Crowdsourcing, humor, participation

It’s been good fun hanging out in Barcelona with my fellow speakers, both the wonderful organizers like Juan Freire and Ismael Peña-Lopez and guests like Carol Darr, Andrew Rasiej and Tom Steinberg. Tom was kind enough to hunt me down for dinner on Thursday, and we had an excellent conversation that I’ve been chewing over for the past 48 hours.

Tom is the brilliant founder of MySociety, a British organization that it relentless in its quest to make UK politics more open and participatory. Smart people around the world look to Tom and the folks he works with for ideas on how to make elected officials more accountable, link disconnected people in local communities and use distributed reporting to document social ills and push for change. Given the opportunity to pin Tom down for insights, I asked him about his thoughts on getting people to connect with people across national and cultural lines.

This can be a tricky topic for political organizers. Most organizers are deeply concerned about the erosion of local civic life, as documented by thinkers like Robert Putnam. It’s easy to misunderstand my obsession with pushing people to connect across lingustic, cultural and national barriers as a lack of interest in connecting locally. I see a great deal of importance in both, though I’m sobered by Bill Bishop’s new book, “The Big Sort“, which makes a pretty good case that Americans are sorting ourselves into homophily traps geographically, and that connecting with our neighbors may increasingly mean connecting with people who share our perspectives and prejudices.

Rather than fighting the local versus global battle, Tom offered interesting and provocative advice about what might work to get people who aren’t otherwise inclined to connect to do so. His projects are finding interesting ways to use games to get participation that would otherwise be difficult to organize. For instance, MySociety wanted to align thousands of hours of taped debates in the House of Parliament with transcripts, so that these videos would be wholly searchable. When automatic methods failed, he and his team built a tool that asked users to complete a simple task – watch a video and push a button when a certain person began speaking. Participants would be scored on “league tables” for the number of times they’d pushed the button, aligning the video – some participants ended up coding hundreds of videos for the project, and all the video was tagged within a few weeks.

Using the same technique, Tom’s now trying to get people to classify Yahoo groups for him, specifically Tahoo groups that mention the term “residents” or “neighborhood”. There are 45,000 of these groups, and Tom wants to know what geographies they address. That way, he can build a service where you send a text message containing your zip code to his servers, and they respond with information on online groups you could join that cover issues in your neighborhood or community. (You should pitch in and help him, if you have a chance.)

So riffing on the idea of games and league tables, Tom wondered whether the way to engineer more international connection is football. Specifically, he suggested that Global Voices or some similarly globalizing entity organize online chats around World Cup matches. Chats would invite nationals from both sides represented – Ghana versus Brazil, for instance – to chat online during the game. Trashtalking would be heartily encouraged, but the hope would be to get beyond insults to an actual conversation about football heroes, national pride, politics, etc. I’m guessing this would require a certain amount of careful engineering – we’d probably limit participation so that one side didn’t overwhelm the other (20 Brazilians, 20 Ghanaians per chatroom, for instance) and recruit some bridge-figures, people who spoke both English and Portuguese and had some understanding of each country and culture.

Tom offered another idea, which is either a great way to start intercultural conversations or a surefire way to start a war. He proposes putting together an online database of regional and national prejudices, offering as an example a recent trip he took to Germany where conference organizers declared they’d be taking “a Belgian lunch”, i.e., a very long lunch. What do expressions like this reveal about what we think about one another? Are these opportunities for conversation about cultural quirks, or are these invitations to flamewars and fisticuffs? (I offered the data point that, when I visited Yerevan, Armenia, a few years ago, one of my hosts excused himself to go to the bathroom with the phrase, “I need to visit the Turkish embassy.”)

Ghana’s one of the healthiest societies I’ve ever seen in terms of resolving tensions between ethnic groups. One of the reasons, I think, is a healthy sense of humor. A great deal of Ghanaian humor depends on ethnic jokes and laughing at each other’s perceived quirks. (I watched a Ghanaian comedian bring the house down in Accra by stepping onto stage and singing a song. When my companion finally recovered enough to explain the joke to me, she told me, “He’s an Ewe, and he’s singing a Ga song, but he’s singing it in Twi.” And then she collapsed into laughter again. Guess you had to be there.) So maybe a wikipedia of ethnic stereotype – Tom calls it a “hatebase”, but I prefer the time “haterbase” – isn’t quite as crazy as it sounds in sparking conversation about and across our differences.

I suspect I’ll be rolling around Tom’s ideas around games and crowdsourcing until I can think of a clever way to harness this power for Global Voices. We are, after all, a community based around voluntary participation – finding a way to make that participation more fun, less involved and easier to accomplish is probably a smart thing to think about. I susect that folks like Tom are likely to find a profitable line of work somewhere soon figuring out how complex problems can be broken into crowdsourcing tasks and outsourced either to volunteers or to systems like Mechanical Turk.

Talking about the decision to use volunteers rather than Turkers, Tom argues that people are looking for ways to participate in useful projects. That squares with my experiences as well. In the wake of Hurricane Katrina, I lent a hand recruting volunteers to enter data on missing persons in the southeastern US – we were completely overwhelmed by people’s interest, and a task we thought might take two weeks was done overnight. I’ve lately been wondering whether one of the keys to getting people interested in international news is attaching the ability to get involved in social change projects. It seems logical that people who are interested in changing circumstances in Darfur are likely to be especially interested in news from Sudan. Is it possible that this runs in the other direction as well, that attaching an opportunity to get involved with a protest or a fundraising effort would make people more likely to read a story on Somalia?

In other words, it was a very good decision to have dinner with Tom, rather than staying in my hotel room and answering email. That’s now my official excuse for everyone’s email I’ve recently failed to answer…