Photosynth

A few days ago a friend sat me down and made me watch this video. In it, Microsoft shows us some software they’ve recently acquired. The software is called ‘Photosynth’ and it is absolutely amazing.

First the demonstration shows how the software can be used to zoom in, indefinitely, on an object. Objects are no longer restrained by the dimensions of the eye (which can only go down so far and out so far). Instead, a page of a hundred photos can be shown and we can zoom in on one detail and go down so far that was a dot before, now fills the whole screen and is revealed to contain, say, an entire book in digital form.

I really can’t do it a great deal of justice in words and instead suggest you watch it.

The second part of the demonstration shows something equally amazing. Apparently, with new software, it is now possible to trawl the internet and have this software recognise images and then put together a compilation of images that are related to a specific object. In the case of the demonstration they use the Notre Dame, but it can obviously be used for a great deal more than that.

Astonishingly, the computer on its own can find images related to the Notre Dame and then calculate in a three dimensional framework where those pictures would have been taken around the actual object. That is absolutely stunningly amazing and it has huge repercussions for image linking and the internet’s interconnectivity, with any picture you take being automatically linked to other people’s pictures of the same area.

The thing is, though I realised that this software was very important in a positive sense, I also understood that there is a far more sinister way to use this software. Namely, it can be used to invade our privacy in a way so far unimagined.

Right now, around the world, governments and companies are putting up ever more cameras to watch the going ons of citizens and employees. Fortunately for us, even though they might put up ever more cameras, the people that watched those cameras could only process so much information. If you wanted to process more information about an area, you needed more people, which then would cause a breakdown in communication and cost a great deal more money. This protected the individual to at least a certain extent. Big activities would be noticed, but us small fries could still do what we liked in relative security.

Now (or very soon) that will no longer be the case. Give the computer enough computational power (something that’s much easier to do than to hire more people and have them watch screens) and it can find whatever you want it to find. Looking for somebody? Just give the computer a couple of pictures of that person and it will scan all cameras, internet pictures and even sketches to find that person. You would never be anonymous again.

Some would then argue that that’s for the better, seeing as it could be used to find criminals and other dangerous individuals, but what happens when that kind of software gets in the hands of oppressive dictatorships? What happens if a group uses this software for political ends?

It would mean the end of any possible opposition. I don’t know about you, but that worries me slightly.