Tag Archive: Privacy


For the first time, Facebook has revealed details about how it tracks users across the web.

Through interviews with Facebook engineering director Arturo Bejar, Facebook spokesman Andrew Noyes, Facebook corporate spokesman Barry Schnitt and Facebook engineering manager Gregg Stefancik, USA Today‘s Byron Acohido was able to compile the most complete picture to date of how the social network keeps tabs on its 800 million users.

Here is what Acohido learned:

    • Facebook doesn’t track everybody the same way. It uses different methods for members who have signed in and are using their accounts, members who are logged-off and non-members.
    • The first time you arrive at any Facebook.com page, the company inserts cookies in your browser. If you sign up for an account, it inserts two types of cookies. If you don’t set up an account, it only inserts one of the two types.
    • These cookies record every time you visit another website that uses a Facebook Like button or other Facebook plugin — which work together with the cookies to note the time, date and website being visited. Unique characteristics that identify your computer are also recorded.
    • Facebook keeps logs that record your past 90 days of activity. It deletes entries older than 90 days.
    • If you are logged into a Facebook account, your name, email address, friends and all of the other data in your Facebook profile is also recorded.

Data about web searches and browsing habits could be used to figure out political affiliations, religious beliefs, sexual orientations or health issues about consumers. According to USA Today, this type of correlation doesn’t seem to be happening on a wide scale, but the concern of some privacy advocates is that selling data could become a tempting business proposition — both to social networks like Facebook and online advertising players such as Google, Microsoft and Yahoo that similarly employ cookie tracking techniques.

Facebook told USA Today that it uses data collected via cookies to help improve security and its plugins and that it has no plans to change how it uses this data. It has, however, applied for a patent on a technology that includes a method that correlates ads and tracking data.

“We patent lots of things, and future products should not be inferred from our patent application,” Facebook corporate spokesman Barry Schnitt told USA Today.

Regardless of how Facebook is handling the data it collects through cookies, by doing so it has entered a very sticky debate about whether consumers should be able to opt out of being tracked by such methods. Aproposed law that would create this option was introduced in February.

While a recent poll found that about 70% of Facebook users and 52% of Google users were either somewhat or very concerned about their privacy, some argue that online commerce would suffer without online tracking.

Thanks: Mashable

 

Advertisements

image

Internet safety is a topic that we are all commonly reminded of as we move to an increasingly digital age. Now, another reminder of how much we willingly share with those we don’t know has been shown, as reported by the BBC.

Using what is known as a ‘socialbot’, researchers were able to acquire information that a Facebook spokesperson rebuked as being “overstated and unethhical”. A socialbot is a botnet adapted for usage on social networks. The worst part of the socialbot’s power is how affordable it is. Dubious websites offer the bots for sale over the internet for as little as 29USD, or 18GBP.

A socialbot differs from a normal botnet in the sense that it can pass itself off as a normal Facebook user. A regular botnet is a type of virus that can infect a user’s computer, and can make use of this to send out spam or partake in digital attacks against other websites. The socialbot takes control of an existing Facebook account, and is able to perform normal activities, such as posting statuses and sending friend requests.

The research was performed by four members of the University of British Colombia, with 102 socialbots being commanded by one ‘master’. The master sends commands to the other bots, which they then act upon. These commands would likely consist of seeking profiles and adding them. In the space of eight weeks, the bots sent out 8,570 friend requests and had 3,055 acceptances. The research showed a relation in the number of Facebook friends a user had, and the likelihood of the socialbot being accepted as a friend.

Remaining within Facebook’s limitations for sending friend requests, the bots sent only 25 requests per day. Any more and the bots risked triggering the fraud detection and prevention system existing on Facebook. According to Facebook, the research is not reflective of how they prevent socialbots operating, as the accounts operated from ‘trusted’ university IP addresses. An IP address used by a real-life criminal operating socialbots would apparently raise alarm bells within the company.

Many people are now growing more aware of friend requests coming ‘out of the blue’, so to speak, and it reflects how people could be growing more aware of the people seeking to acquire more information, whether you intended to give them the information or not.

Thanks: Neowin