Do Online Information Cascades Translate to Crowds?

Clay Shirky brought wide-spread prominence to “information cascades” with his 2008 book “Here Comes Everybody” (which I “reviewed” here). At its most simplistic, an information cascade occurs when a number of people observing a group participating in an action and decide, independently, to join in. The most often cited example is of a protest or similar action, but it is also similar to the concept of innovation diffusion, which I’ve previously discussed.

Much of the research on cascades occurred before, and therefore out of the context of, the rise in prominence of the internet. Shirky attempts to connect cascades to the internet in his book, citing the example of mid-2000s Belorussian protests. But recently, I’ve seen some examples that indicate that perhaps culture has not developed to the point where the internet is an effective tool for information cascades to create crowds. The first, is noted by Evgeney Morozov in a recent Prospect magazine cover story:

It’s not hard to see how the internet might amplify information cascades and so strengthen the position of activists. The point is made most famously by the American web guru Clay Shirky. He is a darling of the social media world, a consultant for government, corporate and philanthropic bodies, and a source for reporters seeking quotes on how the internet is changing protest. He is also the man most responsible for the intellectual confusion over the political role of the internet. Shirky adapted Lohmann’s theories for the age of MySpace in his bestseller Here Comes Everybody (2008). The major lesson he drew from Leipzig is that people should “protest in ways that the state was unlikely to interfere with, and distribute evidence of their actions widely.” Why? Protesters are in a win-win situation: “If the state didn’t react, the documentation would serve as evidence that the protesting was safe. If the state did react, then the documentation of the crackdown could be used to spur an international outcry.”

But the truth is often different. In Belarus, most fence-sitters watched the state’s response and, acting rationally, went searching for higher fences. In Iran this year, the famous photograph of Neda Agha-Soltan, murdered in the streets, went viral and became a symbol of the “green revolution.” Whether it encouraged any fence-sitters is much less obvious.

Information cascades often fail to translate into crowds, even without state fear-mongering. Last year’s anti-Farc protests in Colombia—aided by Facebook—attracted huge crowds. But this year’s anti-Chávez protests did not, although they were organised by the same group using the same methods. The aim was for 50m people to rally worldwide but only a few thousand turned up. The same has been true when people have tried to organise protests in Azerbaijan and Russia.

Now, two very important points: He is absolutely correct about Belarus, and Shirky even concedes this point in a follow up to the Prospect piece. Iran, on the other hand, is much less clear. The Ashura events that began just after Christmas. So, Morozov’s points seem to be a coin flip, but he has others as well, as this is one of his much revisited sticking points.

But another event, far less serious, came to my attention and represents a similar inability for information cascades to translate into action or crowds. This is the case of “Ronna and Beverly” a comedy show first optioned by Showtime and then canceled. The shows stars were apparently able to convince Showtime to air the pilot program only, and then initiated a viral/web campaign to round up support for the program, in hopes of striking some sort of long term deal. The ratings numbers for the pilot aren’t in yet (too my knowledge), but the campaign did round up a large amount of celebrity support on Twitter and elsewhere on the web. However, Mashable’s Christina Warren points out that there is dubious precedent for this type of campaign resulting in any particular action.

Although social media and online campaigns have had an impact on many other mediums, television has remained an elusive nut to crack. Campaigns to get followers to tune into television premieres have consistently failed (or failed to sustain themselves after the initial flurry is over) and online-specific campaigns to take a show to the small screen have been equally unsuccessful.

In 2005, Scrubs creator Bill Lawrence — as well as Scrubs and Family Guy writers Neil Goldman and Garrett Donovan — created a pilot for what was then The WB network called Nobody’s Watching. The WB declined to pick up the show, but undeterred, the pilot was leaked to YouTube in June 2006.

The show became a YouTube sensation, with coverage across industry magazines, The New York Times and other outlets. In fact, the initial buzz was so strong that there seemed to be promise and confirmation that the show would get picked up by NBC. Despite additional webisodes, plans for a full pickup never actually materialized and the actors and writers all moved on to other projects.

Likewise, when TV veterans Marshall Herkovitz and Edward Zwick’s TV show Quarterlife — which was developed alongside a social networking site of the same name — couldn’t get network pickup in 2007, it aired on YouTube and MySpace instead. Despite relatively high totals for the webisodes, when NBC actually picked up the show in February 2008, the ratings were the worst for that time-period on the network in more than 17 years.

So, what we have is two completely disparate events, but both examples of information cascades. Both are unresolved, so it will be interesting to watch as they unfold for insight into the maturity of the internet as a conduit for crowd building. My own insight is that we probably aren’t just there yet, but if Iran erupts into a full fledged revolt and  (to an clearly lesser degree) “Ronna and Beverly” gets picked up by Showtime or another major network, we will have two  sturdy examples of this process in action.