Friendster and Publicly Articulated Social Networking A Social
Friendster and Publicly Articulated Social Networking, A Social Network Caught in the Web By: Mitch Lederman Date: 4/12/07 Shivnath Babu
What is Friendster? • Friendster – Its an online dating site utilizing social networks to encourage friend-of-friend connections. – Built under the assumption that friends-of-friends are more likely to be good dates than strangers – Built to compete with Match. com – Friendster only allows you to access those friends with four degrees – Friendster encourages users to join even if they are not looking for a dates, under the assumption that they know a wide variety of friends who are looking and would serve as a connecter and recommender.
Friendster’s major growth • Launched its public beta in the fall of 2002 – As of early January 2004, the site is still in beta and has amassed over five million registered accounts and is still growing
The value of the network • Friendster assumes that users will define their identity by their profile to ensure meaningful connections. – The users will see the value in connecting to actual friends. – Is the value of the network good? • Many Problems? – Friendster fails to recognize that publicly articulated social networks and identities are not identical to the private articulation. • Public identities are not the same as private ones
Problems cont… • Relationship indicators in Friendster are binary: Friend or not – No way to tell what the weight of the relationship is. Some list anyone as friends, some stick to a conservative definition, most list anyone they know and don’t dislike. – This means that people indicated as Friends even though the user does not know or trust them • Because of this weakness, the weight of a friend connection is often devalued because trust cannot be guaranteed. • Publicly articulated social networks disempower the person performing.
Presentation of self • Friendster Profile (Presenting themselves based on specific time and audience) – – – • Demographic information Interest and self-description Pictures Friend listings Testimonials Context is missing – Individual is constructing a profile for a potential date and also one must consider all the friends, colleagues, or other relations who might appear on the site. – Social Appropriateness • Truth – One is simply performing for the public, but in doing so, one obfuscates quirks that often make one interesting to a potential suitor. • Making it so confusing as to be difficult to understand behaviors that make one interest • Teachers fear the presence of their students on Friendster – Everyday activity we present different information depending on audience
Friendster as a site of connection • People use Friendster to connect to others for a variety of reasons. – Connect with people that they know, reconnect with long lost friends, and colleagues. (Individual connections) – Private elite clubs and pub gatherings – Memorials – Communication – Auctioning connections on EBay – Women advertise their porn sites by attracting potential clients – Fraudster profiles to deal drugs, using the bulletin board to announce events – Many are using Friendster for its intended purpose: dating. • Dating falls into three categories
Dating Categories • Hookups – Three to four degrees away • Direct Pestering – Look at friends’ friends and bug the intermediary about potential compatibility. • Familiar Strangers – Strangers that one sees regularly but never connects with. – Browsing site, users find people they often see out and look at their profile. – Then by that, they can send them a message or approach offline.
Fakesters • “Fake Personas” • Three forms of Fakesters – Cultural characters that represent shared reference points with which people can connect (God, George Bush, Tim Mc. Graw) – Community characters that represent external collections of people to help congregate known groups (Duke University, San Francisco, Burning Man) – Passing characters meant to be perceived as real (duplicates of people on the system) • Fake female character • “Fraudsters”
Fakester Dilemma • Company has never approved of this behavior (collapse network &devalue meaning of connections between people) • Most people love fakesters • Tension between company and users • Company outraged users by deleting fake profiles – “Fakester Revolution” – Site became less interesting when Fakesters removed – Is anything actually real on friendster?
Learning from Friendster • Major problem around publicly articulated information • Reshaped how groups of people verbally identify relationships • Importance of creative play in social interaction
Club Nexus • Stanford in the fall of 2001, reflection of the real world community structure • System to serve communication needs of Stanford online community • Send e-mail and invitations, chat, post events, buy and sell used goods, and search for people with similar interests. • Attracted over 2, 000 students early on. • How they connect people
User registration and data • 1 st step-enter name, e-mail addresses, birthdays, major, year in school, home country and state, phone number-etc. • 2 nd step-users asked to list their friends at Stanford- “buddies” • 3 rd step-list interests and hobbies • 4 th step-select adjectives to describe personalities, what they look for in friendship-etc. • Using data, they were able to deduce attributes contributing to the formation of friendships
Network Analysis • Nexus Net-large social network with 2, 469 users and 10, 119 links between them. • Number of buddies a user has is distributed unevenlymost users had just one buddy, some had dozens of friends, and one had more than a hundred. • Small world effect– distance between two users, measured in number of hops along the Nexus Net is only four on average. • Counterintuitive aspect: people tend to socialize in smaller cliques, yet they are separated by only a small number of hops. • Separated/far away/not connected to many other people because in a small clique. – How can be separated by only a small distance?
Properties of individual profiles • Z-score- (number observed)-(number expected)/(standard deviation) – Used z-scores to characterize the relationship between different attributes the users chose. Also, they indicate how likely it is to find a connection between two attributes by chance. • Ex-Funny and enjoy watching comedies • Personality and preferences (factors influencing friendships) – Used this analysis to find correlations between users personalities and preferences. – Described attractive=appearance is important – Described funny=sought laughter in relationships – Described weird= weird friends, spend time alone and at home – Described successful- activities, fulfilling commitments • Homophily
Properties of individual profiles cont. • Academic major and personality – Examined relationship between persons academic major and what adjectives they chose to describe themselves • Physics, math, engineer majors-nerdy stereotype, learning, weird. English majors-reading. Undeclared-doing anything exciting • Gender differences – Examined how gender influences personality and preferences. • Men-Football, war movies, sex, activities • Women-gymnastics, romance movies, trust, socialize
Association by similarity • Tendency of individuals sharing interests to associate with one another • Activities or interests shared by smaller subset of people showed stronger association ratios than very generic activities that could be enjoyed by many. – Ballroom dancing vs. partying – Duke Football vs. Football • Similarity and distance – Similarity with a friend decreases as distance between users increases – Higher likelihood we share a characteristic with a friends’ friend than we share it with someone four hops away.
Nexus Karma • By e-mail as a new feature-users who were ranked by three buddies were sent an e-mail to do the same. • Users could rank how trusty, nice, sexy, and cool their buddies were. – Some variably in scores given • Demonstrates a clear correspondence between the way that individuals perceive themselves and the way that they are perceived by others. – Described responsible-received higher trusty scores – Described attractive-higher sexy scores
How does this relate to Google? • How we interact with people socially on the social networks has a lot of information. – – Profiles Registration Friends Pictures • Google=make worlds information accessible. • Need to improve indexing and storing for all this information (not indexing images of me on facebook nor my information)-Google desktop is helping • Study on Club Nexus done by Orkut Buyukkokten helped him start Orkut. com, which is a social network associated with Google.
- Slides: 19