Social network sites such as Facebook facilitate the formation of new relationships and the maintenance of existing personal relations between friends and sharing of information through shared common interests, blogging via comments, applications, photo sharing and status updates (Boyd & Ellison, 2007).
During the course of this paper it will be shown how Facebook compromises user’s privacy by disclosing and sharing their personally identifiable and accurate data to a user’s contacts and other users on the network due to the inherent design weakness of the “Friending” mechanism. User’s privacy is also compromised by third party developer API’s in the form of applications and the new Connect Platform running on external websites. Privacy is also jeopardized from the insecure open default Privacy Settings and from users themselves because they do not understanding how the Privacy Settings function and their consequences. User’s also do not read nor understand the Privacy Policy detailing data collection of their profile data or the Terms of Service relating to intellectual rights that Facebook has of their profile data.
This paper will argue that these privacy design flaws inherent in the design of Facebook permit the disclosure of large amounts of personally private identifiable user data to other users, third parties and external websites and that the privacy design flaws are deliberate in order to facilitate the sharing of information thereby posing a massive threat to user’s privacy.
Facebook has an active user database of more than 400 million accounts, with 35 million users updating their profile pages 60 million times daily (Facebook Statistics, 2010). Personally identifiable data that is visible and accessible on user’s profiles consists of their name, status updates, wall comments, profile photo, photo albums, date of birth, phone number, email address, college name, residence, address, dating preferences, relationship status, political views and interests related to music, books, movies and their friends list (Govani & Pashley 2005; Gross & Acquisti 2005). If a Facebook user fills in their profile page, it will contain approximately 40 pieces of personally identifiable information (Grimmelmann, 2009).
According to a study conducted by Gross & Acquisti (2005), 89% of Facebook users provide their real name and that only 8% provided a fake name. The same study also showed that the profile photo uploaded by users was identifiable either partially or fully in 80% of the accounts surveyed. This means that approximately 320 million users provide personally identifiable and authentic data that is visible and accessible on their profile jeopardizing their privacy (Gross & Acquisti, 2005).
To facilitate information sharing, Facebook has created three basic levels of privacy that provide different levels of access to user’s profile data on their network. These groups are “Friends”, “Friends of Friends” and “Everyone” (Privacy on Facebook, 2010).
Personally identifiable and authentic data is made visible and accessible to a user’s “Friends”, “Friends of Friends”, “Everyone” including third parties, due to the inherent privacy weakness of the “Friending” mechanism Facebook utilizes to establish connections between users. Dwyer, Hiltz & Passerini (2007, p1) state that, “Members connect to others by sending a “friend” message, which must be accepted by the other party in order to establish a link. Friending” another member gives them access to your profile, adds them to your social network, and vice versa.”
The “Friending” mechanism is a great privacy concern for all users because by accepting a “friend” request from someone, you are by proxy permitting all that sender’s Friends access to your profile and personally identifiable information. Clay Shirky (cited in Grimmelmann, p38) has neatly summarized this predicament by stating that a, “‘[F]riend of a friend of a friend’ is pronounced ‘stranger.’” Herein lies the problem and privacy issue with the “Friending” mechanism as it exposes personally private and identifiable information to strangers beyond the immediate person you have just “Friended”.
According to Facebook Statistics (2010), the average Facebook user has 130 “Friends”, however up to 50% of people surveyed in the study by Govani & Pashley (2005, p6), have accepted “friend” requests from people they have met just once and a staggering 30% accepted requests from people they have never personally met.
A recent case demonstrating the privacy dangers of “Friending” was highlighted by Gloria Gladsden, a sociology professor at East Stroudsburg University. Gloria thought that some critical comments she made about her students on Facebook would be visible only to her “Friends” and family and remain private. Somehow, these comments were made visible to her “Friends of Friends” which included some of her students. The resulting controversy caused by her comments, forced the university administration to suspended her pending an investigation (Stripling 2010). “Friending” is a great way to connect to other users, but it is major privacy issue that makes user’s profiles and data not only accessible to actual friends but also to complete strangers.
User’s personally identifiable profile information can also be made public through the Facebook Application Platform Interface (API). Third party developers write applications that users then download and run on their account as a plug-in. These applications enhance a user’s experience of Facebook but they also allow the developers of these API’s access to their profile and personal information making it a severe privacy issue (Grimmelmann, 2009).
API’s and their developers have access to the following personal information on user’s profiles according to Facebook Platform (2009),
…your name, your profile picture, your gender, your birthday, your hometown location (city/state/country), your current location (city/state/country), your political view, your activities, your interests, your musical preferences, television shows in which you are interested, movies in which you are interested, books in which you are interested, your favorite quotes, your relationship status, your dating interests, your relationship interests, your network affiliations, your education history, your work history, your course information, copies of photos in your photo albums, metadata associated with your photo albums (e.g., time of upload, album name, comments on your photos, etc.), the total number of messages sent and/or received by you, the total number of unread messages in your in-box, the total number of “pokes” you have sent and/or received, the total number of wall posts on your Wall, a list of user IDs mapped to your friends, your social timeline, notifications that you have received from other applications, and events associated with your profile.
In addition to accessing this personally private and identifiable data, Facebook state in another section on their website that whatever information and data a user shares with the group “Everyone” is also made available to API’s (Facebook Privacy Update to Settings, 2010). Facebook also mention that by allowing an API to run on your account, it can access information related to you and your “Friends” contacts and publish stories regarding the actions taken on that application or website without the users consent (Facebook Platform, 2009).
To understand the potential for massive intrusion into user’s privacy, consider that there are more than a million developers writing API’s with 500,00 active applications running on user’s accounts. A staggering 280 million Facebook users actively use API’s every month exposing their personally private and identifiable data to potential data theft, defamation and other serious consequences (Facebook Statistics, 2010).
Facebook Connect is another service that has the potential to be an even greater threat to user’s privacy than API’s because it runs on participating external websites allowing Facebook users to login and share information about their activities and their identity with other Facebook users and the web (Facebook Connect, 2010).
Lowensohn (2010) explains that, “A site owner just needs to sign up for a Facebook apps API key, then link it up with their Webs account. In return, their users get all the usual Facebook Connect perks, like being able to carry over their Facebook profile picture and other information to that Webs site profile. And whatever comments they leave, along with content on those Webs pages, can be shared back through their Facebook news feed.”
Facebook Connect works on the same principle as API’s, meaning, that a participating website has access to users profile data and the potential to steal this information (Grimmelmann, 2009). There are also issues with trust, because third party websites that participate only need to apply for an API key, and there are no rigorous standards enforced as to who can be a developer. 60 million users actively use Facebook Connect on external sites every month putting their private data at risk.
A good example illustrating the privacy issues with Facebook Connect is given by Tynan (2010), retelling the story of his friend Bob who registered on a dating site called OK Cupid. During registration, the site asked if it could fill in the form using Facebook Connect. He agreed. A few days later he checked his Facebook application page and found Ok Cupid listed there and that it was set to publish one line stories of his activities from the Ok Cupid website on his Facebook Wall by default. Had Bob not checked his application settings page, his activities on Ok Cupid would have been broadcast to his “Friends” and others with potentially embarrassing and/or serious consequences.
The reason why API’s and Facebook Connect are able to access user’s profiles and their personally private data is due to the open default privacy settings that govern visibility and access rights on user’s accounts (Onwuasoanya, Skornyakov & Post, 2008).
As mentioned previously, Facebook provides three basic levels of privacy that provide different levels of access to user’s profile data; “Friends”, “Friends of Friends” and “Everyone” (Privacy on Facebook, 2010). What users do not realize is that by opening an account with Facebook, by default, their privacy settings are open and set to either “Friends of Friends” or “Everyone” making their profile visible and accessible not only to their “Friends” but also to all Facebook users and the Internet as well as API’s including Connect (Facebook Privacy Policy, 2009).
Facebook even state on their website that the recommended privacy settings should be set to “Everyone” so that user’s profiles and personal information is able to be searched by their friends, by everyone on the Internet as well as search engines and RSS feeds (Facebook Privacy Policy, 2009).
Facebook say regarding contact information, that the privacy setting should be set to “Friends” (Privacy on Facebook, 2010) but as we know this setting is no safeguard nor guarantee of a user’s privacy and personal information as 30% of Facebook users add people whom they have never met as “Friends” (Govani & Pashley, 2005).
What Facebook has done in regards to privacy is in-fact the opposite of privacy. By setting the default privacy settings to open, they have effectively disabled privacy on all Facebook accounts and created breach of privacy on a massive scale making accounts viewable and accessible to all users on Facebook, API’s and everyone on the Internet (Onwuasoanya, Skornyakov & Post, 2008).
Facebook users do have a choice in protecting their privacy and data by changing these open defaults on their Privacy Settings page in their account, but according to Bonneau & Preibush (2009, p18), who make reference to a previous study, “between 80% and 99% of users never changed their privacy settings”. If we apply this statistic to Facebook users even at the lowest level of 80%, that means 320 million users have left their privacy levels unchanged since opening their account leaving their personal data vulnerable and accessible to everyone on the Internet.
The fact that so few users change their default settings is not surprising when according to Bonneau & Preibush (2009, p19), “…found that 24% of Facebook users did not understand the implications of their own privacy settings”. This finding is backed up by another study of college students which found that between 20% and 30% were not able to understand how the Privacy Settings in Facebook functioned, nor how to make changes to it (Grimmelmann, 2009).
This lack of understanding is in part due to the confusing layout and design of the Privacy Settings page. Although Facebook provide user’s with a complex set of technically competent privacy settings (Grimmelmann, 2009), there are at least 27 different privacy categories for which a user needs to select either “Friends”, “Friends of Friends”, “Everyone” or “Customize”. Some settings like photo albums even require the user to change settings for each and every album displayed. Due to the complex layout, it’s no wonder that users do not understand these settings properly and how they apply to their profile’s data accessibility (Facebook Privacy Settings and Fundamentals, 2010). The resulting navigational confusion places people’s privacy and personal data at risk.
A Privacy Policy is required by the Federal Trade Commission for any commercial website that collects personal information (Mitrano, 2006). It is a legally binding contract between the user and the social network provider such as Facebook, providing information about the terms of data collection on users accounts and what is done with this information (Bonneau & Preibush, 2009). Facebook has a privacy policy, but this does not automatically mean that it provides privacy quality regarding data collection. As stated by Mirtano(2006, p2), “…a privacy policy meets the requirement even if it states that the owner of the site will sell the personally identifiable information to the lowest bidder.”
Facebook’s privacy policy states its data collection terms quite clearly in that they want to share user’s information, and that by default, Facebook make available a variety of user profile information to search engines and that profile data like name, profile photos, list of friends etc, are not covered by privacy settings and considered public (Facebook Privacy Policy, 2009). The onus is placed back on the user to select the correct privacy setting specifying what information can be shared and distributed to “Friends”, Friends of Friends”, “Everyone” and third parties like API’s and Connect (Facebook Privacy Policy, 2009).
Considering how important is for users to know this information, it is staggering to find that 80% of the users surveyed for the study had not read the Facebook Privacy Policy according to Govani & Pashley (2005). As the majority of users do not read the Privacy Policy, they will be ill equipped and uninformed about what happens to their profile data and how it is shared to other parties.
The last major privacy concern is Facebook’s Terms of Service that applies to the intellectual property rights to the user’s profile content. The Statement of Rights and Responsibilities (2009) states that, “For content that is covered by intellectual property rights, like photos and videos (“IP content”), you specifically give us the following permission, subject to your privacy and application settings: you grant us a non-exclusive, transferable, sub-licensable, royalty-free,worldwide license to use any IP content that you post on or in connection with Facebook (“IP License”). In short, Facebook own their user’s data and the intellectual rights to use the content for any purpose they chose without user consent (Mitrano 2006, p6). Even if users delete their profile and content, if profile data has been shared through applications (API’s), Connect or with other users like photos, they still own the content.
Users presume that the activities they perform on Facebook take place in a private space where their identity and privacy is protected. The fact is however that Facebook is hosted in a public space where user’s profiles and content are by default accessible to everyone (Barnes, 2006). Facebook is now a commercial business whose ad-based revenue model sustains the Facebook network (Facebook Advertising, 2010). In order for Facebook to be successful in its business model it needs to sell ads that are targeted to specific Facebook users (Johnson, 2009). Users have presumed incorrectly that Facebook protects people’s privacy simply because they themselves have not read the Privacy Policy or Terms of Service and informed themselves.
In this paper it has been shown that Facebook in its design of the “Friending” mechanism, API’s and Connect, permit developers and external websites to access users profiles and data due to the open default settings on users accounts. Each and every major design flaw discussed specifically permits sharing of user’s personally identifiable profile data with other Facebook users, developers and external websites. Although Facebook mentions that the privacy of users data is important in their documentation, the onus is placed back on the user themselves to ensure that settings are correctly configured. Why is this? The conclusion I have reached is that Facebook’s main purpose is to share information and connect people but also to make money through an ad based revenue model. To achieve both of these goals their system by design cannot be closed with strong privacy settings for its users. This is the apparent paradox because people want privacy but they also want to share information freely. Facebook has made the deliberate choice in its design to allow information to be shared easily with everyone at the expense of users privacy with unfortunate consequences. As Barnes (2006) said, social network sites create an illusion of privacy, and Facebook does exactly this.

Exposing the inherent privacy design flaws in Facebook by Jason Knight is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 2.5 Australia License.
References:
Barnes, S. (2006). A privacy paradox: Social networking in the United States. First Monday, 11 (9). Retrieved from http://firstmonday.org/htbin/cgiwrap/bin/ojs/index.php/fm/article/viewArticle/1394/1312#note4
Boyd, d., & Ellison, N. (2007). Social Network Sites: Definition, History, and Scholarship. Journal of Computer-Mediated Communication, 13(1).
Retrieved from http://jcmc.indiana.edu/vol13/issue1/boyd.ellison.html
Bonneau, J. & Preibusch, S. (2009). The privacy jungle: on the market for data protection in social networks. WEIS 2009: The eighth workshop on the economics of information security. Retrieved from http://preibusch.de/publications/Bonneau_Preibusch__Privacy_Jungle__2009-05-26.pdf
Dwyer, C., Hiltz, S. R., & Passerini, K.(2007). Trust and privacy concern within social networking sites: A comparison of Facebook and MySpace. Proc. AMCIS 2007. Retrieved from http://csis.pace.edu/~dwyer/research/DwyerAMCIS2007.pdf
Facebook Advertising. (2010). Retrieved from http://www.facebook.com/advertising/
Facebook Connect. (2010). Retrieved from http://www.facebook.com/help/?faq=13458
Facebook Platform. (2009). Retrieved from http://developers.facebook.com/about_platform.php
Facebook Privacy Policy. (2009). Retried from http://www.facebook.com/policy.php
Facebook Privacy Settings and Fundementals. (2010). Retreived from http://www.facebook.com/help/?page=839
Facebook Privacy Update to Settings. (2010). Retrieved from http://www.facebook.com/help/?faq=16377
Facebook Statistics. (2010). Retrieved from http://www.facebook.com/press/info.php?statistics
Govani, T. & Pashley, H. (2005). Student Awareness of the Privacy Implications When Using Facebook. Retrieved from http://lorrie.cranor.org/courses/fa05/tubzhlp.pdf.
Grimmelmann, J. (2009). Saving Facebook. Iowa Law Review, 94, 1137-1160, Retrieved from http://www.law.uiowa.edu/journals/ilr/Issue%20PDFs/ILR_94-4_Grimmelmann.pdf
Gross, R., & Acquisti, A. (2005). Information revelation and privacy in online social networks. Retrieved from http://www.heinz.cmu.edu/~acquisti/papers/privacy-facebook-gross-acquisti.pdf
Johnson, B (2009, September 16). How exactly is Facebook making money?Guardian. Retrieved from http://www.guardian.co.uk/technology/blog/2009/sep/16/facebook-money
Lowensohn, J. (2010, March 30). Webs taps Facebook Connect for log-ins, sharing. CNET News. Retrieved from http://news.cnet.com/8301-27076_3-20001384-248.html
Mitrano, T (2006). A wider world: Youth, privacy, and social networking technologies. EDUCAUSE Review, 41(6), 16-29, Retrieved from http://www.educause.edu/EDUCAUSE+Review/EDUCAUSEReviewMagazineVolume41/AWiderWorldYouthPrivacyandSoci/158095
Onwuasoanya, A, Skornyakov, M & Post, J. (2008). Enhancing Privacy on Social Networks by Segregating Different Social Spheres. Rutgers Governor’s School of Engineering and TechnologyResearch journal, Retrieved from http://www.osd.rutgers.edu/gs/08papers/Paper08-Facebook.pdf
Privacy on Facebook. (2010). Retrieved from http://www.facebook.com/privacy/explanation.php
Statement of Rights and Responsibilities. (2009). Retrieved from http://www.facebook.com/terms.php?ref=pf
Stripling, J. (2010, March 2). Faculty on Facebook: Privacy concerns raised by suspension. USA Today. Retrieved from http://www.usatoday.com/news/education/2010-03-02-facebook-professors_N.htm
Tynan, D. (2010). Facebook’s Sneaky Apps and Privacy Issues. Retrieved from http://www.pcworld.com/article/193423/facebooks_sneaky_apps_and_privacy_issues.html

Comments
Leave a comment