Surface training

On Wednesday last week, me and Jouni (the other writer on this blog) went to Munich in Germany to attend one of the first Surface developer courses outside of Redmond. Teacher was Dr Neil Roodyn from Australia. Dr Neil has been working with the Surface team for almost a year, developing applications and giving training. This was one of three occasions in Europe.


Dr. Neil Roodyn


The course was titled Building Applications for Microsoft Surface and lasted two days. We were 16 attendees and had two lovely Surface tables at our disposal.

Dr Neil talked a lot about the Surface vision and the history about the table, how it all started and how it came to be what it is today. The work with the Surface table started back in 2001 and in 2005 it got the current look and feel. In 2007 the first public pictures appeared and at PDC in 2008 it was made available to the public. If it took Microsoft seven years to get Surface to the public I can’t help thinking about what kind of projects they are starting right now that we will se more of in 2015!

It is really important to think of Microsoft Surface as something else then just a big touch enabled screen. Cause if touch is all that you are looking for then there are much cheaper solutions! Now the unique thing with Microsoft Surface is the interaction with what happens on and above the surface of the Surface.

Thanks to the vision recognition system Surface is able to detect things on top of the actual surface. About an inch from the surface things are starting to get picked up by the cameras. This information for instance used to detect fingers and the orientation of the finger.

As usual when you attend a course and have some knowledge of the subject, the first hours or the first day is not really challenging. Although Dr Neil managed to keep my interest and soon we got to start writing code and play with the tables.

One interesting thing to see on the table compared to the simulator is when you have recognized a tagged object and move it fast over the surface. The vision recognition system is not able to track that it is a tagged object that moves and it becomes a blob instead. But when you stop moving it is recognized as a tag again and the TagVisualizer is animated to the new spot. On the simulator you can move a tagged object with a TagVisualizer at any speed with no problem.

One part of the course was focused on unpacking, configuring and setting up the table. This part was really good since we haven’t gotten our table yet. It was also great to talk to Dr Neil about different deployment and development environment strategies. He had some interesting thoughts about how to use Live Mesh with Surface that I will try.

Anyway a really great course but I would have enjoyed more time coding and hands-on Surface-time!

Contact: The act or state of touching

While being in contact with the Surface you handle more information than when just placing your finger on a traditional touch-enabled screen. A Contact in ‘Microsoft Surface language’ is the kind of object that touches your Surface screen. The Surface SDK gives you the possibility to recognize different kinds of objects. The ones you can use are Finger, Tagged object or Blob.

• Finger is self-explanatory; your finger in touch with the touch-enabled screen.
• Tagged object are objects with an identity tag placed on them. This makes it possible to interpret exactly which object is being placed on the surface table.
• Blob is an arbitrary object placed on the screen. You might have seen the demo with a wine glass with a surrounding aura – voila a blob.

Ok, so what to do with contacts. The touch-enabled controls for Surface can handle events from user interactions – that is different states of the contacts on the screen. These interactions are called gestures or manipulations.

Gestures are user activity which doesn’t work differently on whatever element is affected. A tapping or pressing action is a tap or push whether it’s a button or custom user control. Manipulations on the other hand are interactions that depend on what control is being touched and how. A manipulation user action has a context. That means that for one control a twirling motion means to rotate the control, but for another the same user interaction means to twist the control’s content.

Using the WPF layer of the Surface SDK the Surface controls are enabled to capture the following events:

• ContactDown
• ContactUp
• ContactTapGesture
• ContactHoldGesture
• ContactEnter
• ContactLeave
• ContactChanged

More about the mentioned contact events when we come back to the WPF layer in a coming blog posts.