The finger seems to be the input device of choice judging by the success of the touch screen equipped smart phone and slate portables. But don't you just wish you could use a pen to point more accurately than your finger can manage? Microsoft Research thinks that you might be right.
The pen has still got life in it yet and according to Microsoft research complements the finger perfectly. You could say that pen+finger is greater than the sum of their parts.
Microsoft Research has long been looking into the intricacies of touch-based computing. Even though Apple got to the mass market and made touch successful with the iPhone and later the iPad it was Microsoft which pioneered, unsuccessfully, tablet computing. Early Microsoft tablet machines were pen based and with the rise of the touch screen you might well think that the pen is nothing more than a memory of times past but...
Microsoft Research, or more specifically Michel Pahud, thinks that if you use your left hand as a multi-touch input device and use your right hand as a pen wielding pointer then you have the ideal user interface. It is a bit like playing the guitar - both hands don't do the same thing or play (pun intended) the same role. In this case the idea is that one hand can be used to select menus and modifiers and the other hand can do the detailed and acturate location-based work with the help of a pen pointer.
The point (pun intended) is that the un-pen hand has big fat fingers but up to five potential input devices and the pen hand has a single input device that can select a location far more accurately than a finger.
Does it work?
You have to see the video, which is part of an on-going series of tours around the Microsoft Campus, to judge the outcome.
It is fairly convincing but less so with the smaller handheld devices.
The big problem is that they simply haven't tackled the main disadvantage which could be summed up as "hey dude where's my pen?". The finger has the big advantage that you never have to go looking for it.
One of the most common applications of AI or advanced stats, depending on how you look at it, is the recommendation engine. All too often the first impulse is to build your own - but it's tough and th [ ... ]