Rethinking Social Media

Today’s web is social, we all know that. But it may be more social than we think. Some platforms, like Facebook, are obviously social: we see the group of peers who are creating our experience through content creation or recommendations. Other platforms, like Google search, are “phantom social” – you don’t see the group of peers that have created your experience, but they are there nonetheless.

Let’s start with the obviously social: Facebook, Flickr, Tumblr, blogs, Twitter, Yelp, Reddit, Meetup, 4chan, YouTube…. Whether you are looking at a restaurant review, a photo, a video, or a post, you know who has written or created it. The creator has an identity, even if it is just an online identity. You are constantly aware that your experience is being created by your peers.

Phantom social applications like Google Search or Cleverbot are more tricky. It may seem that you are interacting with a computer, but the responses you receive are still created by your peers, you just don’t see them. In the case of Google Search, the ranking of a search result is determined by the number of incoming links. Who creates those all those links? Other people. Notice how Google can predict possible queries as you type? Again, those suggestions are gleaned from other people’s common searches. And instant search? You get the idea. Each Google Search is made possible by millions of user actions collected and crunched by Google algorithms, yet few of us think of a search as a social activity.

Cleverbot is an application that allows you to carry out a text chat with a computerized personality. It seems the archetypal anti-social online experience: talking to a computer. Yet Cleverbot is also phantom social. According to Wikipedia:

Cleverbot differs from traditional chatterbots in that the user is not holding a conversation with a bot that directly responds to entered text. Instead, when the user enters text, the algorithm selects previously entered phrases from its database of 20 million conversations. It has been claimed that, “talking to Cleverbot is a little like talking with the collective community of the internet.”

Then there are the hybrids, platforms that have some features where peers are visible and some features where peer input is hidden and you seem to be interacting with a machine. The classic example here is online shopping sites like Amazon and, which incorporate user reviews (visible peers) along with product recommendations and star ratings (phantom peers). All three are based on the input, buying, and browsing patterns of other users. All three are social.

What does this mean? It means the web is a lot more social than we think. We know that interacting with hundreds of friends on Facebook or thousands of followers on Twitter is social , but when we are interacting with millions through an algorithm we forget the social aspect. Whether through a email from Mom or a recommendation gleaned from millions of strangers, the web is becoming an ever more elegant medium for meaningful human interaction.

From Our Book: The Problem of Too Much Participation

NOTE: We’ve posted a free downloadable copy of our new book Digital Activism Decoded and on July 1st the paper version will go on sale at For the next month we’ll be posting more brief excerpts from all the chapters in the book. To learn more, visit our book page.

Today’s excerpt is by Tim Hwang, founder of the Web Ecology Project and ROFLCon. Looking at an example from Twitter, Tim describes how the ease of online participation can damage a digital activism campaign when the “signal” of a useful knowledge meme is lost in the “noise” of commentary and chatter, resulting in entropy (chaos). You can find Tim blogging and on Twitter.

…For one of the Web Ecology Project’s first reports, we conducted an overview of the Twitter conversation happening around Iran’s recent postelection crisis. Working with a set of more than two million tweets (short posts) produced by users, our study captured the universe of content created in the weeks leading up to and directly after the election. Rather than focus tightly on content produced by activists in Iran and around the world, our group attempted to enlarge the scope of the study by examining the broader cultural landscape of discussion around the topic. This included the activists, certainly, but also incorporated the commentary of celebrities, journalists, casual observers, and many others.

This examination revealed that the biggest issue facing some activists was not so much getting the word out—where an issue strikes a nerve with a large enough audience, content spreads easily. Instead, the threat came from the door to participating in the conversation being open to contributions from all sides, whether or not they are constructive or useful. Indeed, in an era of social platforms that are tightly networked, the bar to merely getting a message out from a circle of activists into the public sphere is much lower. Consequently, on increasingly popular platforms like Twitter, the period between when a conversation becomes widespread or trending and the point at which it becomes “polluted” with a high amount of noise in the form of spam and tangential conversation is rapidly shrinking.

This problem is particularly significant to the work of activists, since any goal beyond “fostering discussion” online will require that certain kinds of information be far more important to spread than others. Particularly where a campaign attempts to motivate people to real-world action, a massive, unfocused discussion around a topic might, in fact, hide or inhibit effective mobilization. This happened quickly in the case of the discussion around Iran—the initial voices of activists reporting on events from within the country became rapidly washed out in the noise of the commentary from a mostly U.S- and Europe-based group of reporters and celebrities (with some standout exceptions, including, most notably, @persiankiwi, an anonymous opposition Iranian activist reporting about events on the ground during the crisis). For the interested viewer, however, a casual glance at the stream of content emerging about the Iran election on Twitter would be overwhelming—impossible to keep up with and read, and most of it would not actually be from the activists on the ground.

This phenomenon is not confined to conversations about “serious” topics or political issues. A common occurrence on the Web is the flurry of reposting, commenting, and spam that springs up around a particular piece of popular content, which inhibits the ability of an interested reader to assess where, when, and why content is being created. Most important, without any particular intervention, the most relevant and actionable content often becomes increasingly obscured—a process that we call “memetic entropy.” The word “entropy” refers to the process of disorganization, “memetic” comes from the word “meme,” a transmittable unit of cultural content. Thus, memetic entropy refers to the observation that the dissemination of cultural content itself seems to engender an increasing disorganization of that very content, in this case because no limits on participation in dissemination are present, and, as a result, no way is available to clearly identify which user-generated content is most salient to the activist.

A simple but useful lesson for the online activist: build methods of filtration and curation into an overall strategy of awareness and mobilization. While a rapidly spreading discussion online might foster awareness, creating easy access to relevant information and ways to sift through information in a semi-stable way play a key role in sustaining attention. After all, the question isn’t just whether or not a group of activists can get the word out, but whether or not those activists get the proper information to the right people once the spotlight of attention is turned on them….

Proudly powered by WordPress
Theme: Esquire by Matthew Buchanan.