On “smart” computers

Computers are everywhere in our modern society. And they keep getting better, faster and smarter with each new model to hit the market. Or at least so we’re told. But two simple facts don’t change:

1. A computer – any computer – is only as smart as the software it’s running.
2. Unless the software corrects for known errors, the quality of the data coming out is only as good as the quality of the data going in.

A few weeks ago, my esteemed employer went and bought me a new phone – and a “smart” one at that - a brand new HTC Desire S. I got to pick it out myself and it was actually the cheapest HTC on the list I got to choose from. I like to keep it within reason, regardless of who’s paying, and my research says this phone is plenty good enough for my purposes. So, here I am with my new HTC and those who have been around me or followed me on Facebook or Twitter will have already seen it in action or heard / read comments about it – one of the first and most often repeated being that it really is a smart phone, but it also eats battery like pop corn.

When I first got this phone and turned it on for the first time, one of the first things it did was to figure out where I was and tell me what the weather was like there at the time – just in case I hadn’t already noticed these things myself. It also provided a detailed and - subsequently proven - pretty accurate forecast for the next several days, which of course could be more interesting and useful in most situations. Better yet, the phone can keep track of several different e-mail accounts, contact lists and calendars so I can have “everything” accessible in one place and that place is right there in my pocket and always with me. And yes, this can be very convenient and very smart, but also potentially a major security hazard – my colleague Per Thorsheim can tell you more about that side of the story. His blog can be found at  http://securitynirvana.blogspot.com    I will not go into more detail here except to say that I put “everything” in quotes because I myself don’t put everything on my smart phone – and neither should you. When deciding what to put on your phone, you should always keep in mind that what’s convenient for you could also be very convenient for the totally wrong kind of person if he or she ever gets hold of your phone. Therefore, as a rule of thumb, if it’s secret and sensitive, it shouldn’t be on your smart phone. Period. And enough of that (for now anyway).

Back to my basic point about computers only being as smart as they’re programmed to be. As I have already mentioned, my new phone has the capability to determine my location. It has a GPS function, but it’s also possible to use information from the mobile network itself and/or any wireless networks in the area to determine my whereabouts with a variable but usually pretty good degree of accuracy. Also, the phone can sense movement: it knows when I’m moving around and when I’m staying put – or at least it has the ability to. When I’m not actively using it to navigate, I leave the GPS off to conserve some battery. As a result, my phone often doesn’t know my exact location, but then again, I don’t need it to. Besides, like I said, it’s usually got a pretty good fix on my whereabouts anyhow.

However, yesterday I had my first experience of the phone totally misplacing me on the map, when I drove in to my childhood home in Andvik, Masfjorden and my phone claimed I was near Hosteland – on the other side of the fjord. It said “+/- 4 673 meters” and drew a big circle around my plotted position on the map to show the area of uncertainty, but even so, it missed me by about 300 meters – I wasn’t even inside the circle at the time.

So, evidently, the local mobile network alone does not provide solid ground for accurate navigation, which is why we have a GPS function for that purpose. I stepped outside and turned it on for a bit - and sure enough: the thing took about 5 seconds to figure out exactly where I was. But, since I already knew my location, I turned GPS back off to conserve some battery.

About an hour later and less than 20 meters from my last known GPS plotted position, I discovered that the phone had attempted to update my location – using data from the mobile network - and reverted back to saying I was about 5 km away and on the other side of the fjord…