Friday, July 1, 2011

Here Comes Everybody by Clay Shirky

Here Comes Everybody ("HCE") is a nickname given to Humphrey Chimpden Earwicker in James Joyce's novel Finnegans Wake. HCE is referred to by literally thousands of names throughout Joyce's book. Shirky has adopted it for this seminal work (2008) on the effect of the Internet on group dynamics and traditional business models and the implications for new economies. Shirky is ambitious in his examination of the tectonic shifts impacting communications and media. My approach is to paraphrase the ideas in the first 5 chapters, chapter 7, and the epilogue. Before we start, note that Rosenbloom in The New York Times (6/26/11) comments that Shirky hardlly uses Twitter himself.

It Takes a Village to Find a Phone. Shirky uses an intriguing example of web based methods to retrieve a phone lost in a taxi cab to demonstrate the power of the Internet in solving otherwise intractable problems. This story illustrates how the Web enables groups to assemble and cooperate. The story bears repeating. So Ivanna leaves her Sidekick phone in a NYC taxi, giving up, she buys a new phone. Lo and behold the photos taken on her old phone were being transferred to her new phone, thereby revealing the face and email address of a girl in Queens. Ivanna's techie friend, Evan, emailed Sasha requesting return of the phone. Sasha and boyfriend sent racist invectives, so Evan decided to take the story public. He created a "StolenSidekick" webpage with Sasha's photos and description of the events and added it to his personal website. Friends soon found Sasha on MySpace. An NYPD officer saw the site and contacted Evan to file a claim. Sasha's brother, Luis, Military Police member, contacted Evan to say Sasha had purchased phone from a cabbie, contradicting Sasha's earlier account that her brother found the phone. By now, Evan's story appeared on Digg, a collaborative news website. Soon, the nation's media picked up the story. Someone identified Sasha's home and posted a drive-by video. Now members of Luis' MP unit took notice and became concerned that he was threatening Ivanna. Evan now set up a bulletin board, hosting thousands of members. Evan and Ivanna filed a police report, who classified phone as lost rather than stolen, meaning no action would be taken. NYC Government officials got involved and explained how the complaint could be amended. The public airing of the NYPD's refusal to treat the case as theft generated public outcries. The NYPD caved and arrested Sasha, a 16-year old from Corona. After 10 days, Ivanna had her phone back.

This is a story about the power of leveraging a massive international audience of millions. Thanks to the Web, the cost of publishing globally has collapsed, due to the tools and social structure now in place. The story went from local to international almost immediately. Without the audience, the NYPD would not have changed the complaint. For Sasha, the recovery of the phone was not the only loss she suffered, since Evan's bulletin board named and shamed her on a global platform. This is also a story about the architecture of participation (p. 17). Evan was able to avail himself of capabilities previously reserved for professionals. He told a story without being a journalist, he found Sasha without being a detective. The Web makes it easier for unsupervised groups to self-assemble without requiring formal management (and incurred overhead seen in hierarchical organizations).

Sharing Anchors Community. What is the chance that in a group of 36 people, no two people share a birthday ? With 36 people and 365 birthdays, a 1/10 chance seems reasonable. But in fact it is 8/10 ! This is called the Birthday Paradox (surprise really). Most people get the odds wrong for two reasons. They think in terms that one of the 36 shares their own birthday, which is 1/10. But in a group, you need to count links between people. Consider yourself with 2 other people, A and B. It's not 2 comparisons, but 3: you and A, you and B, and A with B. For 36 people, there are more than 600 pairs of birthdays. A count of "any two people" rises much faster than the number of people themselves.

This growing complexity applies to social settings as well. As a group grows in size, getting universal agreement becomes difficult, then impossible. Ever try to decide on a movie with a group ? This complexity carries over to organizations, hierarchical to render interactions at a minimum. Every transaction an organization undertakes requires time, attention, or money. Because of the high transaction costs, no institution can put all its energies into pursuing its mission. Nobel laureate Coase (1937) considered what would happen if we abandoned organizations, and transacted directly in the marketplace ? Coase reasoned the transaction costs would be too high.

Shirky looks at the photo file sharing site Flickr. Flickr lets the users label or tag their photos as away of arranging them. The photos are linked when two or more users share the same tag. It is vital that Flickr does not bear the cost of organizing the pictures. Nor does it have any obligation to broadcast the categories it has ! Flickr provides a platform for users to aggregate the photos themselves. Flickr reverses the old order of group activity, transforming "gather, then share" into "share, then gather." Flickr escaped the traditional management overhead costs by abandoning any hope of oversight.

Traditional managerial oversight is often illustrated by an org chart, which had its origins in the railroad industry where trains ran in both directions on a single track (2 tracks were too expensive). One gets pretty organized after a few collisions. Hierarchical management of track sections became the answer and lowered transaction costs. At some point an institution cannot grow anymore and remain functional, because the cost of management destroys the profit margin. This is the Coasean ceiling. But what if transaction costs collapse ? Inconceivable in Coase's era. What happens to tasks that aren't worth the cost of managerial oversight ? Until recently, they didn't happen. Think of Flickr. These activities now lie under a Coasean floor. They are valuable to someone but too expensive to be taken on in any institutional way, since they are not worth pursuing. Coasean logic dictates that large decreases i transaction costs create activities that can't be taken on by businesses, because there isn't enough payoff to support the cost of being an institution. Social tools provide an alternative: action by loosely structured groups, operating without managerial direction and outside the profit motive.

For 100 years, a big question has been whether any organizational task was best taken on by Government in a planned way, or by business competing in the market. There has been a universal assumption that people cannot self-assemble. But now we can operate below the Coasean floor. We can now see coverage of tsunamis, coups in Thailand, etc. as the beginning of intense experimentation with these tools. Sharing, cooperation, and collective action form rungs on a ladder of activities enabled by social tools.

Everyone is a Media Outlet. Shirky looks at the traditional role of professionals (gatekeepers) who possess a specialization to solve a problem in a world constrained with scarce resources. Collapse in communications costs have undermined media businesses that traditionally rely on running a printing press or record label. The commercial viability of media businesses involves providing solutions, so preservation of original problems becomes an economic imperative. The Web did not introduce a new competitor in the ecosystem, it introduced a new ecosystem. The future presented by the Internet is mass amateurization (mass professionalism is an oxymoron) of publishing and a switch from "Why publish this?" to "Why not?' What happens when costs of reproduction and distribution go away ? What happens when there's nothing unique about publishing anymore, because users can do it for themselves ? What happens when the basic link between newsworthiness and publication no longer holds ? The lack of trustworthiness in blogs is made up for in the volume of reporting.

Shirky relates an amusing anecdote on 15th century scribes. Only a tiny fraction of the populace could actually write. The scribe would handcopy fragile and decaying manuscripts as an essential service to refresh cultural memory. A scribe's function was indispensable as a bulwark against intellectual loss. At the end of the 1400s, Gutenberg's movable type undermined the scribe profession. The Protestant Reformation was not caused by Gutenberg's invention, but was possible only after Martin Luther could spread complaints against the Catholic Church and print Bibles in local languages. In 1492 Johannes Trithemius launched an impassioned defense of the scribal tradition. Ironically, in order to get the word out, he used a printing press, undermining his message (see image of his book In Defence of Scribes). It is interesting to think that the subsequent spread of literacy was an early example of mass amateurization (p. 79).

In 2006, New York Times reporter Judith Miller was jailed after refusing to reveal her sources in an ongoing federal investigation. This case created a great deal of unease about journalistic privilege. Federal law has no equivalent to state laws (in 49 states) that provide a shield. Members of Congress introduced a bill to provide a federal shield law. It was unanticipated that it would get hung up over who is eligible as a journalist. Definitions of journalist was easy in an era of scarce resources, where journalists were rare. Now, anyone can become a publisher, so anyone can be a journalist. Given that, journalistic privilege becomes a loophole too large to be borne by society. Shirky fails to answer how the federal case was or was not resolved. This conundrum also applies to photography and the mass amateurization of photographers, i.e anyone with a camera phone. These amateur photos may be viewed by millions with no money changing hands. Absent the scarcity that made publishing such a serious-minded pursuit, the written word no longer has a special value in and if itself.

Publish, Then Filter. Personal communications and publishing now blend into each other. AS a result, the old pattern of professional filtering of the good from the mediocre before publication has yielded to social filtering, after the fact. User-generated content is a term used for amateur raw media that users create and share with one another, with no professional oversight. Most user-generated content is not content at all, any more tan a phone call to a family member generates family-generated content. Besides, people don't want professionally produced content for community matters. Shirky considers the amusing case of playing a recorded "Happy Birthday " on the stereo, instead of personally singing it off-key, likely preferred.

Facebook and MySpace are home to millions of users, but most users have an audience of a dozen or less. It is tempting to think of this as a failure, but user-generated content is only intended for a small audience. The posts seem inane because they aren't intended for us, but rather a community that has a social density that large audiences lack.

Communications media (phone, fax, telegrams) are designed for 2-way conversations. Broadcast media is between one sender and many recipients, a one-to-many pattern. Social tools now enable many-to-many conversations that are 2-way. The distinction between communications and broadcast is evaporating. A driving force behind user-generated content is that conversation is no longer limited to social cul-de-sacs like the phone.

While the Web makes interactivity technologically feasible, famous participants promulgate an imbalance between inbound and outbound attention. For someone to be famous, they have to be unable to reciprocate. Removal of technological limits has been replaced by social ones. Oprah has email, but if it became public, it would become useless. Even when a medium is 2-way, the most popular practitioners will be forced into a 1-way pattern. Oprah cannot hope to read thousands of blogs pointed towards her.

Because publishing is hard and expensive, the written word comes with an implicit promise: someone besides the writer thought this was worth reading. The converse of this effect explains our skepticism about self-published books. Mass amateurization has created an enormous filtering problem. Mass amateurization of publishing has invoked mass amateurization of filtering. We have lost the clean distinctions between communications media and broadcast media.

Communications tools don't get socially interesting until they get technologically boring. The invention of a tool doesn't create change; it has to have been around enough that most of society is using it. It's when a technology becomes normal, then ubiquitous, and finally so pervasive as to become invisible, that profound changes happen (p. 105).

Personal Motivation Meets Collaborative Production. Wikipedia is the best known example of distributed collaboration. Founded in 2001, it was an offshoot of the failed Nupedia, an encyclopedia to be written, reviewed and managed by experts volunteering their time. The first wiki was created in 1995 (Hawaiian for "quick"), which was a daring user-editable website. A wiki allows reader/writer crossover. A test wiki was set up on Nupedia, but the BoD objected, so was founded. In 2007, over 2 million entries were achieved. Why is it reliable? Because Wikipedia's self-correction process is so robust. Wikipedia has transcended the traditional functions of an encyclopedia, becoming a coordination tool for gathering and distributing information quickly. Within minutes of the 7 July 2005 London bombings, a Wikipedia page was set up and over 1,000 edits were added in the first 4 hours.

Without management, Wikipedia relies on spontaneous divisions of labor. Once an article is started, it attracts readers. Every edit is itself provisional, as is human knowledge, the articles are never finished. Wikipedia's contributors work on what they want to, when they want to, avoiding traditional obligations of workers in an institution. Wiki does not need to make sure its contributors are competent, productive, or even showing up. Nupedia failed because it was an expert-driven system. Mass amateurization removes the technological obstacles to participation. Interestingly, equality of participation does not follow, but obeys a power law distribution, which tracks the imbalance of participation. Wiki articles with hundreds of contributors typically have a few that do the most work. Apparently, the imbalance drives large social systems rather than damaging them. The "average" user is nonexistent (implying "bell curve" distributions). The 80/20 rule (Pareto's Law) applies to a store making 80% of revenue from 20% of inventory. This is the basis of long tail economics, where at Amazon, most items don't sell well, but in aggregate they sell huge, brick&mortar stores cannot afford the overhead to stock them. In a power law system, most participants are below average. [If Bill Gates walks into a bar, everyone becomes a millionaire on average!]. To understand the creation of a Wiki article, you can't look for a representative contributor, you have to concentrate on the behavior of the collective.

Faster and Faster. Collective Action is explored using the protests against the GDR in Leipzig in 1989. Gradually growing in numbers, they took advantage of an information cascade. The 3 levels of shared awareness were key - (1) when everybody knows something, (2) when everybody knows that everybody knows, and (3) when everybody knows that everybody knows that everybody knows. Shirky also talks about the rise of flash mobs and the ice cream events in Minsk, Belarus. Effectively, these flash mobs are not planned, but coordinated at the last minute using social tools. An historical example was the Nazi's use if radios in tanks, which gave them superiority over France. Mobile phones now provide a shift away from advance planning. People now make less definite plans - they can call at the last minute to make arrangements.

Evan Williams developed a website that would take text that a user entered into a form, and post it onto a webpage, known as Blogger, later acquired by Google. That social tool enables this blog you are reading. He then focused on mobile phones and created Twitter. Twitter was first used in a political activist context to make sure that Egyptian activists successfully passed through borders or checkpoints.

Epilogue. The epilogue is interesting as it provides a springboard for the social tools that have come to fruition in the Arab Spring with the ouster of Ben Ali from Tunisia (14JAN2011) and Hosni Mubarak from Egypt (11FEB2011). Shirky introduces the concept of "net value." One has to wonder if this was the inspiration for Evgeney Morozov titling his book Net Delusion. Regardless, Shirky makes the case that increased flexibility and power for group action will have more good effects than bad ones, making the current changes positive. The Abbot of Sponheim never understood this tradeoff - he was operating under the "lump of labor" fallacy, meaning any labor-saving device must make society worse off. He never stopped to think of new jobs in the printing press industry. One weakness of net value arguments recognizes that the good and bad changes are incommensurable, which is to say that the value of new sources of knowledge like Wikipedia cannot be measured against the increased resilience of networked terrorist groups. Further, when a real revolution is going on, net value is problematic. Societies before and after revolution are too different to be readily compared. Regardless of incommensurability, the ability of people to do and say as they like is inarguably good.

Shirky cannot resist a final anecdote about Venetian printer Aldus Manutius, who published a translation of Virgil's works in 1502. Manutius chose to publish in the octavo size, small enough to fit into saddlebags and enable easy transport. This was a small revolution, allowing the book to also shrink in cost. In an echo of the salacious nature of new content in many modern media (after all, pornography drove VHS), Manutius published the Hypnerotomachia, a contemporary novel with erotic passages, creating a market for new fiction. This certainly hints at audio downsizing with MP3, creating new markets for teenagers who live the principle of "good enough is good enough."

No comments:

Post a Comment