Posted on July 2, 2007
Filed Under Web 2.0 Kool Aid |
The Web 2.0 phenomenon is often referenced alongside the concept of the “wisdom of the crowd,” which essentially states that under the right conditions, crowds are extremely intelligent and are more likely to come up with answers or solutions that are more accurate or better than could have been developed by any individual member (or subset of “experts”) found within the crowd. This concept has been popularized by the book “The Wisdom of Crowds: Why the Many Are Smarter Than the Few and How Collective Wisdom Shapes Business, Economies, Societies and Nations” by James Surowiecki.
Because many Web 2.0 services are driven by “crowds” and enable the aggregation of opinions, rankings, etc., it’s only natural that Web 2.0 and the wisdom of the crowd have become associated with each other. Many argue that the wisdom of the crowd, as implemented by Web 2.0 services, provides a superior means by which important data, information and media can be organized most effectively. For example, instead of relying on paid editors at a newspaper, many believe that the collective wisdom of an online community is better capable of determining what news is most important.
As I’ve read posts and comments throughout the blogosphere that relate to this issue, it’s become fairly obvious to me that quite a large portion of the people who espouse Web 2.0’s role in making the wisdom of the crowd part of our daily reality have not even read Surowiecki’s book. This is very problematic, because Surowiecki himself logically establishes that a crowd can only be wise if a number of elements are in place. Without these elements in place, you are more likely to have a stupid crowd (or mob).
As per Web 2.0 wisdom of the crowd darling Wikipedia, the four elements Surowiecki believes are prerequisites for a smart crowd are:
- Diversity of opinion. Each person should have private information even if it’s just an eccentric interpretation of the known facts.
- Independence. People’s opinions aren’t determined by the opinions of those around them.
- Decentralization. People are able to specialize and draw on local knowledge.
- Aggregation. Some mechanism exists for turning private judgments into a collective decision.
When crowds are too homogenous, centralized, divided, imitative and/or emotional, their potential to be sources of wisdom is severely compromised. Thus, any service attempting to create a wise crowd truly needs to ensure that Suroweicki’s four elements are in place.
When I look at Web 2.0 services often mentioned alongside the wisdom of the crowd concept, I can’t help but notice that most fail to meet the criteria for creating a wise crowd. In this post, I will analyze in-depth one of the Web 2.0 services most commonly associated with the wisdom of the crowd concept: Digg.
Digg: Wise Crowd or Stupid Mob?
One of the services most commonly associated with the wisdom of the crowd is Digg. Digg’s founder and “Chief Architect” Kevin Rose has stated publicly that Digg harnesses the wisdom of the crowd, and the description of the service is in line with this:
Digg is a place for people to discover and share content from anywhere on the web. From the biggest online destinations to the most obscure blog, Digg surfaces the best stuff as voted on by our users. You won’t find editors at Digg — we’re here to provide a place where people can collectively determine the value of content and we’re changing the way people consume information online.
How do we do this? Everything on Digg — from news to videos to images to Podcasts — is submitted by our community (that would be you). Once something is submitted, other people see it and Digg what they like best. If your submission rocks and receives enough Diggs, it is promoted to the front page for the millions of our visitors to see.
This all sounds great in theory, but an analysis of whether Digg truly incorporates the four elements Surowiecki believes are prerequisites for a smart crowd leave a lot of questions about whether Digg is anything but wisdom of the crowd vaporware.
Diversity of opinion.
Is Digg’s audience truly diverse? That is, does it reflect a wide cross-section of society and variation in opinion?
According to a September 15, 2006 , “Recent published Digg demographics indicate that the Digg community is 94% male and generally twenty or thirty something techies earning $75,000 or more.” While I wouldn’t dismiss the possibility that Digg’s demographic makeup has changed somewhat since then, it’s fairly well-known that Digg has a “techie” leaning. A cursory look at the top stories on Digg at any given moment typically confirms that this seems like a reasonable claim. As such, it is obvious that Digg’s userbase is not only not diverse, but actually quite homogenous.
Conclusion: Digg does not have diversity of opinion.
Does Digg offer a service in which the opinions of individuals aren’t determined by the opinions of those around them?
Digg has repeatedly faced criticisms that a small cabal of its users collude to promote stories they like to the homepage and to bury stories they don’t like. A look at the top Digg users in July 2006 (before Digg stopped publishing this data), revealed that the top 100 Digg users controlled 56% of Digg’s homepage content, and 20 Digg users actually controlled about 20% of Digg’s homepage content. This, at the very least, casts serious doubt on the independence Digg provides. Although Digg no longer publishes this data, it is available here and seems to indicate that not much has changed over the past year.
It is also important to consider the following:
- Because Digg enables users to invite others and add users as “friends,” independence is directly compromised by virtue of the service’s structure. Enabling connections to friends obviously supports the creation of “special interest groups” and introduces the possibility of peer pressure.
- A number of businesses, such as Subvert and Profit, claim to have achieved success in manipulating Digg and are commercializing their ability to do so.
- Because Digg enables individuals to see what other members of the crowd are doing, it promotes information cascade, in which the actions of others influence the actions of an individual. Information cascade is a key component of groupthink and herd mentality.
Conclusion: Digg does not provide the requisite level of independence necessary for a smart crowd.
Does Digg create a decentralized platform where the people closest to the problem are the ones who solve it and where no individual or group of individuals “at the top” are dictating the group’s judgment?
As noted above, a small cabal of Digg users appear to exercise immense control over what content is promoted to the homepage and what content gets buried. By virtue of the fact that this cabal exercises such control, it logically follows that in many cases the outcomes on Digg are the result of a more centralized process instead of a completely decentralized one. The Digg cabal is successfully able to promote content that is in line with its own interests and beliefs while marginalizing content that is not in line with its own interests and beliefs, regardless of whether or not the cabal has local knowledge that puts it closest to the problem (i.e. best able to provide some key insights on the value of a specific piece of content).
When we speak of decentralization, it is also rather interesting to note that Digg’s founder, Kevin Rose, has a “popular ratio” of 100%. He is the only such user on the service with a 100% “popular ratio” and this essentially means that every single piece of content he has submitted has been made “popular.” This is no different than if the middle managers at a company always went along with the suggestions of the CEO. While this does not necessarily mean that there is no decentralization (perhaps the middle managers evaluated the suggestions independently and came to the same conclusions privately), it is a red flag that hints at a flawed system.
Conclusion: When all the available data is taken into consideration, it is fairly obvious that Digg does not appear to be a decentralized platform.
Does Digg provide a mechanism by which the private judgments of its users are synthesized into a collective judgment?
On this point, Digg does succeed, however, as noted above, the mechanism by which individual judgments are collected does not make this a private event and thus promotes information cascade. For instance, by virtue of the fact that certain content will be promoted to the homepage and made more visible, it is likely to influence the subsequent “votes” of other individuals.
Digg meets only one of the four prerequisites for fostering a wise crowd, and the one prerequisite that it does meet is not implemented in a fashion that attempts to reduce or eliminate the problems of information cascade. If we take all of this into account, the only logical conclusion is that it’s disingenuous for Web 2.0 commentators to intelligently use Digg as an example of how the wisdom of the crowd is being harnessed by Web 2.0 services.
In a second post, I’ll provide a high-level evaluation of whether other popular Web 2.0 services meet Surowiecki’s requirements or not, and in a third post, I’ll address some of the challenges in successfully implementing an online service that is truly capable of creating a wise crowd.Print This Post