Lifelogging Data Aggregators

Today there are devices and apps that can record what you eat, what you weigh, how many steps you take, how well you sleep, how far you walk, how much time you spend on Wikipedia, your heart rate, your LDL cholesterol level, and more. It has been called self-tracking, lifelogging, quantified self, behavior monitoring, and probably other things too.

(Some people think it’s new, but it’s not. It’s just easier and cheaper now, so more people are doing it.)

Unfortunately, self-tracking data often gets stored in different places. For example, I use RunKeeper to track my walking and jogging and the data gets stored on RunKeeper’s servers. I also played with RescueTime, which recorded how much time I spent using various software and visiting various websites. That data got stored on RescueTime’s servers.

If I want to see if there’s any relationship between how far I jog and how much time I spend reading news websites, I have to bring the data from RunKeeper and RescueTime into one place (or app).

Some of the devices and apps let you get your data in common file formats, such as CSV files (spreadsheets). Sometimes you have to pay for the “premium” or “pro” account before you can get the data that way. Also, many of the services let other apps access your data using a web API (Application Programmer’s Interface). (You usually have to give those other apps permission to access your data.)

I looked for apps (i.e. software) to aggregate my self-tracking data. As it happens, there are several options. Some open source ones are:

• ThinkUp – Currently, it only gets your data from the larger social networks (e.g. Facebook, Twitter), but they keep adding more sources. They will offer a hosted version soon. Written in PHP. GPL license.
• The Locker Project – Being developed by a company named Singly, which offers a hosted version called Singly DataFabric, currently aimed at developers, not consumers. Written in JavaScript on node.js. BSD license. Update: Singly was acquired by Appcelerator in Summer 2013.
• Manybots – wrote an aggregator written in Ruby, using the Ruby on Rails framework. MIT license. It seems they changed into a different business at some point, but the original code is still up on GitHub.
• Fluxtream / BodyTrack – A project that came out of Carnegie Mellon and a company named Fluxtream. They seem to be regrouping at the moment, but their code is available on GitHub (here: Fluxtream, BodyTrack). Written in Java.
• Reportr. Apache 2 license. Written in JavaScript.

You could install one of those on your desktop computer, but there’s no easy desktop installer to download and run. They’re really meant to be run on web servers, along with other required software (e.g. database and web server software).

Proprietary quantified self data aggregator services include:

• Tictrac – also makes nice charts and graphs.
• Exist.io
• The Wolfram Language along with the Wolfram Connected Devices Project . As I understand it, this is a project to make it easy to pull data from your tracking devices (or their online data storage) into Mathematica (?) for doing calculations, making graphs, making maps, etc.
• Open.Sen.se
• RunKeeper and HealthGraph – RunKeeper started as an app to track runs, but it has grown into a software-development platform (HealthGraph). The platform can store and compute with data from many apps (once permission has been given by the user).
• Endomondo
• EveryMove – gives you points for doing stuff; you can then get real-world rewards with the points
• A service from Qualcomm Life and WebMD, announced in March 2013. It’s built on top of Qualcomm Life’s 2net Platform and WebMD’s Health Cloud platform.
• Apple’s iPhone Health app (new with iOS 8). iPhone developers can develop apps which talk to the Health app using Apple’s HealthKit API.
• Google Fit
• Microsoft Health
• Sony Lifelog

If I missed one, let me know.

Photo Credit: The photo, titled “Umm…” is by Clearly Ambiguous on Flickr and is licensed under a Creative Commons Attribution 2.0 license.