Header Image for Over-Social Media Post

As the news has come to light over the past week regarding Facebook and Cambridge Analytica a campaign to delete Facebook has arisen that has gained some media attention; although it’s not clear yet if this has actually translated into people actually deleting Facebook. From the initial revelations it’s not clear what this can actually be called. Some people have referred to it as a data breach; Facebook calls it a ‘breach of trust’ and some people have simply called it Facebook’s normal mode of operation.

I don’t think the relationship between Facebook and Cambridge Analytica will become clear for sometime yet and so won’t go into much depth regarding what happened; what it has shown some people and reminded others is that you should thoroughly think through what you share online. There has been an adage passed around for sometime that “if you’re not paying for the service, then you are the product” and this story has only highlighted that fact.

A personal example of how pervasive the tracking is within social media is whenever I browse for something on Amazon, for a number of days after on Instagram I’ll have Amazon ads based around the product I searched for. Across my ads on Facebook the theme of the ads I’m shown will change depending on the links I’ve clicked and I’m sure I’m not alone in finding this.

The Cambridge Analytica story has another element however, one method they used to build a profile of the people they wanted to target was to create innocent looking surveys and then link this with user profiles so they could more effectively target users based on who their clients were. The mundane surveys people fill in without a thought as to the aim behind them, may actually be divulging huge amounts of information about that person.

People updating their status on sites are also allowing organisations to gain an incredibly detailed picture of their behaviours, locations and what may influence them. For example in a previous blog post I used a Twitter analysis tool which outputted information that included the most common times a user sent a tweet, the most common geotags, as well as who they tweeted most. This data alone is enough to glean a huge amount of information and the conclusions drawn will only become better as data science advances further, and this is information found without looking at the contents of the actual tweet. Similar tools are widely available online, Tinfoleak being just another I have also come across.

Not only can the data behind tweets and posts be revealing, but so information we actively post. Many people have come to regret what they have included within their posts with employers, press, and public finding the content of an individuals social media account and severely criticising a person based on those tweets.

When the story of a persons historical tweets just adds credence to the claim that social media appeals to some of the basest personality traits of people. In the cases I have linked to; discounting the possibility they believe what they tweeted, their tweets seem to be an effort to stand out from the crowd and gain a large following by being ‘edgy’ in their humour. There remains only one solution to avoid this scenario; think hard about what you’ve written before you hit the post button if you don’t actually believe what you’ve typed. I try and follow the mantra “Would I say what I’ve typed either directly to that person or through a megaphone in the middle of a city”. If it’s a no, in the bin that tweet goes.

In my opinion the tracking element of websites is more passive in nature. While a user may sign-up to a social media website and expect themselves to be tracked throughout that site, should they also be expecting to give that site permission to use their browsing data from that site’s advertising ‘partners’. Going back to my Instagram/Amazon example, I don’t ever recall being informed that the two companies have their properties linked in anyway; should there be a flag on their respective sites indicating this? The one foolproof way to at least reduce the amount of tracking you are subject to may be the way most people aren’t willing to do; don’t sign up.