I love their games, even if the learning curve looks somewhat like a cliff...
It's important to realize though, that they're publishers and well as developers. Of the five games that are listed for 2010 release on their wikipedia page, only Victoria 2, and perhaps Magna Mundi are actually developed by Paradox themselves (I say perhaps for Magna, since there's nothing on their website, and I can't imagine it's going be released in 2010).
The games that they're best known for developing are the Europa Universalis series, Hearts of Iron series and games like Victoria and Crusader Kings that fit into the same universe. They're all grand strategy games with huge amounts of detail - and they've all got Mac ports as well!
I doubt we are seeing anything new here. I assume they just use the accelerometers to determine how much they should crop away from the current sample, and then in the end stitch everything together.
It doesn't do anything like that - what you're talking about is a form of image stabilization where each frame is sharp, but the frames move slightly compared to the frame before and after since the camera is moving slightly.
This is about deblurring by working out how the camera moved while the picture was being taken and then reversing that effect back out, which isn't very simple at all.
Think about a picture of black circle on a white background. Then apply a directional blur to it. Now work out how to get from that blurred photo back to the original. By knowing how the camera moved while the picture was being taken you now know the direction of the blur, which makes the problem a lot simpler, but it's still pretty hard to try to get to the unblurred version while losing as little data from the picture as possible.
I think that after reading quite a few of the very anti-copyright posts above, I took an overly negative reading of your post - which reading it again is not the case.
On most levels I think that we pretty much agree with each other, we're just coming at it from slightly different points of view. However, I think that you're confusing copyrights with patents to an extent.
Copyright isn't meant to help creators at all, except in an incidental manner. The goal of copyright is to promote the progress of science, which consists of 1) causing works to be created and published that otherwise would not have been, and 2) having any restrictions on the public with regard to those works be as minimal and as short-lived as possible.
The above is the purpose of patents - to ensure the progress of science by ensuring that works are created and published such that they end up in the public domain after a period of time. This, as it turns out, is in many ways the opposite of what copyright was designed to achieve.
Copyright law as we think of it now, came into being in 1710 in England. At that point, it was intended to protect an author's 'natural right' to benefit from his works first, and for the work to be placed into the public domain second. This came into being as both a reaction to the monopolistic practices of the Publisher's Guild (the Stationers' Company) at the time, and as a response to the unregulated copying of texts at a time when printing was seen as a threat to monarchy.
When the Statue of Anne was passed in 1710, the right to have exclusive control over the publication of a work was moved from the Guild to the Author, and a limit was placed (initially 14 years plus 14 years) on how long that right would last. There was no mention of the limit existing for the good of the public - it would seem that the view was that after that exclusive right had expired, other people should be able to make money from it as well.
None the less, I think that we're mostly on the same wavelength here - especially when it comes to the point of view that copyright as it stands is in sore need of reform.
Never trust an operating system.