What You Really Need To Know About Your FB Footprint, According To This Tech Guy & PI

Facebook, We Hardly Knew Ye (But Hoo Boy, Did You Know Us!)

So, you use Facebook. Well, so do a lot of people. Recently, there has been a lot of talk about Fbook kind of letting you down in terms of selling your data and or giving up your data to Cambridge Analytica.  So, I thought I would explain the whole thing to you.

Let’s talk about Mad Men first. A long time ago, people realized you could sell things to people that they didn’t really need. You could convince them that they smelled bad (they probably did), that they needed these little paper tubes stuffed with leaves that would ultimately kill them, a shiny new car, or even things that were completely useless. This became known as marketing. Marketing is about several key things: 1) Segmenting people into groups who are likely buyers; 2) Persuading people who are NOT in the group to join the group; and 3) Convincing people that other groups are bad/stupid/dangerous, etc. So, for instance, back in the 60’s (a la Mad Men) cigarette companies would make different cigarettes for different groups. Marlborough, was originally marketed to women, but it didn’t catch on, so they changed the marketing and made a “man’s cigarette” with a cowboy. Virginia Slims, marketed to women (famous ad slogan “You’ve Come A Long Way Baby”). All of this is just about segmentation of the market and identifying traits that that segment has, so you can use the right image. Colors, logos, words, pictures, it’s all just marketing.

Spin forward. Today, we have something called big data. Big data is all about volume of numbers, not just what some guy in a hat thinks in 1963, but rather using large volumes of information to attempt to find segments that we didn’t know about and/or information we can use to manipulate suckers, erm, customers, into doing something. How about a fashion example:  For some reason, a Valentino Garavani clutch purse costs 4,095 dollars and a very similar clutch purse at Walmart (horribly named “chicastic”) costs 14.99. They are both bags that you can put things in, they are probably both made in a factory somewhere out of similar materials, but one is a LOT more expensive. Why? Marketing. 

Someone figured out that a clutch purse with an Italian name from a designer (although the “chicastic” had a designer too) was worth a lot more. Marketing.  Big data then is about figuring this out. Why Italian? Why not Albanian? What about all those Yemini purses? Chinese? When you think of these things they influence you and create an image in your mind. Big data can search through large volumes of information about people and find complex correlations which can be used to segment, create segments that didn’t exist before, and merge segments to create larger segments. I assure you, Valentino Garavani has some idea of who will shell out 4,095 dollars for that purse. I also assure you, Walmart has a really, really good idea of who will buy the 15 dollar chicastic (that is really how it is spelt on the site).

So, how does all this relate to Facebook? Well, when you join Facebook you agree to their terms.  They collect information about you. Truthfully, it seems harmless enough. “likes cat videos”, “looked at purses today”, also searched for “farsi translation”. Eclectic? Maybe, but what if I can make this a MASSIVE data collection with everything you do, who you are friends with, who they are friends with, and so on. Suddenly, statistics and the law of large numbers kicks in. This means that I can start to draw finer and finer conclusions due to the large number of observations. Ok, so what? I mean, wouldn’t you rather see ads for purses in the pop up then ads for Bulldozer Supplies or whatever random ad pops up. Targeted marketing works and it’s kind of nice. I get a lot of guitar ads and never get ads for accordions. Cool.

So, how does all this relate to Facebook? (Yes, I said it again.) Well, Facebook has been collecting all sorts of things about you but doesn’t stop there. Suppose you use other apps - Farmville, games, photo management, each and every one of these collects information from you, and based on the agreement, may share this information with other apps. Eeek.

Now, enter the API. An API is a programming library of tools which are shared on a platform. This means that developers can all use the same set of tools to work within a platform and produce content. So, Farmville uses an API for Facebook which allows it to function. Butterfly Hunter uses the same API and tools. This means that all these tools can talk to each other using a common language and they all talk to Facebook.

[Ominous chord, F Minor I think.] Enter Cambridge Analytica [Did you imagine Darth Vader, I did!]. So, CA shows up and uses the API too. They realize that they can use an app to talk to other apps and extract data about people who have clicked on agreements which allow their data to be shared. What’s there? Oh boy, wait till you look. Go to the down arrow, click settings, and at the bottom of the page, click Download my Facebook Data. After a while, you will get a link to your download. Take a look at the index.  Mine shows my address, spouse, relatives, birthday, favorite athlete (?I have a favorite athlete? It was Julia Mancuso.), and about 150MB of information about me and what I like and don’t like. Wow!

Now, enter big data. What if we could take all this massive amount of information and get back to marketing basics. Let an AI churn through all this data and find trends and tendencies. Are you a member of the NRA?  Do you like Julia Mancuso? Like a website about Valentino Garavani?  It has books you liked, movies, TV.  If I could find trends here, I could identify market segments for my candidate for political office. Let’s just suppose that we realize that people who tend to like my candidate also tend to like Valentino Garavani and really hate Walmart. Let’s make a meme. I photoshop a picture of my opponent (or get a real one) of my candidate shopping at Walmart and talk about how she not only doesn’t have a Garavani, but is actively working to ensure that Garavani purses are made illegal! (It doesn’t have to be true, it just has to get you angry.) All of a sudden, people are rioting in the streets to protect their interests and sending death threats to the opponent for her “extremist anti-Garavani views”.  These types of things have been shown to die slow deaths and as such what if it starts to actively sway public opinion?

Add to this, a more complex issue. What if I could find other tendencies that were localized?  I could have memes targeted at segments based on this data and buy very specific ad time. Some people get the Walmart pic, some people get other subtler references, and so on. All of a sudden, I can start to shift the opinion based solely on big data findings that are not even obvious. (Like say, people who like Garavani also really hate hot dogs, so I show lots of pictures of my opponent eating hot dogs). Wow!

All this means that we are going to have to be careful with our data, I am not sure how to stop it or even if we can but it means we have to be ever diligent if we want to avoid being manipulated by big data. Now, we have been being manipulated since the first caveperson convinced their neighbors that there were better rocks over in the swamp but we need to protect ourselves from this kind of manipulation as well. What can you do about Facebook? Well, you can get rid of it but there will just be something else to take its place. Better, you can start controlling what you allow. Choose your permissions for apps, and be careful what you read. Russ and I did a show about settings in Facebook on Secure Digital Life (#57) if you want to see more about how to fix this in your own Facebook. Or you could just go with it and hope it all works out…

For more from Doug White, check out his podcast at SecureDigitalLife.com.

About Doug White

Doug White is the Chair of Cybersecurity and Networking programs at Roger Williams University. He has worked in the technology industry for many years and specializes in networking, disaster, forensics, and security. He has been paid to break into buildings, talk tech people out of their usernames and passwords, steal money, and figure out horrible scenarios like “What if a rabid shark swarm was caught up in a tornado while a core meltdown occurred? Could we still watch Netflix?” Doug has a PhD in Computer Information Systems and Quantitative Analysis from the University of Arkansas, is a Certified Computer Examiner, A Cisco Certified Network Administrator, A Certified Information Systems Security Professional, and a licensed private investigator. 

Photo Credits: Doug's foto by Ashley Farney