Tag Archives: geolocation
Quantified Self Labs is dedicated to the idea that data access matters. Moving forward, we’re going to be exploring different aspects of how data access affects our personal and public lives. Stay tuned to our QS Access channel for more news, thoughts, and insights.
On January 13th Uber, a wildly popular and often scrutinized ride share company, announced they have entered into an agreement with the City of Boston to share anonymized data generated by users of the service. This is the first partnership between Uber and a local government body, but points to the ability to potentially partner with cities that want to take a peak at the vast amount of data about when and where people are traveling within their municipality. Our first reaction to this was to explore if Uber has provided any method for it’s own users to access and export their trip data. Surely if they can able to export and pass along data to a third party, they can pass that data to their own users?
In our exploration of the mobile and web user platforms we found that Uber currently does not offer users with an easy way to access their data. As an Uber customer, you are provided with email receipts of your trips that include travel information, a route of the ride, and cost. This information is also available through their online user account page. However, it is not exportable and accessible in a method that allows individuals to store information in a consistent and machine readable format (such as a csv file). In our search for methods to assist in exporting Uber ride data, I stumbled upon this data scraper on Github developed by Josh Hunt. It’s useful to know that Uber has a standard no scraping clause in in it’s Terms of Service, but individual users accessing their own data for their own reasons is probably not what these clauses are meant to protect.
Aside from data access issues there is of course open questions about how Uber will implement privacy protections governing sensitive user data. Of course, Uber is not without fault in this space. The now infamous blog post pointing to their ability to track one-night stands (archived here) was enough for some users to question ethical standards within Uber. In their announcement, Uber touched on this issue by stating that they will provide some privacy protections by only offering anonymized aggregated data to third party partners. Protecting user privacy through data aggregation and anonymization is a step in the right direction, but there remain these open issues around data access for users. Uber and the cities they partner with will learn a lot about how we travel, but the partnership between Uber and their users could be improved by helping users (myself included) understand their own data and behavior by allowing easier access to the data we contribute when we use the service.
We’re interested to hear from our readers about their experiences using the above mentioned tool, or similar tools to access and export their Uber trip data. Please let us know. We’ve also reached out to Uber for comment.
I reached out to Uber Support over Twitter and received the following response:
“Unfortunately this is not currently a feature, however we’re always looking to improve and I’ll pass your suggestion along! *NM” (link)
We hope you enjoy this weeks list. Feel free to submit articles, show&tell self-tracking stories, and QS data visualizations. Just email me!
Why can’t you track periods in Apple’s Health app? by Nat Buckley. With the recent re-release of Apple’s HealthKit enabled self-tracking and personal data system it no wonder that people are taking a long hard look at what data is being excluded. With the popularity of menstruation tracking apps (this app has nearly 30,000 ratings) it’s surprising this was overlooked. This excellent post is a must read on the topic.
Now That Cars Have Black Boxes, Am I Being Tracked? by Popular Science Editors. Questions and concerns about surveillance are becoming more commonplace. As someone who is looking to purchase a car in the next year or so I was happy to see this post come across my stream.
The Quantified Self community, lifelogging and the making of “smart” publics by Aristea Fotopoulou. I love it when people take a thoughtful look at the Quantified Self community and write about their experiences:
For me, the potential of QS for public participation lies in the show and tell meet-ups that constitute a central feature of this community. Meet-ups enable the exchange of stories about the success or failure of lifelogging practices; they allow people to connect and form synergies around common interests, and to explore wider questions such as personal data management and ownership. [...] members touch upon key political issues and create temporary spaces of dialogue: what happens to personal data, who has access to these data (is it private individuals, governments or corporations)? For what purposes (medical research)? And how can these data be interpreted (by algorithms, visualisations) and used to tell stories about people?
Stepping Down: Rethinking the Fitness Tracker by Sara M. Watson. Sara uses her personal journey of recovery from hip surgery to frame an interesting question: Should we trust our fitness trackers to prescribe movement goals?
Practical Statistical Modeling: The Dreaded After-School Carpool Pickup by Jamie Todd Rubin. Jamie wanted to understand if there was a way he could reduce how much time he spent waiting in line to pick up his son from school. Why not track it and model it!
Bulletproof Diet and Intermittent Fasting: 1.5 Year Results by Bob Troia. Bob takes a deep dive into his data to see if this particular diet is having beneficial health effects. Click for the great data, stay for the wonderful discussion and very, very thorough write-up.
Quotidian Record by Brian House. I’ve been a fan of Brian House since his early days visualizing Fitbit data. I was reminded of this work during a conversation about geolocation data and thought it would be a nice addition to our visualization list.
Visualizing My Daily Self-Management by Katie McCurdy.
What does my daily medication and self-management look like? How could I visualize this regimen? How can I communicate the ‘burden’ and work of caring for myself?
I decided to draw pictures of the things that I need to do on a daily basis; that way I could show the workshop attendees what my day was like instead of just telling them.
It’s Time to Eat by Karl Krehbiel. Karl, a data science intern at Jawbone used the data from their global community of users the determine the likelihood of food and drink consumption during the day. Really fun and interesting visualizations here.
Jamie Aspinall was interested in what his location history could tell him. As a Google Location user, his smartphone is constantly pinging his GPS and sending that data back to his Google profile. Using Google Takeout Jamie was able to download the last four years of his location history, which represented about 600,000 data points. In this talk, presented at the London QS meetup group, Jamie describes his process of using a variety of visualizations and analysis techniques to learn about where he goes, what causes differences in his commute times, and other interesting patterns hidden in location data.
You can also view his presentation here.
This guest post comes to us from Mark Moschel and Eugene Granovsky, the co-organizers of the Chicago Quantified Self meetup group. At their recent meetup on March 26, 2014 they had three great talks from community members. If you live in the Chicago area why not join the group!
Dan Abreu on GeoTracking
Dan travels a lot. I mean… a LOT! He stepped through an airport well over 300 times in 2012. He started documenting his travel a few years back and has used a variety of tools since: TripIt, Track My Life (discontinued), Google Latitude (discontinued), QStartz, and myTracks. During that time, his technique for tracking evolved and gained complexity. He’s now able to develop very detailed maps of his trips (see below). What has he learned from all this? “Not much” he said. However, he enjoys the practice and consistency of it and is excited to continue finding more uses for this data in the future.
Zak Boswell on Sleep
Like many of us, Zak was on a very inconsistent sleep schedule for most of his life and would often stay up too late. However, unlike many of us, Zak was experiencing severe fatigue during the day. In the span of just a couple years, he had 4 car accidents from falling asleep at the wheel (in two, his car was totaled). Realizing this was a problem, he started exploring traditional solutions. He saw a handful of doctors and participated in a very expensive (and ineffective) sleep study. During this time, he also started tracking his sleep and decided to go to bed at a consistent time each day (around midnight). In the data, he saw his sleep quality beginning to improve. He also stopped falling asleep during the day. At first, he struggled with the change, but he’s since changed his whole philosophy and loves it. You can view Zak’s presentation here [PDF].
Ovetta Sampson on how tracking helped her become an Ironman (or “The science of Faith”)
Let’s start with the end on this one. Here’s what Ovetta accomplished: 2.4 mile swim, 112-mile bike ride, 26.2 mile run – all in under 17 hours. Wow! Even more impressive is that she was never an athlete growing up and weighed 270 lbs in 2012. In just a year, she turned a seemingly impossible goal into a real accomplishment. How? She found faith in her data. “Tracking data helped me change my behavior” she said. By tracking her times, weight, speed, and distance, a few things happened: 1) she quickly saw progress and was motivated to keep going, 2) she became competitive with herself, always trying to beat her last score, and 3) she could ignore the thoughts in her mind. As she said, “you have to trust something and the mind is not to be trusted. Trust the data.” Her thoughts kept telling her to quit, but the data proved she was doing well. She didn’t quit and now she’s an Ironman. You can view Ovetta’s presentation here [PowerPoint].
For those of you in the Chicago area Elmhurst Art Museum is hosting a new exhibit called “Lifeloggers: Chronicling the Everyday.” Check it out here.
In the Quantified Self community we focus on projects and ideas that help people access and get meaning out their personal data, including the information you can collect with your smartphone. If you have an iPhone, Android, or Windows phone you’re already have carrying of the world’s most sophisticated self-tracking tools. The GPS, accelerometer, the microphone, all of these tiny sensors make up a great set of tools you can use to understand how you move around the world.
I’m going to focus this short “how to” on geolocation data and mapping your movement, specifically using data gathered by the Moves application. Moves is a passive activity and location tracking tool available for the iPhone and Android. We’ve written a bit about it in the past and had a chance to interview their CEO, Sampo Karjalainen. I’ve been using it since May, 2013 and I wanted to share some neat tools and methods for getting a bit more out of the data Moves collects.
I find that visualizing my data on a map to be incredibly powerful. It might by my inner cartographer, but seeing my patterns of movement (or lack there of) in reference to known places and landmarks is a great mechanism for inducing recall and reflection on where I’ve been and what I’ve done. Hopefully you’ll use one of the tools or methods below to map you data and learn something new!
Moves Connected Apps
Like many self-tracking applications and devices, Moves has a API that many different developers have built services on top of. Here are just a few of the services that allow you to see your data on a map. Be advised that each of these services has access to your data. Make sure to read their Terms of Service before agreeing to the data transfer.
WebTrack. This is by far the most utilitarian data mapping tool. However, you shouldn’t get discouraged by the lack of fancy design because it gives you an very unique data view. When you use Moves on your phone you typically only see the “storyline” and the detected places you’ve spent time at. However, Moves is constantly pinging and recording your location when it detects movement. WebTrack allows you to see all those movement points by hovering over the associated timestamp.
Fluxtream. You might know Fluxtream as Friend of QS and a great open-source data aggregation tool. They’ve set up a “Moves Connector” that allows you to import and visualize your Moves data. Because Fluxtream is set up as an aggregation and visualization tool you can also map other interesting data sets. Want to know where you were tweeting last week? Fluxtream will map it for you. (You can see me tweeting on a CalTrain ride between San Francisco and Palo Alto below.)
Zenobase. Another interesting data aggregation service here. Zenobase treats your Moves data bit differently. Rather than importing all the movement geolocation data it focuses on your place data and visualizes those locations. I like the high-level view it start with, but make sure to keep zooming in to see more specific place data.
Resvan Maps. This mapping application adds a unique twist to the typical mapping visualizations. It will plot your places, paths, and categorize paths depending on the activity (transport, walking, running, and cycling). Additionally, you can create “analysis cirlces” and have the application compute the time you spent in a certain location you bound (it aggregates to hours:minutes per day).
MMapper. This method for mapping your data, developed by Nicholas Felton, is by far the most technical, but it produces some really neat visualizations. You’ll have to download Processing and follow the instructions Nicholas provides on the Github repository page here. The great thing here is that the mapping and data access is all happening locally.
Move-O-Scope. Another great mapping application here from the folks at Halftone.co. They’ve probably completed the most thorough mapping and exploration tools for your Moves data. After linking your Moves account you can explore maps by activity type, day of the week, and custom data ranges. Additionally, they’ve implemented a neat feature for exploring place data. You can see how many times you’ve visited a specific place, where you’ve come from and where you go next, what days you typically visit, and your typical time of day at that place. (See this post for background on why they created this nifty tool.)
Map It Yourself!
If you don’t want to trust your data to a third party, but you still want to explore your movement maps there is really great option for you. Our friend and co-organizer of the QS LA Meetup, Eric Blue, recently published a method for easily exporting your data: the Moves CSV Exporter. You’ll have to login and use the Moves pin system in order to download your data, but Traqs isn’t storing your data, just providing a way for you to access it. The tool allows you to download and explore your activity, summary, tracks and place data. We’ll focus on the place data for creating maps. You can also use your full tracks history for mapping all the geolocation points Moves collects.
Because this data is based on latitude/longitude coordinates there are many different methods available that you can use to map your data. I’m going to focus on two here: Google Fusion Tables and CartoDB (if you know of others share them in the comments or our forum).
Google Fusion Tables
Fusion Tables are a new Google Drive tool that you can use to store, analyze, and visualize many different types of data. Once you download your Moves places.csv file you can upload it to a new Google Fusion Table. Once you upload your data, which takes about 2 minutes, you’ll see a menu bar and three tabs: Rows, Cards, Map of longitude. Just click on the “Map” tab and you’ll see your data already placed on a map. If you want to see a heatmap rather than a point map just navigate to Tools -> Change Map and you’ll see an option for a heatmap on the lefthand side. This is just the tip of iceberg for mapping fusion table data. You can learn more about different mapping methods and tricks here.
CartoDB is a visualization and analysis engine for geospatial data. I’ve been using it to play around with a few of the different geolocation datasets that I have (I actively keep three). Although it is paid service, they do offer a free plan for smaller datasets, which is perfect for your Moves data. Again, you’ll have to upload your places.csv file to a new table once you set up your account. Once the data is uploaded there are quite a few different map visualization wizards you can use to view your data in different ways. Pesonaly I like playing with the “Torque” visualization that gives you a real feeling of space-time to your data.
TileMill is an interactive map design tool from the folks over at Mapbox. If you’re looking to create custom maps with your data that you can format, style, and share then this is a wonderful tool to use. At first glance it’s a little daunting because it looks like a mashup of a CSS editor and map tool. That actually gives it the unique power to drive customization. Don’t be afraid, it’s not too hard to get started with. Mapbox has provided a great “crashcourse” to get you started with importing data, saving it as a new layer on your map, and then manipulating how it looks on your screen. If you want to go just a bit farther you can also add legends and informative popups to describe your data points. Mapbox also offers a free hosting plan if you want to share your interactive maps on a webpage. For example check out my MovesMap here, where I added a quick styling to manipulate the point size in relation to the time spent at a location.
Hopefully you’ve learned something new from this. If you map your Moves data (or any other geolocation data) we want to see it! Leave a link in the comments, post it in the location mapping thread on the QS Forum or get in touch on twitter!