Emergence on facebook
From Physical Programming
“Netscape browsed the Web, Yahoo organized it, Google searched it, and now Facebook has made it social”
Seth Goldstein, co-founder of SocialMedia
Introduction
The phenomenon of Facebook has transformed the way we interact socially over the internet. Research and discussion on the net concerning Facebook is a huge industry, and the big capitalist question of how to make money from it is at the head of the debate. Many sites talk about the ‘ultra-application’ which is what developers term the application which will hook everybody in. It would be like starting a fad within a fad, an irresistible trend within Facebook.
With an understanding of graph theory and emergent behavior, I hope to write a Facebook application which taps into the theories of Malcolm Gladwell’s book The Tipping Point on how to predict social trends to create a popular, emergent Application.
The Tipping Point
Fads which were previously unfashionable suddenly take the world by storm, seemingly unexplainably. Malcolm Gladwell attempts to dissect these trends, fads, fashions and the flow of information through social networks. He does not address Facebook until the newly written Afterword.
Gladwell defines three causes of trend formation; The Law of the Few, The Stickiness Factor and the Power of Context. The Law of the Few describes 3 types of people, Connectors (people with lots of active connections to other people), Mavens(collectors of information) and Salesmen (charismatic sellers and encouragers. ‘Successful’ trends use all three of these types of people. Interestingly, these people can be profiled and identified on Facebook within an application, and ‘used’ to spread the flow of information.
The Law of Stickiness rules that trends need something to grab the attention. This is usually a minor detail which ‘sticks’ in each person’s mind and makes them take up the fad, and in turn become another point to introduce the fad to others. This minor detail is hard to define, is something unusual but not necessarily a major part of the fad, and is usually developed against regular advertising advice. It’s hard to pinpoint, and to a large extent seems to be a matter of chance
The Power of Context describes the time, place, method and general environment into which a fad is introduced. In a similar way to Buckminster Fuller, Gladwell suggests that fads which are engineered need to take on this context, and in a way developed for the context. Fads which succeed randomly are because the context and the fad randomly combine into a winning combination, which is why they are reasonably rare.
All three of these causes need to be taken into account when engineering a fad (though Gladwell does not detail this engineering side of the discussion). Facebook is a perfect platform to create a fad, and in itself is a fad. It allows networking so that all people can become Connectors, Mavens and Salesmen. It will be interesting to see if people’s real life social structures and roles break down in a social network due to the new cyber possibilities. In the new Afterword to The Tipping Point, Gladwell writes about the use of fads on the internet, but suggests that the information there is ephemeral and ubiquitous because there is a large mass of it and people are bombarded by it from all directions. For example in Facebook, the function of applications to flood friends of people who already have the application is a tool to spread the word – but people become desensitized to the requests and ignore them on mass, like SPAM.
Types of Applications
Reading no-mans-blog (ref 2), he splits the top 100 applications on Facebook (as ranked by adonomics – ref 3) into 3 sections: identity formation, phatic communication and other (expanded upon later). Out of all applications, Identity forming/displaying applications make up 42%. They are tools to display our character in the public realm. Self displaying characteristics make up 18%, while applications which ask a user’s friends to determine their character make up 24%. Phatic communication is the subject of 38% of the top 100 applications. Phatic means that people use the applications to communicate their presence or social being. It’s an act of being social without transmitting any ideas, just base information. An example of this is in Facebook is the action ‘to poke’ and in real life, the phrase ‘how are you’, which in passing is just a greeting, not a question of the persons health. These phatic communications can be divided into pokes, themes, gifts and seasonal hellos. Perhaps this popularity implies a move further away from personal communication like letters, and towards a status of ‘hey! I’m still here’. It can also be a form of social grooming, a translation of a physical presence and touching to a constant stream of simple electronic gestures.
Surprisingly, friend connection, communication and games make up the other 20% or so. I would have thought that connection, socializing and networking would be the main focus of Facebook. This shows how people really are living on Facebook: creating a personal profile of themselves and keeping in minimal but detattched ‘social grooming’ relationships with friends.
Writing an application within Facebook
Facebook assumes that developers have a server which can host php5 – which is similar to FBML, the programming script in which Facebook Applications are written. Facebook applications are written in FBML to make the appearance and use of all facebook applications the same or at least similar. Free software – ConTEXT – is available to edit the script. Facebook hosts a tutorial which goes through options within Facebook as well as with the php/FBML script. It seems quite complicated, and difficult for a novice to understand, let alone code in. Understanding Facebook, the limits of the API, the possibilities of ‘networking’ and the design of the system is probably where I should first focus. Perhaps by looking at the problem from a design perspective, instead of a development perspective I can come up with something different.
Limitations of the API
The method of retrieving information about people on facebook (like who someones friends are, or what their age is) is limited to various API functions. For example, an application can only view the friends of someone with that application, so not the friends of friends. These limitations need to be taken into consideration when designing an application.
What I want to write
I would like to create an application which utilizes the extreme connectivity of the Facebook social graph and develops emergent behaviour from it. This might mean asking people on facebook to add their friends to the application which then effects the application in some way. It might have some way of connecting people who don’t yet know each other but have many similar friends. It should be developed from the known popularities of facebook applications and engineered to be popular – to demonstrate the best function of an emergent system – to have many nodes.
Quotes on how to make your application successful:
• tap into the "social graph”
Senior platform manager Dave Morin (ref 7)
• Applications that augment or mimic existing features on Facebook - such as the wall (a space for writing messages) or a poke (a way for friends to say a quick hello) - are also more likely to take off
• death to an app… is complexity
• Any application that is more complicated than the most complicated feature in the core of Facebook will be penalized.
• [applications that] stumble on even the smallest bug are likely to become roadkill
San Francisco-based programmer Jesse Farmer (ref 7)
Even if someone creates a killer supergroup app the idea will probably be copied by facebook itself and added into every group on the site. Then the “killer” app won’t be so killer anymore. People have to think WAY out of the box to create something that will be successful and facebook won’t copy. Andrew
References
1. Summary of the Tipping Point http://www.wikisummaries.org/The_Tipping_Point
2. Dissection of facebook statistics
http://no-mans-blog.com/2007/11/facebook-applications-trends-report-1/
3. Facebook analytics. Compares different companies who are writing Facebook applications and how much they are making, lists the most used applications and their worth (via advertising I imagine), has a marketplace where people write applications and sell them; and Adonomics services which will grow your app ‘virally’ and then help you sell it (for a fee of course…). http://adonomics.com/
4. Info on the United Application Developers Alliance, which is a collective or company creating the most successful applications. A group of developers who seek to make money off their applications and help each other get better money etc. join if you have “either 10,000+ daily active users or 100,000+ installs”. Join and get an equity stake in the profits http://adonomics.com/uada.php
5. Data on UADA http://adonomics.com/company/The_UADA
6. Gives a quick history of Microsoft’s role in building windows based software then divides facebook applications in 10 categories to rate the best in each category. http://blog.adonomics.com/2008/02/22/building-the-social-suite-of-category-killer-apps-for-facebook/
7. CNN Money report on how to turn facebook into a money generator “the stickiest applications are those that tap into the "social graph." That's Zuckerberg's oft-quoted term for the web of connections between users and their friends. "Most apps are only interesting if there is much more content below that widget," Morin says. "It needs to take you someplace different, do something more."” 4 ways to make money (sell ads, attract sponsors, sell products, sell services) http://money.cnn.com/2007/08/22/technology/facebook_economy.biz2/index.htm
8. chinks in facebook’s armor, or ways it could be improved (or why its not worth $100 billion) http://blog.adonomics.com/2008/01/11/a-z-on-why-facebook-cant-be-worth-100-billion/
9. counter argument (by the same guy) on how it is worth $100 billion, and how. But more interestingly, evaluated the platform and facebook users http://blog.adonomics.com/2007/12/06/why-facebook-is-worth-100-billion/
10. Predictions on a new application to kill all others and rise to the top with the most people involved http://blog.adonomics.com/2007/10/12/is-supergroups-the-next-killer-app-for-facebook/
11. Tips from David McClure on how to create a successful facebook application http://www.techcrunch.com/2007/10/03/seven-steps-to-graphing-your-facebook-strategy/
12. Article concerning companies vying for space on facebook, and how to build on that platform http://www.technologyreview.com/Biztech/19122/?a=f
13. Anatomy of a facebook application http://www.yeeguy.com/cs377/fb_app_anatomy_20071001.pdf
14. Semantic web – where computers are agents on the net and process information without the need for human input. Writing pages in RDF instead of HTML allows them to be read by computers and therefore analyzed by them. http://en.wikipedia.org/wiki/Semantic_Web
15. friend of a friend organization. Maps social networks without the need for a central hub. THIS IS AN EMERGENT SYSTEM. Basically writes people’s contacts and ‘friends’ or edges to other vertices (nodes) in RDF to be read automatically by computers, making searching for certain people within the map easy. Also incorporates a site for making your own FOAF page in RDF. http://www.foaf-project.org/
16. Decoder of RDF files to make the files readable by computers readable by humans again! Basically displays the information about a person (in their FOAF information) in a way that makes sense to us. http://foaf-visualizer.org/
