Facebook and the cookie jar
It’s been a tough week for senior management at Facebook, as the world’s press has finally caught onto what many of us in the tech industry have warned our friends about for years: all those little quizzes you’re doing aren’t just harmless fun.
It’s like the world has finally noticed the hand in the cookie jar.
There are two separate issues at play here.
1) The seemingly innocuous little quizzes and the like which – individually – aren’t all that significant, but when taken as a group and analysed in a ‘big data’ fashion, become quite useful at identifying different traits in groups of people – information which is undoubtedly useful to the advertisers who are ultimately Facebook’s customer (remember, you, dear reader, are the product, *not* the customer).
2) The second, and in my view, far more insidious, issue at play here is the less than transparent data collection and analysis Facebook has been shown to carry out. This is especially apparent through its apps – just try installing the Facebook or Messenger apps on an Android device and have a look through the list of permsisions it demands. It goes well beyond capturing your interaction with Facebook, and appears to collect all sorts of other data from other – unrelated – services and apps on your devices.
“Somehow it has my enire call history with my partner’s mum…”
If you want to have a look for yourself, you’ll need to go to www.facebook.com/settings and download your data as a zip file. Extract it and you’ll be able to analyse the data Facebook has stored about you.
“a historical record of every single contact on my phone, including ones I no longer have…”
Dylan McKay recently posted a lengthy Twitter thread with his analysis of his data dump from Facebook – it’s an eye-opening read: https://twitter.com/dylanmckaynz/status/976368845635035138
Spoiler: Dylan found all sorts of *non-FB* data being collected by the various FB apps on his devices, including text messages (not sent via FB), phone calls (not made via FB).
“metadata about every cellular call I’ve ever made, including time and duration…”
If that’s not enough to concern you, let’s then consider what Facebook does with all that data. Cast your mind back to 2014, and you might recall the brief flurry of news when it was discovered Facebook had – under the guise of a psychological experiment – manipulated the order in which items appeared in users’ feeds to see if it had an impact on their mood. Put simply, they tried up-ranking positive posts for some users and up-ranking negative posts for others, then analysed those users’ posts to see if they could identify a mood change.
Read about it in more detail here: Facebook sorry – almost – for secret psychological experiment on users
Fast forward a couple of years to May 2017, and that ‘pyschological research’ appears to have well and truly left the academic sphere: Facebook told advertisers it can identify teens feeling ‘insecure’ and ‘worthless’
It’s easy to see that we’re now way beyond simply providing demographic data to help marketers target their adverts more effectively – something I suspect most of us would consider ‘fair game’.
A line has well and truly been crossed. It’s no longer just about targeted advertising, but a deliberate attempt to mislead users about what data is captured and what is done with it.
See Part II of this article for a few suggestions for how we might (from a tech angle) combat it.
