Big data, data mining and algorithms. Whose the villain?

Let’s ask the right questions.

Being in design we are always made to think a lot. About technologies, systems, ethics, people, psychology etc.. We are expected to form opinions, have a “stance” and a critical approach towards everything present in the world and beyond. There is one thing that I have noticed. Whenever there is a choice between technology and users, designers tend to blame the technology and advocate for people, the users. Users seem to win designer’s hearts as they seem quite innocent to us designers. We have this inherent need to protect the users like a mother protects a child. And thats obvious as design is for the people. It’s user centric design. But the question is does that make technology a villain?

Let’s talk about big data. What is big data? It is large data sets that may be analyzed computationally to reveal patterns, trends, and associations, especially relating to human behavior and interactions[1]. So simply put it is huge amounts of data that is stored to be analyzed for patterns and trends of the society. Let’s not forget that this data is not new. Since many years marketing companies have used this kind of “metadata” coined by Jack E. Myers which involves people’s likes, dislikes, preferences, choices opinions etc.. to come up with new trends, policies and products that appeal to their audience. It’s what we all call “know your audience”. Designers use this same approach in user research and user interviews to understand the likes, dislikes, their drives, blocks and to understand user’s psychology behind making decisions. This helps us design better. Then why does big data sound so scary? The reason for this is the word “big”. The scare is about the sheer numbers. Since the birth of the internet people’s attitude and behaviors have changed. They no longer feel insecure or apprehensive to display their information online for the world to see. Just a few years back transferring money through online banking was a frightening process. Now everything related to the economy operates online. So if we understand it correctly, internet has slowly brought about behavioral change. Behavioral change is always frightening. Its new and it takes time to wrap our heads around it while analyzing its implications. Now everyone has an online presence, the personal individual data that was limited to a few is now open source for the world. It does sound scary when put this way but we have to learn to accept this new age where the whole world is online. You are not alone. Everybody and everything has an online digital presence. It’s similar to when your kids can officially drink in public. You can’t stop them as parents and take it away but maybe help them understand the rules and some repercussions. Being aware and vigilant about what information you post is one of the tools to avoid giving out too much information.

Let’s understand how this open source digital data turns into actual currency by referring to the documentary “Generation Like” on Frontline with Douglas Rushkoff. They talk about companies like Audience that help amplify the social network and build big data.[2] Big data is created with a series of “likes” and “follows” thus making it into an interlinked network managed by the companies like theAudience. All this data is adopted by the big players(brands) of the market. Companies like Audience sell the audience, the users to big brands for profit.[2] These big brands utilize this information for their own benefit to gain more loyal customers and make more money.[2] The irony of this video by frontline is towards the end of the video they ask viewers to “follow” Frontline to know more and view more about this and various other topics.

The real question is not about building social network and big data and how its built. The real question is are users aware that they are helping brands create new marketing strategies? Are these strategies used to manipulate the users? The personal data put up by users is by their own will and many times users give consent for private companies to use this data. The problem lies when this data itself is responsible for “data discrimination” as rightly put by Alice E Marwick in “How your data are being deeply mined”. She talks about this personal digital data of who you are, what you do to what you buy is analyzed by private companies and sold to marketers, corporations, governments and criminals.[3] Though I felt this was a little far fetched as criminals may not even need data to be sold to them. Were they to access personal data for criminal activities then that is not a new trend. With or without big data there is no saying that crime can be prevented. Alice asks an important question which is “Can we take full advantage of personal data without playing into larger corporate interests?” The answer is no, there will always be a tradeoff. Every coin has two sides. There is no system in this world that is an ideal system. This is not how the world functions. If we as users want to take full advantage of using social media for personal data and building our own media company then can we really blame big brands wanting to leverage social media to the benefit of their companies and getting more consumers? Alice talks about big data being made of little data that is deeply personal data. But is this data really personal? Is metadata, personal data? David Cole talks about metadata in “We kill people based on Metadata.”[5] He introduces that fact that NSA collects metadata and keeps a watch on foreign nationals apparently to keep the country safe from threats.[5] Being a foreign national I was taken aback. So am I being scrutinized too? This is a good learning for me. Being a designer I should ask some ethical questions like “What if I was in the position of the user?” “Would it have been acceptable then?”. These can help me be a better designer while designing experiences for users based on insights from big data.

Let’s look at a positive example of data mining being used for benefit for humanity, to save lives.

The data being used here is also incredibly personal, but the user is aware of the data being used. The difference is, its being put to right use. Nothing more noble than saving lives is there? Then going back to Alice’s article what is the problem area? The part where Alice talks about “behavioral targeting” leading to “predictive targeting” is the perhaps the right question.[3] Eli Pariser somehow rephrases this perfectly when he talks about “being aware of online filter bubbles”.[4] The problem is not that this data has been mined to be used for new marketing strategies, the problem I feel lies with the data discrimination and “data personalization” rather than “data customization”. Don’t decide for the user but let the user decide for themselves what information they want to see and what they don’t care about. Companies like Google and Facebook are failing to understand that personalizing user’s experience by completely deleting all the things that user does not click is not a safe way to assume that the user does not want to see it. This is also a problem because of it being entirely invisible to the users.[3] Users want control and choice. Taking away control from the users and catering to them what the calculations of the “algorithm” suggest is the worse possible method of enhancing user experience. In fact its the opposite. The single biggest question is where is all this leading us? Is it a future with no privacy? This is where we as the designers will play a huge role to shape the future. We can chime in and help build a better experience than what “algorithms” are building. Or maybe help create better algorithms that not only protect privacy but build an equal trade off between users and their favorite brands.

Sources:

[1] https://www.google.com/search?q=implications+of+genetic+algorithms&oq=implications+of+algorithms&aqs=chrome.1.69i57j0.9278j0j8&sourceid=chrome&ie=UTF-8#q=big+data+meaning

[2] http://www.pbs.org/wgbh/frontline/film/generation-like/

[3] How Your Data Are Being Deeply Mined by Alice E. Marwick

[4] https://www.ted.com/talks/eli_pariser_beware_online_filter_bubbles?language=en

[5] http://www.nybooks.com/daily/2014/05/10/we-kill-people-based-metadata/