Posts Tagged ‘wikipedia’

Take any crisis in the past 10 years and there has been an increasing amount of digital content created immediately during and after the crisis events. For events like the Mumbai terror attacks in November, the amount of information is enormous and lasts multiple days. For shorter, more localized events like US Air Flight 1549 crashing in the Hudson, the amount is understandably lower. But in either case, pervasive new media allows citizens and professionals alike to create massive amounts of data. Sifting through and making sense of it all can quickly overwhelm an individual person or news organization.

Recognizing this, Ushahidi, the brilliant crisis mapping application, is approaching the question of how to filter through the information overload. Their simple approach is to “crowdsource the filter.” Writing on their blog, Erik Hersman calls this project “swift river” and will allow connected individuals to “go and rate the information as it comes in… where the more people you have weighing in on any specific data point raises the probability of finding the right answer. The information with greater veracity is highlighted and bubbles to the top, weighted also by proximity, severity and category of the incident.”

According to Erik, the prototype has successfully filtered a large amount of data and can combine experts and amateurs. When I first read this, I was understandably excited to see this in practice. During the Mumbai attacks, I watched as Gaurav Mishra chronicled the events by scanning blogs, YouTube, Flickr, Twitter and the MSM in order to synthesize and explain the story to his audience (in classic bridge-blogger style). His incredible public service kept me glued to my screen over Thanksgiving break. Amid his heroic effort, Gaurav occasionally mentioned to his appreciative audience how exhausted he was, and I can only imagine how much work it took. A reliable “Swift River” style tool would have allowed Gaurav to grab a couple more hours of sleep, knowing that the important truths would rise to the top. A look at the mock-up they provided shows a bit more what they imagine – individuals transparently describing their sources and adding “facts” that they trust to be true.

A mock-up of Project Swift River

No news consumers are completely autonomous; we all rely on some sort of filter. There is a spectrum from the obsessive consumers whose filter is minimal – reading and watching a ton of first-hand reports from Twitter, Flickr, etc. – to the mainstream consumers who get their news from CNN or even water cooler conversations.

My belief is that there is a sweet-spot for this spectrum of signal-to-noise filters – somewhere between (1) the exhausting work of reading all the information/analysis coming out of a crisis and (2) the lackadaisical, third-hand accounts which are colored by personal bias and memory.

The Spectrum of News Sources and Potential Downfalls

No one’s crisis news intake will ever be completely autonomous – even eye-witnesses only experience so much. But filters, especially the unreliable ones that exist in our world, can act like people in the children’s game “telephone” where each node can obfuscate the truth. Although news sources also add analysis, insight and context, my hunch is that during and immediately after a crisis, what matters most are verifiable facts. Ushahidi’s idea seems like a great way to find the sweet-spot for crisis reporting.

Although it will have to answer the same questions that were sneered at Digg and Wikipedia – who will trust a bunch of amateurs?! – a wealth of literature supports the ability of crowds to quickly and reliably reach consensus. Certainly there are some serious questions given the importance of the information that will be filtered (and I wish I had my copy of Infotopia to revisit how the wisdom of crowds can fail), but the open source nature of the project will allow for experimentation and revision that the New York Times or other news sources cannot do very adeptly.

I’ve been on a National Geographic Magazine binge; so many of their feature articles are fascinating investigations of society through the lens of science, conservation or travel. One, entitled The Genius of Swarms, takes a look at the ability of certain groups to be smarter than their individual components.

“Ants aren’t smart,” Gordon says. “Ant colonies are.” A colony can solve problems unthinkable for individual ants, such as finding the shortest path to the best food source, allocating workers to different tasks, or defending a territory from neighbors. As individuals, ants might be tiny dummies, but as colonies they respond quickly and effectively to their environment. They do it with something called swarm intelligence.

The tendency manifests itself in diverse species who seem to act as one even though no one is giving orders.

Even with half a million ants, a colony functions just fine with no management at all—at least none that we would recognize. It relies instead upon countless interactions between individual ants, each of which is following simple rules of thumb. Scientists describe such a system as self-organizing.

Instead, ant colonies, caribou herds and bee hives all rely on the signals and actions of their peers. Relying on local information and simple principles, groups exhibit the capacity to solve complex problems.

Businesses and governments are taking note. Firms with complicated logistical challenges, say delivering expensive, flammable gas in the least expensive and safe manner, are learning from ants and reaping the rewards of facsimile. The military is having success with “Centibots project, an investigation to see if as many as a hundred robots could collaborate on a mission.”

Wikipedia, everyone’s favorite example, has used swarm intelligence to create a resource of immense value and seen through biological understandings, it is clear why principles like NPOV have come to be enshrined in Wikipedian policy: they are the simple rules of thumb which help shape collective action.

The NGM article is really only a small piece of the growing literature on what James Surowiecki calls the “Wisdom of the Crowds” and what Cass Sunstein investigated in “Infotopia.” I’d be interested in seeing some research into the methods of governance for swarms: although they are distributed actions, what are the norms and principles which govern them? How do these come into being? How are wild mobs without reason replaced by thoughtful decision-making groups? The article passes briefly over this,

Crowds tend to be wise only if individual members act responsibly and make their own decisions. A group won’t be smart if its members imitate one another, slavishly follow fads, or wait for someone to tell them what to do. When a group is being intelligent, whether it’s made up of ants or attorneys, it relies on its members to do their own part. For those of us who sometimes wonder if it’s really worth recycling that extra bottle to lighten our impact on the planet, the bottom line is that our actions matter, even if we don’t see how.

But the answer to avoiding fads or market bubbles remain elusive.

Wikimedia, the non-profit which oversees Wikipedia, is holding its annual conference in Alexandria, Egypt. One of the debates making headlines is the discussion about how to facilitate the growth of the Arabic language Wikipedia which is the 29th largest Wikipedia – well behind the imaginary language Esperanto. This is an issue which has been persisting within the Wikipedia community for a number of years: how do you create an online, user-generated encyclopedia in a language whose speakers have little access to the Internet or free time to contribute?

I was first introduced to this issue by a question posed by Ethan Zuckerman at Berkman@10. He asked Jimmy Wales how the desperately poor could be expected to take the time to contribute to Wikipedia. His concern was perceptive and suggests that generating online content is a post-materialist pursuit. Only after attaining a certain level of material security can people afford to contribute to human knowledge. Others, too, have noticed this. At the 2nd Wikimania Conference in 2006, a Dutch Wikipedian, Kasper Souren, spoke about paying $1 for speakers of Bambara, a language of 3 million speakers in Mali, to contribute an article. Through this method he was able to amass more than 100 articles by the 2006 conference, but as of today, the Bambara Wikipedia only has 211 articles. I’m not sure if Mr. Souren is still soliciting articles in this manner or if he stopped, but, either way, I’m not convinced this is the way to create a viable native-language Wikipedia.

In The Gift, Lewis Hyde goes to great length to explain the ways in which the gift economy operate differently than the commodity economy. When money is introduced into a system, it undergoes substantial changes in ethic. The market efficiently allocates many things, but Wikipedia exists in a non-monetary economy. People contribute for many reasons, but the English language Wikiepdia did not reach more than 2 million articles because people wanted a new source of income. To convince Bambara speakers to contribute to Wikipedia, one needs to understand the motivations of a Wikipedian – reputation, principled support of universal access to knowledge or social.

In preparation for the 2006 Wikimania Conference, Ethan wrote an insightful post about the bilingual Wikipedians who decide to write in English instead of their native language. He notes that the tendency to edit the English Wikipedia probably comes from the desire to “influence perception and opinion on topics important to them by creating articles on political figures, important issues, issues of national or regional pride. And it makes sense to contribute to the wikipedia which has a broad audience and, therefore, a maximum chance of being read and influencing opinion.” I would tie this directly to the well-documented “reputation economy” which motivates open source programmers and many Wikipedians. As he points out, until a critical mass is reached in the Bambara Wikipedia, or any other smaller project, it may not be worth the time to edit it, even if you have the time, because your edits are falling trees in an empty forest. How, then, can Wikipedia make it a reputable task to edit smaller languages? Leader-boards or some other public displays of affection for editors? One might think that it comes down to paying for contributions until a critical mass is reached and it is worth people’s time to contribute. But what is that critical mass? And how would one go about stopping payments? The hypocrisy of paying for smaller encyclopedias until they reach a certain size would disillusion many would-be contributors.

Which brings us back to the Arabic Wikipedia. Cultural (Arab opposition to creating technology) and technical (lack of Arabic keyboards) reasons have been proposed to explain the lack of articles, but I don’t buy either. Sure, technicalities may cause incremental difficulty, but plenty of non-Roman languages have flourishing Wikipedias and the idea that Arabic-speakers are for some reason opposed to editing online encyclopedias is not very convincing.

Which makes an attempted solution so promising:

One volunteer, Abdel Rahman Hussein, an engineering student in Alexandria, pointed to the Facebook page he help administrate for the conference (593 members) and showed the events they had staged to spread the word long before the visitors had arrived. He said he had shown more than 1,000 people how to contribute to Wikipedia at his college in May. Bloggers have been enlisted, and some have added a banner in Arabic encouraging people to build up Arabic Wikipedia.

This takes advantage of the social and web-native nature of Wikipedia. By teaching people to edit Wikipedia and showing that it is socially acceptable, Abdel and the blogging community are providing an important impetus for the enabling of a community around the Arabic Wikipedia.

Update: See Ethan’s response to this post, where he considers the popularity of blogging in Egypt compared to Wikipedia.

The Starfish and the Spider is one of the better books I’ve read recently. It is a fun book packed with great examples about the “unstoppable power of leaderless organizations.” Much of what I am interested in recently are the networks which are increasingly important and prevalent; Ori Brafman and Rod Beckstrom provide a wonderful explanation of the benefits of decentralized networks.

The title of the book comes from the fact that if you cut off a spider’s head, it dies; however, if you cut a starfish in half, what you’ve done is create a two animals. In fact, some starfish can even grow from the removed leg of another. Their persistence comes from the fact that they do not have a centralized, hierarchical control system. In the past, starfish organizations did exist, but the Internet “is a breeding ground and launching pad for new starfish organizations.” As such, the lessons the two authors outline are important.

Brafman and Beckstrom are masters of finding perfect examples to illustrate their points. They easily tie together Native American history, the automobile industry and P2P networks to provide insight into the nature of starfish organizations. For example, why were the Aztecs so easily conquered by Cortez but the Apache tribe remained fiercely independent for hundreds of years? The answer, it turns out, is the same reason record labels are languishing and Napster is dead, only to be replaced by BitTorrent and eMule. The Aztecs were highly centralized with a capital city, chain of command and stationary lifestyle; in contrast, the Apache’s relied on disparate communities led by example and willing to live nomadically – with no supreme leader to kill or towns to raze, the Apaches proved much more adept at avoiding conquest. Similarly, whereas Napster decentralized much of music sharing, it relied on a central database which was vulnerable to copyright enforcement. Newer approaches, like BitTorrent are fully decentralized making it near impossible to shut down.

These open systems are also, understandably, more adept at evolving or mutating. One prime example of a starfish organization is Alcoholics Anonymous which doesn’t have a real headquarters or leader. AA is just a system and belief in the power of people to help each other overcome addiction. Though it began as a system for alcoholics, it spawned decentralized programs for other addicts including gambling and eating.

By being decentralized, starfish organizations take advantage of distributed knowledge. Wikipedia does this famously by allowing anyone to edit an encyclopedia page; Draper Fisher, a leading VC firm, does it by having dozens of offices around the world to hear pitches from entrepreneurs. Starfish organizations rely on the community they foster and the willingness to contribute to a cause.

One of the things I like most about “The Starfish and the Spider” is the authors’ decision to structure it, in parts, as a business book with clear “how-to” and lists of strategies. One section in which they do this is how to combat decentralization. Al Qaeda is one of prototypical starfishes: they are more an ideology than an organization. Would-be terrorists do not need Osama bin Laden’s approval to carry-out an attack. In this light, it makes even less sense to hunt the 9/11 masterminds while not combating the ideology that motivates terrorists. Brafman and Beckstrom say that there are three ways to combat starfishes, none of them are to further centralize yourself, even though that is the tendency (witness the rise of the Executive branch post-9/11 or the record labels in recent years). You can change the ideology of the decentralized adherents by giving them reasons to not attack the West (a future or respect, perhaps?), centralize them (as America did by providing Apaches with cattle and therefore property), or decentralize yourself.

The key, though, for decision-makers and organizers, is to find the hybrid form like Toyota or eBay. Both major corporations, Toyota relies on their much-flaunted production system which relies on outsiders for much of it and eBay has a decentralized network of sellers and buyers, but a central payment system.

At 210 pages of fun examples, plain English and smart-thinking, this book was great (I finished it the day it arrived). Though I found their discussion of the catalysts (people who empower decentralized networks) a bit tiresome, I cannot complain about the rest. “The Starfish and the Spider” will provide an important way of thinking about the world.

David Weinberger, in an early essay for Publius explained that “Rules are norms that have failed.” In his reasoning, the majority of human action is not governed by explicit rules; instead, tacit governance – norms, conventions and expectations – dictate the appropriate behavior in most cases. Where explicit governance is needed, the norm based approach has failed. Roads need explicit speed limits to avoid people’s tendency to cause accidents. “The overwhelming preponderance on the Net of tacit governance over explicit is a sign of the Net’s depth, importance, humanity, health and success.”

This tacit governance is connected to the civic technologies extolled by Zittrain. Those technologies, like Wikipedia or the Internet, which require constant care and effort to be successful, rely on mix of tacit and explicit governance. On the one hand, the Internet Engineering Task Force decides through “rough consensus,” often by humming. On the other, Wikipedia has an extensive list of rules and policies (of which one is to “ignore all rules”).

This mix of governance strategies hints at an effort to capture the civic ethic which allows these technologies to avoid formal, external institutional rule-making. Partly as a result of a technopanic over online porn, the Communications Decency Act was passed to regulate online speech. Because badware is so pervasive, McAfee and Symantec have a tidy business of combating it. Both of these are purely explicit governance which can have numerous complications.

How do we govern the Internet and technologies in either purely tacit manners or through a mix which minimizes explicit rules? I think it comes back to the civic ethic which can motivate heroes like Ghandi or just a simple Wikipedian who deletes a malicious edit. The question, then, is how to capture this civic ethic and expand it to new fields? This is more than a technological question, but by designing the tools and understanding the motivations of civic engagers, we can seek to expand this.