Microsoft Chairman Bill Gates once talked about putting a PC on every desk. Now Gates is talking about turning the desk itself - or a tabletop - into a computer. Microsoft is set to announce an ambitious new computing category today called "surface computing" to try to make it happen. Microsoft Surface previously known as Milan, technology behind Surface is called Multi-touch. If you go back to history you can find that this technology was evolving from past 25 years. It was came from University of Toronto (multi-touch tablets) and Bell Labs (multi-touch screens) in 80's. However Microsoft brings this idea into reality and introduce to market.

Surface is nothing but a 21 - by - 42 inch table with base acrylic cover (touch screen). Inside this windows Vista PC along with arrays of camera which is mounted beneath the screen. Infrared enabled cameras will track the activities on touch screen and these signals are sending back to operating system for processing. Here users can interact with machine by touching or dragging there fingertips and objects such as paint brushes across screen or they can even can keep there credit card or an object with barcode.
There four main components being important in Surface's interface are direct interaction, multi-touch contact, a multi-user experience, and object recognition.

Direct Interaction:
Users can actually "grab" digital information with their hands, interacting with content by touch and gesture, without the use of a mouse or keyboard. Surface computing images and then there are things that are just plain fun, like a program that makes the surface look like a shallow pool of water-touch it and you make ripples as if you're skimming a real pond. The cleverest trick so far is a demo program that uses seemingly transparent tiles as puzzle pieces. When you put the tiles on the table, the computer reads hidden codes and lights each of them up with a section of a video image. It's quite a kick to physically manipulate those puzzle tiles to complete a picture in motion.
Multi-touch content:
Surface computing recognizes many points of contact simultaneously, not just from one finger like a typical touch-screen, but up to dozens of items at once. Surface is optimized to respond 52 touches at a time. Some real time example is that you can take a digital camera that's Wi-Fi enabled, put it down on the tabletop and the machine recognizes it and downloads the photos. Then, you can interact with them much like actual physical photos-you can pass them around the table, shuffle them into piles to sort them, pull on the corners to zoom in or out. It's intuitive, quick, and brings a fun social aspect to a task (photo editing) that can be the very definition of tedious
Multi-user experience:
Several people at once can interact with the Surface tabletop - to play games, choose music or whatever. Below the tabletop are cameras with infrared filters to sense objects plus custom software built around Windows Vista. Projectors display what you see on the surface.
Object recognition:
Users can place physical objects on the surface to trigger different types of digital responses, including the transfer of digital content.
This technology allows non-digital objects to be used as input devices. Credit cards, for instance, could have a bar-code-like tattoo that opens up your Internet account to the surface computer. Your photos could automatically be imported to the table, and you could grab the ones you like by simply touching it and moving it where you want. By putting two fingers on a photo and spreading them apart, the photo gets bigger; pinching two fingers on a photo makes it smaller. The same can be done with documents, and the same techniques could navigate a map in Microsoft's Virtual Earth.

Prices will reportedly be $5,000 to $10,000 per unit. However Microsoft said it expects prices to drop enough to make consumer versions feasible in 3 to 5 years.

Find more resources on surface computing here