One of my longtime interests has been “personal telemetry,” or, in the current vogue, “the quantified self.” Since my early involvement with the Plazes project I’ve had a particular interest in “geopresence” – the record of where I’ve been, when. My breadcrumbs, in other words.
Over the years I’ve been dropping digital breadcrumbs in a variety of ways that I hope to aggregate, archive and develop tools for the exploration of:
- 10,973 Plazes check-ins from 2004 to 2012.
- 2,176 Foursquare check-ins from 2009 to present.
- 6,245 Google Latitude records from 2010 to present.
- Some proportion of the 20,010 tweets I’ve posted to Twitter (tweets where my Twitter client has added geolocation data).
- Some proportion of the 34,839 photos in my iPhoto library (photos with geolocation in the EXIF data).
- Some proportion of 60 months worth of Metro Credit Union bank statements (transactions that I can attach to a specific place).
- Some proportion of 48 months worth of personal MasterCard statements (transactions that I can attach to a specific place).
- Some proportion of 60 months worth of corporate Visa statements (transactions that I can attach to a specific place).
As a starting point, I’d like to work to convert all of these streams into KML Placemarks, which seems like a nice widely-adopted XML standard that can capture both the time and location of a “geopresence.” Here’s my first Foursquare checkin, for example, as a Placemark:
Reinvented HQ Not sure that I completely understand Foursquare - seems like Plazes, but harder to use Mon, 26 Oct 09 18:12:56 +0000 Mon, 26 Oct 09 18:12:56 +0000 1 1 relativeToGround -63.12968790933351,46.236265592340494
This was easy to grab because Foursquare has a handy page that will export all your checkins directly into a KML file. Twitter appears poised to release the entire back-catalog of tweets for each user any day now – there were tests of this in the wild in late 2012. When Nokia shut down Plazes they allowed Plazes users to export their complete history as delimited ASCII. Google Latitude provides for of history as KML (albeit via a little URL hacking to get the date range right). The other streams will require some PDF file parsing and some intellligence to convert credit card and bank statement description lines into georeferenced locations.