities, can do so. Those that are either not willing or able, through circumstance or whatever, can still participate at another level. Tick the box. As I say this is just the germ of an idea. It may be a silly idea. But I offer it for consideration.

No Paul, it’s not a silly idea – or at least I don’t think it is.  It is the beginning of building a bridge between the simple vox pop and something which is offering some tentative possibilities of finding a way ofordering preferences. To do so it imposes opportunity costs on voting for one thing – because voting for one thing means not voting for something else – just as it does in a ballot box. Another thing I think is important is finding ways of building reputation on the net.  Of course e-Bay has been doing it for a long time, as has  Slashdot has a protocol through which people can qualify to be unusually worthy commenters.  We need to build those kinds of mechanisms to add depth to the discussion, rather than just have a cacophony of voices full of sound and fury and signifying nothing (well at least less than it might).

8 Responses
  1. 2009 June 28

    You are never going to get a legitimately representative sample of public views unless you take the jury approach of randomly selecting people. The best approach is a sampling method called stratified random sampling, which also ensures that the sample has equal gender, proportional age and education to the general public.

    The list priority problem is one of process. If you are running a visioning exercise, then listing priorities is precisely the aim. But if you want a specific solution (or solution set) to a policy problem, then micro-processes can be used by a citizens’ jury to reach that goal that are more consensus-seeking.

    The last point I make about such deliberative processes is that they are facilitated, which tends to iron out the noisy participants and encourage critical thinking.

    • 2010 April 14
      Madeleine Kingston permalink

      Ron has a very good point here regarding a representative sample. Stratified samples are used by most marketing research companies.

      I remember how very expensive such exercises were if farmed out for professional data colleciton and interpretation, and ahve done my share of collecting longitudinal data for studies commissioned by various government bodies and by commercial firms.

      Facilitated dialogue would deal with the challenge of obliterating noise interference is well known.

      Setting up citizen’s juries, allowing sufficient time for consensus etc will be challenging, but will provide a broader base and cater for several viewpoints.

      I can see some hiccups with people’s individual approaches. My data is stored electronically. I am able to respond promptly because I have stored data that is often relevant. I could not do the same if sitting with a group of people without access to my own data or make the types of responses I do in the depth that I do. I often prefer to search out material on topics that I am unfamiliar with or wish to use to support a point.

      In my view some of the existing policy bottlenecks need specific solutions not general approaches

      On the other hand I believe it was Nicholas who observed how very difficult it was to identify citizens out there with sufficient interest and stamina to stay involved and contribute to the dialogue.

      To some extent I feel I did Ron an injustice in one of my recent postings as I failed to acknowledge his strong points or take proper care in response, – which I will have to return to another time, since I too am interested in the concept of deep penetration of policies and real collaboration that is citizen-focused.

      Ron is far better versed in the concepts of deliberative decision making and to make any sort of further comment I would need to read further.

      That’s all I can manage on this for now.



  2. 2009 June 30
    Martin Stewart-Weeks permalink

    Good points Ron. There’s a great article in the current edition of the Stanford Social Innovation Review ( – check that?) which looks at three different types of collaborative platform…one for exploring, one for determining (which usually involves more depth, deliberation and difficult choice-making) and one for executing. As Ron says, you need to be sure you know which one you need and which one you’re on, because the rules change considerably each time.

  3. 2009 July 2
    Paul r permalink

    Wow, what a buzz to see my comments being taken up here by Nicholas Gruen :) Actually, I’m even more pleased to see a shared approach to emergent thinking on display by TF members.

    In contemplating a possible role for layered participation in government/citizen interaction, I would treat the poll results as one set of indicators of public preference along with other sources. Poll results are what they are – snapshots of perspectives at the time and it depends on who engages in the polling. Still, as well as polling platforms, there are others to take advantage of. The social web has scale and depth like no other social network. Just as there could be layered participation so the potential is there to tap into multiple social networks using multiple platforms. And let the computers to the aggregating and categorising.

    Self-selection is an issue but there are ways to build depth into the conversation. The platform in question in this posting has space for participants to discuss issues, upload documents and share information across web platforms. That whole issue of linking and sharing has massive networking potential.

    Participant reputation is a significant factor in evaluating preferences. There are new influencers active in the social web and their views would carry more weight…that may be positive or negative depending on value judgements and so on.

  4. 2009 July 4

    I recommend you review some of the functional designs used in Open Source software for collaboration. Bug tracking tools like JIRA and Mozilla have a kind of voting system attached to bugs or issues. People can subscribe to the items they are interested in, but votes don’t always determine what gets worked on.

    Slashdot uses a very different approach, feedback from registered users, in two layers. Moderation and Meta-Moderation.

    The most interesting part of the Slashdot is how they deal with the nasty undercurrent, anonymous users are labelled with an embarrassing name and they are also given a credibility rating of zero. Users with very high credibility ratings are generally those who participate responsibly, non-disruptive to the community, have been in the community for some time and moderate others fairly.

    You can choose to view any level of content you like, but the default is not zero.

    It is important that ALL moderation is reviewed and critiqued so that no moderator can edit the conversation to a point where it’s altered completely.

  5. 2009 July 7

    The best thing about environments, virtual or real, that enable unfettered, unmoderated access is that anyone – absolutely anyone – can get in there and participate.

    The worst thing… is that anyone – absolutely anyone – can get in there and participate.

    Why do we tend to worry so much about whether or not such environments are “…representative…” or even free of cranks, wannabes, fraudsters and/or self-interested or conflicted-interested contributors? Sure, these types often muddy issues, they can make it harder for more ‘genuine’ or more ‘valid’ contributors to get their voices heard, but in the end, is that so bad? In the Obama example given by N Gruen, the bias was identified – not via moderation at the point of input, but by reasoning people looking at the outputs. So what if your cranky retired neighbour spends all day writing letters to the editor or messing up comments pages on blogs like this? In the final analysis, we end up tuning out such background noise, just like we do in ‘real’ life, usually by simply ignoring it.

    I say that we should worry less about moderating inputs, and we should worry more about understanding and judging what people are saying.

    • 2009 July 20
      Aron permalink

      Just sharing some thoughts…

      In planning online participatory systems – it’s can be useful to try (or at least offer) to meet and interact with people where they like to ‘hang-out’ rather than forcing them to come to a central place / site.

      Being available to someone in their preferred space can improve the perception of comfort / security and the likelihood and frequency of interaction.

      This may be achievable through broadening the range of potential electronic touch-points eg. access via mobile phone, facebook, twitter, interactive voice relay, redistributable blog widget and even simple email.

  6. 2009 July 7
    Bec permalink

    You could improve your hyperlink text by following the advice at

Comments are closed.