Collaboration | pureinformation.org

Recently, New York University Professor Clay Shirky gave an amazing talk at the Harvard Berkman Center about his new book, “Here Comes Everybody.” In his talk, Shirky says, “the Internet isn’t a decoration on contemporary society, it’s a challenge to it.” He claims that the Internet in action represents an important technological marker in our society such that the world pre-Internet looks very different from the world post-Internet. Clay says, “a society with the Internet is a different type of society, just like a a society with a printing press was fundamentally a different type of society.” We’ve seen this before, but with really only a few things: printing press, telegraph/telephone, recorded media, and broadcast.

Clay observes that the aforementioned media were revolutionary but limiting–technologies that create groups don’t create two-way communications, and technologies that create two-way communications don’t create groups. For the former, he uses an example of a magazine. Magazine publishing can foster a large group of followers, but doesn’t facilitate fans communicating with each other or fans communicating with the magazine staff (at least very well). For the latter, Shirky uses the example of the telephone. Telephones create easy two-way communication channels, but don’t really build groups around it.

We now realize that with the Internet, we have a network that is natively good at group forming. As Shirky asserts, “group action just got easier.” Before the Internet, Clay says, groups get complicated before they get large. He offers a classic diagram that displays the ballooning number of node connections as node size increases (5 people, 10 connections; 10 people, 45 connections). We realize that there are native disabilities to large groups, but today sociologists point out that the Internet provides for “ridiculously easy group forming.”

Clay offers the “reply-all” email feature as one of the first examples of ridiculously easy group forming. But we also see a social lag in acceptance and adoption of these types of technologies. Groups are natively conservative, and in the “reply-all” instance, all members need to be participating for the technology to really spread its wings.

Shirky compares the Internet today to the example of an internal combustion engine. We don’t think about the car engine when we drive to the grocery store, but the engine is crucial for the event to take place. The same thing happens with the Internet now. Clay says, “once the technology has sunk deep enough into the culture the social effects that get built on it require the technology and aren’t about the technology.

Clay then breaks down group development online into 4 different categories: sharing, conversation, collaboration, and collective action. Sharing is demonstrated through social networking sites like del.icio.us, where social effects are almost an afterthought. In this “me first” collaboration world, users discover others they have things in common with only after they’ve done their own work. Conversation can be represented through the development of communities of practice, such as the HDR photography groups on Flickr. Shirky says that the group took a mere 3 months to explode, and that through conversation and posting of tips about HDR, the group can get better together. Furthermore, additional value is created by introducing users to one another. Collaboration occurs in such groups like anime translating and dubbing. Clay observes that this group started writing software to subtitle the anime they liked because those titles were otherwise unavailable. This process required teamwork, and there was no obvious commercial motivation behind it–fans just wanted to increase the overall accessibility of anime. Collective action is the future of the Internet, Shirky says, but we are getting a taste of its potential today. He explains an example of collective petitioning to enact an air passenger bill of rights after travelers were stranded on a Detroit tarmac for 7+ hours. While the social support required to sustain this type of action is high, we see that citizens are using new communication tools to co-opt traditional media for collaboration and synchronizing once-disparate groups for action. Now, the Internet can help people achieve a shared goal, and not just provide information.

Fascinating stuff. Once again thanks to Berkman Center for inviting such interesting thinkers and for providing the podcasts/webcasts so quickly! Check out Shirky’s “Here Comes Everybody” blog. Video at MediaBerkman.

Thursday night Larry Sanger of the expert-driven wiki encyclopedia Citizendium spoke at Eastern Michigan University. Larry was a co-founder of Wikipedia, but left the due to ideological differences about how the project should be maintained.

Sanger began by addressing his main concerns that web 2.0 hasn’t solved yet: high quality and high relevance. Even though there are 1.2 billion people online now, he argued that simply increasing the number of people in an online community won’t solve these problems. He claims that Citizendium can work to address these concerns in the following ways:

By finding meaningful roles for experts

Larry says that the intelligent use of experts in collaborative projects like Citizendium can help improve the quality of the output. He says that experts can be chosen online the same way they are chosen offline–through a demonstration of knowledge on a particular subject. He said that in order to get contributions from experts (like we see in really good Wikipedia articles, he admits), we need to make sure experts feel comfortable working within an open system. Finally, he argues that just because great things can be created without expert advice (Wikipedia) doesn’t mean that greater things can’t be created with expert input.

By requiring contributors to use their real-world identities

Sanger says that it is sensible to require real names for membership in some, but not all Internet communities. He does agree that there are legitimate privacy concerns that sometimes require anonymity. However, he argues that attaching real names to contributions (like is done on Citizendium) improves credibility, makes effective rules enforcement possible, and makes people behave better.

By establishing a rule of law by committing contributors to a social contract that makes them full partners in the project

Larry suggests that online communities should adopt constitutions just like offline communities do–he says this might take the form of some sort of online bill of rights. By doing this, he claims that contributors are compelled to agree to the rules as a condition of participation, thus giving them a more tangible and viable stake in its governance.

Sanger ran through the current stats on Citizendium, reporting that the site now has about 2000 authors and 250 expert editors. He says that there are over 5200 articles under development and that 5 million words were added in its first year (which he claims is over 6x larger than Wikipedia after its first year).

I entered the talk a skeptic of Citizendium, and remain one still. I was waiting to hear two things from Larry–of which I only got in scattered pieces: (1) Citizendium works to provide a better and more accurate resource than Wikipedia and (2) the process by which Citizendium is built is better.

I feel that neither of these pieces are particularly well-addressed. Instead, Citizendium seems like a reactionary project that only seems to address the thugs on Wikipedia. Larry says that the Citizendium community is a “friendly open country fair” while the Wikipedia community continues as a “street fight between rival gangs.” Is this really the point to Citizendium–to make collaborative knowledge production more civil? I agree, there are problems on Wikipedia–defacing and deletion of articles, skewed political rants, plain cruft. But these things can and will be worked out. Articles can be flagged for non-neutrality, incompleteness, and spam. As is central to the wiki form, more information and more transparency can work to help solve these problems. While there are some problems to an anonymous system, Sanger also ignores the enormous benefits to an informal system of usernames and identities. Just because I don’t use my real name on Wikipedia doesn’t mean a useful and positive community has not emerged–and one that respects individual users, develops social norms and fosters positive community relationships.

Citizendium attempts to use revolutionary tools like the wiki, but misses the mark because it relies on old frameworks for knowledge production and maintenance. When Larry speaks about his staff reviewing experts’ CVs before they’re accepted as Citizendium contributors, I can’t begin to understand how this process will be neutral, equitable, scalable, or particularly desirable. Sanger doesn’t see this process as carrying an prohibitively high barrier to contribute, but how can we ignore it?

Larry concludes, “Wikipedia will always be disrespectful of expertise.” I do not agree, and I think Wikipedia will continue to lead the way as a truly remarkable, inviting and accurate collaborative knowledge community.