When I met my boyfriend I sometimes felt a tinge of regret that we never got to make an adorable meet-cute story, despite living in the same city for five years and attending the same university. We met online at the height of the COVID pandemic, and our first interactions were all digital.
Thanks to Google’s location history quietly ticking away collecting GPS data in the background I thought that I could potentially find an answer to how fatefully our paths criss-crossed before we met. Just how closely did we come to potentially having an eyes-meet-across-the-room moment IRL?
The project received much more attention than I ever expected. First it made it to the [top of Hacker News](https://news.ycombinator.com/item?id=33077473), then it was picked up by influential [Software Developer on Twiter Dare Obasanjo](https://twitter.com/carnage4life/status/1577590247143378944?s=46&t=TUAWPBQ-rQtU5cBTiAVA0Q). More recently it was noticed by [Genís Roca, Barcelonian tech writer](https://twitter.com/genisroca/status/1607345162589552641), which led to a feature on [Catalonian TV news station TV3](https://www.ccma.cat/tv3/alacarta/telenoticies/una-parella-descobreix-que-van-coincidir-33-cops-abans-de-comencar-la-relacio-a-traves-duna-app/video/6196113/).
Aside from journalists, this project has received nearly 7,500 likes and upvotes in the places I know that it’s been posted. More impactfully, however, are the at least 600 people around the world who have weighed in commenting with their feelings about data privacy. Many commenters said that they checked their privacy settings as a result, or wanted to do the analysis for themselves. Whether people were capivated or creeped out by the use of personal data collected by big coorporations like Google, the project, I hope, has given consumers a sense of agency over their own data and data collection that they are subjected to. While the data protection laws and regulations are just being developed, public awarness and discussion on this issue is critical.
I used the R programming language to extract and transform data that is downloadable on request from Google Map’s Timeline. This data is recorded with a user’s permission from sources like background GPS tracking. Using data files downloaded by both me and my partner, I used R to extract all the location records into a database of records over the span of more than 5 years. My boyfriend requested, downloaded, and shared with me all his data himself.
I cleaned the data next so that records with low accuracy confidence were removed. I also condensed the volume of data by removing unnecessary detail from the dataset by grouping together records into 10 minute intervals. Similairly, GPS coordinates were also grouped together into a grid of coordinates 111m apart from each other.
Once the two datasets were compiled and cleaned up, I did the analysis. I wanted to know how close together we got at any time so I matched up all of the timestamps to make pairs of records. Then I calculated the distance between the pairs of coordinates. This yielded the result that on 33 recorded occasions we had been on the same grid-point before we met within the same 10-minute interval.
Context about the project:
* In the last ten years the number of countries that have data privacy laws has doubled. Now 71% of the countries in the world have adopted data privacy regulations to protect consumers, and another 7% have draft regulations waiting to come into effect. This speaks to the fact that in an international context countries and their governments are talking about who has what rights to personal data. [(Source: UN Conference on Trade and Development)](https://unctad.org/page/data-protection-and-privacy-legislation-worldwide)
* Due to the budding data protection regulatory scene, it means it’s never been more important for consumers themeselves to have a say in the development and enforcement of these new regulations. My project brings data privacy to the front of public consiousness, but I hope it also shows the benefits of personal data collection with informed consent, and how it can be used to come to deeper insights that would be otherwise unanswerable.
* I completed the project as a portfolio project while job searching, so my piece is written as a code-focused demonstration of my data visualisation skills. I didn’t ever expect it to grab the imaginations of so many people- technical and non-technical!
* Without any platform or promotion the research has momentum. Aside from me doing the project and posting it for little or no followers on [LinkedIn](https://www.linkedin.com/feed/update/urn:li:activity:6970680653004615681/) and [Medium](https://medium.com/@channonperry/i-used-my-google-maps-location-history-to-see-if-my-boyfriend-and-i-could-have-met-before-we-did-c6842c4a8b64), data experts and regular romantics alike have really been drawn to talking about and sharing my story.
* This is a small personal project as opposed to a funded journalism excercise carried out by a team. Aside from the limitations that that presents I want to highlight it because even an honourable mention [(like the one I got from The Pudding)](https://pudding.cool/process/pudding-cup-2022/) would have a huge impact on a young and inexperienced data story teller like myself.
What can other journalists learn from this project?
I think that my project caught public imagination not because of my small story, but because it tapped into the curiosity that readers have about their own lives. Applying technology to the big philosophical nature of the research question, “Could my romantic partner and I have met sooner”, is solidifying the otherwise nebulous unknowns of the past into hard data. This has proven to be polarising, which also prompts discourse between people with different views on areas like data privacy that are still being developed, especially in terms of laws and regulations.