bbtmn

iLog

Thursday 28th April 2011

I’m a little late to the party on this, but there’s been lots of coverage lately about iPhones “tracking” people. The original script that was released to look at this data was Mac-only, and at the time I couldn’t be bothered to hack it to work with Windows machines, so I couldn’t look at my own records. Happily though, other people have done it instead, so I’ve been able to have a look at the data from my own iPhone.

The above image shows every data point in the database, since I first got my iPhone last year. And from that, it looks like there’s some quite detailed data there. Certainly, from memory, the above map doesn’t seem to be lacking anything (although even at that level, I can see some anomalies, places I haven’t visited).

There are obvious concerns that arise from the concept of your phone logging everywhere you go. Clearly, from the data you can get a rough idea of someone’s movements. Each of those data points has a date and time associated with it, so from that you could build up a rough idea of where in the country I was on a given date. But can you tell more than that?

This all seems to come from WiFi and phone masts, so I assume that the most accurate information (in terms of location and time) would be in places with the most WiFi hotspots and phone masts. You can see from the image above that I’ve visited London in the last year. I went there for a day last August, so I thought I’d look to see how accurate the log is.

Now, London probably has the highest concentration of phone masts and hotspots in the country, so I assumed that the “tracking” data would be pretty accurate. In fact, it’s quite mixed. Right off the bat, there are two clumps of data points in areas that I didn’t visit (around South Kensington, and around Shoreditch). Additionally, there’s quite a spread of data points on the top-left of the map. If you look closely, they vaguely (very vaguely) track the train line that I used to travel in and out of London; and in fact the timings do coincide with those journeys.

The stuff in the middle seems to be more accurate. I had an interview that day at the Royal Academy of Engineering office, which is near to St James’ Park; there are a clump of points around that area, because I got there early and so walked around the park for a little while. I also walked upto the British Museum from there, and as you can see there’s a big trail in that general direction.┬áDuring the day, I also spent some time in Hyde Park (time to kill…), so I find it quite interesting that there’s very little in that general area.

Now, looking at the data, I think it’d be pretty hard to work out exactly where I went, without already knowing it beforehand. Many of the data points have the same timestamp (I assume the phone fetches a bunch at the same time), and as you can see there are more than a few in places that I didn’t actually visit.

To expand on this, here’s a similar map showing a visit to Cardiff, last month.

In this case, I’d driven down to Cardiff to pick up my younger brother, and was there for about half an hour. The exact journey I took was West along Eastern Avenue (near the top of the map), south-east down Whitchurch Road and Crwys Road, and then onto the street my brother lives on; and then the same route in reverse about half an hour later. Looking at the data from my iPhone, it’s pretty much impossible to tell precisely where I went.

I’m not saying this data couldn’t be useful. It can paint a broad picture of where someone has been, and if that formed part of a larger investigation into the person then that information could be invaluable. But to gain access to this requires access to the phone, or the computer it syncs with, which isn’t always possible. And, as I say, it’s only a very broad picture. Looking at all of the data from my phone, there is no way that you could even work out where I live, for instance. If someone got hold of my phone and took this information from it, then it’d be fairly useless to them.

When details about this first surfaced, I was a little unnerved about it. However, having looked at it and seen the level of inaccuracy, I think that much of the negative publicity that Apple has received has been a little over egged. It’s a minor concern, especially given that mobile phone networks already record locational information as part of their service (and I’d bet it’s a fair deal more accurate than what’s stored on the iPhone). It’s a minor (and interesting!) mistake, which Apple are now fixing.

Posted at 12:40 am | Posted In: Technology Tagged:

2 Comments:

Lucy

Thursday 28th April 2011, 10:41 am

H’intriguing. I don’t have an iPhone (obviously?!), but I’m willing to bet that if I plotted my movements from the last year, my map would cover pretty much the diagonal half of England that yours doesn’t!

And presumably other people can’t access your log without either your iPhone or your computer?

Dickie

Friday 29th April 2011, 7:19 pm

It’s quite interesting seeing the spread like that. I feel that I don’t really get out enough sometimes, so it’s quite nice to see that I’ve actually managed to branch out more than I thought.

Anyway, yeah, you need access to the phone/computer to get at the logs. Which could perhaps cause issues if someone lost their iPhone. Although as I understand it, it’s not a trivial thing to get at the data from the phone; it’s much easier to get at it from the backup files on the synced computer.

Write a comment: