In November 2014, we held our second hackathon with our YRS participants. This event was much anticipated as it gave us an opportunity to return all of the data we had been collecting over a six month period via the MobileMiner app. The day began with Giles doing a brief presentation, outlining what we had collected and how our young coders could best work with all the information that they were given.
While there were some initial struggles in trying to figure out what to do with all of the data, our researcher participants persevered and came up with some brilliant prototypes for apps and ways to visualize our data. What we quickly realized is that there are just so many things one can do with a lot data, all you need is a good imagination!
In just a few hours, here are some of the hacks that were undertaken by our research participants:
- Cell Tower Time: Featured an app that tried to display different colours on a map to show what cell towers where used at certain times. This decision was made instead of using GPS, because our researchers collectively decided that this form of location tracking would have been too invasive for our participants. As a result, the MobileMiner app converted cell tower data into very approximate location data using the gazetteer of cell tower locations provided by opencellid.org. Had this hack been completed, it would have highlighted all of those Cell Tower IDs that had been gathered during the six months of data gathering, highlighting how easy it is for us to be tracked with the most basic kinds of technology.
- Data Mining: This hack created applications using the SQLlite databases that were created in the MobileMiner application. This collaborative effort tried to produce something that might allow users to gain insight into what the ‘average’ leaks when the engage on social media platforms. In other words, thinking about ways of quantifying the notifications that occur when a user uses an app such as Facebook or Twitter.
- Social Media and the Days of the Week Correlations: This hack tried to develop a graph generating tool to demonstrate when and for how long, the Twitter, Facebook and Facebook Messenger app were used by our research participants. More specifically, it aimed to know what days of the week these apps were being used, alongside the number of times they were accessed and at what time of the day they were most frequently used. He then cross referenced this with the number of notifications that the apps sent back to their home servers. By so doing, he asked three questions of the data: 1) Does using an application result in an increased number notifications? 2) Is there a day of the week that appears to make the user more vulnerable to the app leaking? 3) Is there a time a day that makes the user more vulnerable to the app leaking? In addition, the hack tried to visualize social media use and seeing how this changes over time. From this perspective, the hack tried to extend the quantified self app to encompass social media usage.
- Sonification of Processes and Notifications (Sonosocialdata-ism): This hack tried to process sketch that takes the data from the mobile miner app (as a CSV file) and produces a tone at a higher frequency with processes (popularity of usage) and increases volume with notification count. The participant thus turned data into sound by developing a number of tones and colours to correlate with the frequency in which the apps we had been monitoring were calling home. Had this project been completed, then the app would have generated brighter and more grating sounds to coincide with the leakier and more invasive apps found on any user’s device. This unique approach to visualization came from their belief that sound and colour signify more clearly to the average user. In their own words: ‘Most people would prefer this to numbers on a screen or paper, as its a lot more jarring for the non-savvy, as some could say that higher, louder tones are uncomfortable, whereas seeing numbers is relatively meaningless unless you know the context…This could break down the complicatedness for the end user…It would be great if you could listen to a list of apps, to find the tones, to find the one that are potential problems.’
We are very excited that some of our initial results have already been published! In October 2014, Dr. Tobias Blanke travelled all the way to Washington to attend the IEEE conference, where he presented our article: “Mining Mobile Youth Cultures”. This short paper discusses the co-research we have been involved with over the past year with our YRS participants. Here you can read about the MobileMiner app that we designed with the help of the 20 young people we were working with. The app’s primary function has been to capture the data that is regularly leaked from mobile applications. Specifically, it looked at the number of times in which data was exchanged from the mobile and the various apps that our research participants happened to be using. The MobileMiner then captured the data that gets regularly harvested by third parties to investigate how digital culture is tracked and captured in mobile environments.
Briefly, the MobileMiner uses the Android API, which provides functions that return the total number of bytes transmitted and received by a given app. It does so by polling the APIs every half second and then logs the moments when there is a notable increase. We could not develop the MobileMiner as an iOS app because of the highly proprietary and restrictive environments of the Apple ecosystem. The very fact that we were not able to access even limited data off this highly popular operating system is perhaps a valuable finding, albeit not entirely surprising. However, what were able to create for the Android system proved to be successful, in part due to the enthusiastic collaboration with our YRS participants who continuously strived throughout the project to improve upon MobileMiner’s open source code.
Some of the most notable highlights we uncovered from the MobileMiner are that when it comes to the apps that we download, there are some significant differences in the amount of data that gets captured about users. For example: Three of our young coders played the game ‘Don’t Tap the White Tile’, a popular app with over 130 million downloads. From our records it accessed its server 46, 53 and 42 times over a period of 21, 2 and 3 days for each of those users. According to some marketing blogs, our findings are what could be considered ‘normal’. However another game called: ‘The Line-Keep In’, which only boasts 12000 downloads and has no privacy agreement for the user to accept, accessed its server 1760 times over a period of 27 days, with significant activity found even when the user was asleep!
The paper also highlights some of the finding from our focus groups, which targeted the complex relationship young people have in relation to their online privacy. Here we tried to raise a series of questions around how much control users feel that they actually have and examined if privacy is something they feel they experience individually or collectively. The questions also sought to understand why young people are willing to share so much about themselves online by seeking their attitudes towards information sharing. In our future research, we will work with the YRS communities to co-develop a set of analyses that will utilize the big social data for humanities research in the digital culture.
I was fortunate enough to be invited to Transmediale 2014, the Berlin-based festival of art and digital culture to speak. On January 31st I spoke on the panel ‘The Uses and Abuses of Big Data’ along with Jussi Parikka (http://jussiparikka.net/), Kate Crawford (http://www.katecrawford.net/), and Paolo Cirio (http://www.paolocirio.net/). My paper was entitled ‘Big Social Data: Potential Invented By New Relations?’ and it critically examines the social data we generate with our mobile devices. I ask whether it can function as anything other than surveillance fodder for the NSA/GCHQ or potential surplus value for the information economy?
Here is a link to our panel. My co-presenters Jussi, Kate and Paolo are all very dynamic thinkers and provide wide-ranging insights into big data. Plus the video is in HD!
In my talk I propose the ‘digital human’ as a way of theorising our data-rich contemporary condition and better understanding our constitutive relationship with technology. I also draw on concepts from the French philosopher Gilles Simondon as a means for thinking through our contemporary state of being wherein we are in a collective state of data excess. The challenge is in accessing our own ‘big social data’ outside the proprietary platforms in which they are generated. I conclude by suggesting that we face a fundamental cultural, political and economic challenge in moving beyond our current state of generating unprecedented amounts of personal data over which we have precious little control.
In short, just like a snow-laden slope pre-avalanche, we all generate social data that collectively situates us in what Simondon calls “this initial supersaturation of being.” Right now this state of profound potential is primarily being actualised by capital and the security state. Collectively, however, this potential could be actualised in very different ways through new relations. One of the reasons we are undertaking our research is through a shared belief that big social data can be more than a source of profit for Facebook or digital surveillance. We hope that our research is making a modest contribution to getting there by critically examine and working through our own social data and the ecosystem through which it circulates and is processed.
Today marks my overdue introduction to this blog. My name is Mark Coté and I an the Co-Investigator of our AHRC-funded ‘Our Data Ourselves’ research project. I am also Programme Director of the MA in Digital Culture and Society at King’s College London, where we are also undertaking our research. Along with our Principal Investigator Tobias Blanke, and Research Associates Jennifer Pybus and Giles Greenway, we will be posting updates on our research as well as related reflections. We are always interested in the ideas and research of others and hope this can also function dialogically. So please if you are interested or provoked by what you read, please do let us know. Thanks.