X
Business

Could your iOS device get you wrongly accused of cheating by your spouse?

So, how easy is it for someone to grab that tracking data that your iOS devices stores on you? And once in their hands, what information can they learn about you? How accurate is that information?
Written by Adrian Kingsley-Hughes, Contributing Writer

So, how easy is it for someone to grab that tracking data that your iOS devices stores on you? And once in their hands, what information can they learn about you? How accurate is that information?

To find out, I decided to see how easy it would be to grab my wife's data from her iTunes backup, examine it, and see what I could learn from that experience.

Note: for the record, Kathie did give me permission to to this in advance, but I didn't let her in on how I was going to grab the information.

My chosen method of attack ... log in to the Windows system that she has iTunes installed on remotely, copied over the nphonetrack tool and used that to output a CSV and KML file. The process took literally seconds and I was easily able to remove all traces of the deed.

Note: Using the tools currently available to grab the data from the iTunes backup file you do need direct access to the machine or of a backup of that machine. Also, the iTunes backup file needs to be unencrypted (and in this instance it was ... something which surprised me actually).

My 'breach' gave me access to 42,012 data points going as far back as the end of June 2010 (iOS4 was released June 21st, 2010). The nphonetrack tool adds an artificial 'fuzzing' to the data, blocking all data for a week under a single timestamp and also 'quantizes' the data into grids roughly 2Km x 3Km to offer a certain level of privacy, but as I'll show you later, it wouldn't be hard for anyone with a little programming experience to modify the tool to remove this fuzzing and allow pure output of the data stored.

I then took the KLM file that nphonetrack created and pumped the data into Google Earth, which dutifully placed all the data points onto a map.

42,000 points is a lot of data to visualize, but if you have a reasonably good clue as to what your partner's been up to over time, you start making sense of what you see. You see business trips and personal trips, as well as movements around and about the local area. I can see trips done via car, I can see trips done via train.

But I can also see evidence of visits to places I didn't know anything about.

For example, it seems that Kathie has taken a number of trips to Ireland over the past few months.

Hmmmm, secret trips that I knew nothing about, eh? What have I uncovered here? Something interesting perhaps?

You bet! What I've uncovered here is just how wildly inaccurate this data can be. What's going on here is that the iPhone was connecting to cellphone transmitters in Ireland when it was in fact on the west coast of Wales, UK. The handset was claiming to be at locations over 60 miles/100 Km from where it actually was at that time. Sure, there's a technical reason why (the handset was connecting to the nearest available transmitter), but if you (or your spouse) didn't happen to know this, you could have a lot of explaining to do!

So, what have we learnt from this? Several things:

  • iOS devices collect a lot of information about your movements (I'll have more to say about this in a later post)
  • Grabbing the data is trivial if you have access to the phone, the PC the phone syncs with, or access to a backup of that PC
  • Encrypting your iTunes backup adds an obstacle to grabbing the data, but encryption can be cracked
  • From this data you can learn a lot about someone's movements over a long period
  • The data can be highly erratic and inaccurate
  • Conversely, the data can be very accurate, and the overall volume of accurate information can lead you to think that all the information stored is accurate

Remember how I said that nphonetrack 'fuzzes' the date and location data, well, if you're handy with coding, you'll easily be able to figure out how to remove this privacy guard. In fact, all it takes is editing three lines of code and recompiling ...

21-04-2011-17-27-52.png

So, what do you think? Are you happy with your iPhone logging your movements, or is this a one feature that needs reigning in?

Adrian Kingsley-Hughes, with Kathie Kingsley-Hughes

Editorial standards