Using Google Wave to debate climate change
I am by nature an optimist. I'm a big fan of Google by choice. I've also had a thing for Google Wave ever since I saw the I/O video in May. All of which I declare upfront because as someone who believes fundamentally in the value of honest debate I think it's important to state my influences first.That said, I think Wave is the nuts. I do. Now I'm aware this is not a universally shared view, criticism seems to fall into either the 'it's buggy' or 'what do I do with it?' camp. The first I don't get, we're talking about very new technology in an early stage of development; I'd actually be less impressed if it didn't have bugs.
The second I have more sympathy for. First impressions of Wave can be a bit 'meh' unless you've got people to Wave with and - crucially - things to Wave about. Wave is great for many-to-many conversations where at the end some kind of consensus needs to be formed. Not sure about you but I seem to have more and more of these these days. Use Wave for this once and you wont want to go back. I also think there will be a whole range of new uses it can be put to too, one of which I proudly present to you here.
My name is David Crane, I run the non-profit debating web site debatewise.org, and we put together a group of 1,000 young people from over 130 countries and asked them to use Google Wave to debate issues arising from the Copenhagen climate change conference.
We had people sign-up from Iran, Iraq, North and South Korea, Burma, Bhutan, Vanuatu, Venezuela, Vietnam, 25 countries in Africa, 30-odd in Europe, many parts of Asia and both halves of the Americas. As you'll imagine, not all these people are equipped with fast machines or blessed with deep technical expertise. Most don't speak English as a first language. To cap it off we only came up with the idea for the project in mid-October. Six weeks before the conference started.
So it's fair to say we took on a challenge or two. The tight deadline meant we had to use Wave out-of-the-box. The differing levels of geek knowledge meant we had to get a good few people Wave'd up in a short space of time. And as anyone who's worked with large groups of people knows, working with large groups of people is hard.
Of course the benefit of these kinds of intense projects is the amount of learning that goes on. I hope to share some of that with you now.
Wave is not a whole new world
In many ways, we found managing conversations in Wave much like managing them in any other online environment: create reward mechanisms for the things you want to encourage, develop ways of dealing with the things you want to prevent. Praise good activity publicly, deal with the bad in private. Understand and deliver what people want in return for giving of their time and expertise. Find the champions amongst your users, they want to do more to help. Offer people a variety of ways to get involved and make it easy for them to transition from lurker to participant.
Some people need a bit of coaxing, others will only get involved in particular issues. So we'd create Waves on a range of different debates from the general "Is Copenhagen going to be a success?", to the specific "Should Bangladesh get 15% of any climate fund?". We found provocative topics useful but they needed to be employed with care. The Yes/No/Maybe gadget and other voting tools helped people take that tentative step towards participation.
It is not enough to create opportunities for involvement though, some people need active encouragement. Our approach was to repeatedly invite them in, create a welcoming environment when they arrived, reassure them that their communication skills were good enough and step in before disagreements turned to flaming.
One thing we want to do more of in Phase Two is develop our reward mechanisms, even though there are doubts about their value. The first school of thought is that people give to these kinds of projects simply because they want to get involved and reward mechanisms, even for things as basic as karma, devalue their contributions and can drive people more towards earning points than saying what they feel.
The second, the one I tend to lean towards, is that reward mechanisms are useful in helping encourage the sort of behaviour you'd like to see more of. Multi-dimensional award systems, such as those on reddit.com, reward people for different kinds of activity, not just those the majority agrees with. I think these tools are useful in a very new environment where the ground rules haven't yet been established.
And that's the other thing the hippy in me wants to stress. I don't like lots of proscriptive rules and believe that the best communities are the self-moderating ones. However, I also believe that new communities benefit from having a set of guidelines that state what is and isn't expected, valued, tolerated and appreciated. These can evolve over time of course, but if the basic framework is set from the off it gives the community a good chance of growing in the direction you'd like to see it go.
Wave is different enough
George Bernard Shaw once said: "The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself". An understanding which tends to sail high over the heads of people like drug war warriors but is one web professionals generally get. It's certainly true here.
A lot of people find Wave a very new environment and when in very new environments most people look for clues about how to behave. Not everyone of course, one person told me, and I quote: "Oh I thought I could delete anything because Google Wave is in preview". Luckily, the majority of people are decent and reasonable and want to do the right thing. And they'll err on the side of caution until they're sure what that is.
The problem of course is that people don't generally read manuals. Web users are notoriously short of time, are prone to scan and detailed instructions have about as much chance of getting read as the T&Cs on the next major release of the iPhone software.
Users prefer show to tell, hence the popularity of screencasts. Happily this principle is one of the less-heralded benefits of Wave. Wave is an editable conversation (which, along with 99 levels of undo and a universal find, are the three things I'd like in life 2.0). The ability to edit any part of the conversation sets Wave apart from a wiki and allows us to model the behaviour we'd like our users to adopt.
In our case, we wanted people to form summaries of the discussion in the first blip. We tried asking and telling people to do this. But what seemed to work best was modelling it by leaving little notes in a blip saying "Part of point added to the summary". This rewarded people publicly, highlighted the type of conversation we wanted to see more of and demonstrated the action we hoped others would take.
These notes can also be used to gently steer people back on topic, guide them towards further information or a more suitable Wave or explain why part of their blip has changed (hopefully only to add value, remove abuse or fix spelling and grammar errors). The appearance in your blip of somebody else's icon can be disconcerting. These notes help clarify the situation and demonstrate to all what you'd like to see more of.
Other things we learned, in no particular order:
People route around obstacles provided they think the objective is worth the effort. Access to technology did not in itself prevent people taking part. Confidence with technology, lack of desire and I'm sure a good few personal reasons, did. In the main we were delighted with just how many of our technically disadvantaged panel managed to have their say.
Waves pick up speed. The first blip may take a while to get a response, the second blip less time, the third even less, and so on. I think there are two reasons for this, people prefer joining in busy Waves and more blips means more things to respond to. If you want to bump a blip to the top of a busy inbox a little edit will do the trick.
The real-time typing feature takes a bit of getting used to but often saves time (knowing osirisx11 has entered text was always frustrating for me, post-Wave it seems pointless), this feature also created dynamism and buzz around the debates, which encouraged more participation.
Broadly speaking the feedback we've got from users was very positive. Most people found Wave intuitive and were able to get their head round it quickly. They enjoyed the experience, found it less formal than a forum and thus easier to jump in to.
Phase two of our project starts on the 18th January and will see our panel organised into countries or regions and asked to debate how decisions made in Copenhagen will impact on them. In the true spirit of debate we will also ask people to argue from the opposite position to the one they would normally adopt. For example, Bangladeshis could be asked to argue why they do not deserve 15% of any climate fund while the rest of the world argues that they do.
If you'd like to get involved in this in any way, by writing code, finding panel members or adding expert views please get in touch. We'd love to have your help.