I took around 40GB of photos in New York last month. It’s been a few weeks since I’ve been back and I still haven’t sorted through them yet. That’s because I have something better to play with—GPS data for my entire trip, logged with my old iPhone running Motion X GPS. Everywhere I went, how long I stayed, what routes I took…all of that has been recorded. It’s been fun manipulating the data in different ways such as calculating more efficient routes I could’ve taken*. And of course the killer app is combining this data with photos, tweets, credit card statements to create a more complete and compelling account of my trip. * I was walking in circles a lot and not making optimal use of the subway for the first few days of the trip. Going in, I didn’t plan on doing this nor did I read any blogs on people who have done something similar so the data and my methods aren’t perfect. A standalone GPS unit would’ve been better both in terms of battery life and antenna strength. I had to recharge my old iPhone twice a day and there are gaps and inaccuracies in the GPX files. The good thing is that any inaccuracies have supporting data in the form of geotagged photos and foursquare check-ins. From there, it’s really easy throwing everything up onto Google Maps/Earth, but what would really be neat is an interactive front end with Feltron-eqsue graphics. Make it completely automated and you have yourself a hit.  In a way this is already here in the form of fitness apps with RunKeeper, Nike+ GPS*, Google My Tracks. There’s also Twitter for iPhone (“nearby tweets”) and Flickr. Better yet, take Daytum and add location services and automated tracking. What I am thinking is something that automagically tells a story / replaces a blog post** by  pulling data from a portable device (smartphone really) that logs GPS coordinates, steps taken, and whatever else hardware the is capable of. aggregates data from other services like Flickr and Mint runs statistics*** on 1 & 2 spits out cool interactive graphics from 3. * I’ve been using Nike+ (for runs) and Nike+ Active (daily steps taken) for about a year and still enjoy the stat tracking and graphics. Someone else get this so we can compete. ** I paused for a minute to think if this blog post be automatically generated. How meta. But no, what I am referring to of course are blog posts that recount events e.g. “Today I went to X, saw Y, and bought Z.”  *** I have an Excel spreadsheet of some data and that’s been fun as well even with my basic knowledge of statistics. Food was my biggest expense. Avg for a meal was $12 with StdDev 5. Most interesting regression results: Yelp ratings for the places I ate at (taking into account # of ratings as well as the score) tended to be higher when I had a meal closer to the average price. Again, in a way this is already here in some respects, but not quite the implementation I’m looking for. It can’t be too far away. Ten years ago, most people still used film cameras. Now everyone has a DSLR that can record photos and videos. Three years ago most people had a phone that could record a different sort of data—tweets. Now a lot of people have smart phones. Soon everyone will have one that has enough battery life—all that’s needed really—to automatically track and process the necessary information. At Shake Shack (★★★★) $10 got me a Shack Burger and Strawberry Shake. The wait was 30 minutes and I got my food 12 after minutes later. I took 19 photos in the area.

I took around 40GB of photos in New York last month. It’s been a few weeks since I’ve been back and I still haven’t sorted through them yet.

That’s because I have something better to play with—GPS data for my entire trip, logged with my old iPhone running Motion X GPS. Everywhere I went, how long I stayed, what routes I took…all of that has been recorded.

It’s been fun manipulating the data in different ways such as calculating more efficient routes I could’ve taken*. And of course the killer app is combining this data with photos, tweets, credit card statements to create a more complete and compelling account of my trip.

* I was walking in circles a lot and not making optimal use of the subway for the first few days of the trip.

Going in, I didn’t plan on doing this nor did I read any blogs on people who have done something similar so the data and my methods aren’t perfect. A standalone GPS unit would’ve been better both in terms of battery life and antenna strength. I had to recharge my old iPhone twice a day and there are gaps and inaccuracies in the GPX files. The good thing is that any inaccuracies have supporting data in the form of geotagged photos and foursquare check-ins.

From there, it’s really easy throwing everything up onto Google Maps/Earth, but what would really be neat is an interactive front end with Feltron-eqsue graphics. Make it completely automated and you have yourself a hit. 

In a way this is already here in the form of fitness apps with RunKeeper, Nike+ GPS*, Google My Tracks. There’s also Twitter for iPhone (“nearby tweets”) and Flickr. Better yet, take Daytum and add location services and automated tracking. What I am thinking is something that automagically tells a story / replaces a blog post** by 

  1. pulling data from a portable device (smartphone really) that logs GPS coordinates, steps taken, and whatever else hardware the is capable of.
  2. aggregates data from other services like Flickr and Mint
  3. runs statistics*** on 1 & 2
  4. spits out cool interactive graphics from 3.

* I’ve been using Nike+ (for runs) and Nike+ Active (daily steps taken) for about a year and still enjoy the stat tracking and graphics. Someone else get this so we can compete.

** I paused for a minute to think if this blog post be automatically generated. How meta. But no, what I am referring to of course are blog posts that recount events e.g. “Today I went to X, saw Y, and bought Z.” 

*** I have an Excel spreadsheet of some data and that’s been fun as well even with my basic knowledge of statistics. Food was my biggest expense. Avg for a meal was $12 with StdDev 5. Most interesting regression results: Yelp ratings for the places I ate at (taking into account # of ratings as well as the score) tended to be higher when I had a meal closer to the average price.

Again, in a way this is already here in some respects, but not quite the implementation I’m looking for. It can’t be too far away. Ten years ago, most people still used film cameras. Now everyone has a DSLR that can record photos and videos. Three years ago most people had a phone that could record a different sort of data—tweets. Now a lot of people have smart phones. Soon everyone will have one that has enough battery life—all that’s needed really—to automatically track and process the necessary information.

At Shake Shack (★★★★) $10 got me a Shack Burger and Strawberry Shake. The wait was 30 minutes and I got my food 12 after minutes later. I took 19 photos in the area.

Notes