Folding the Third Dimension into the Second

The birth of the graphical user interface has revolutionized the computer industry. The GUI introduced a new way we interact with computers. It has created a 2 dimensional plane of interaction. Where the computer is used to easily run millions of commands simultaneously without a delay in time.

Gesture control is become more and more realistic. Once upon a time it was something we just saw in sci-fi movies. Now it’s a reality. As cool as it might seem, there is a slight problem; it doesn’t provide efficiency and certainly doesn’t provide additional productivity. This issue needs to be addressed for gesture control to gain more popularity and soon become a standard just as the mouse.

The root of the problem is there exists a disconnect between the human interface device and the graphical user interface. When the interface we currently use, which was what revolutionized the operating system and the way we look at computers, is set to be controlled by a two dimensional manner. A mouse is used on a flat surface that represents what we see on a screen. Both the interaction mechanism and the interface are in the second dimension. Introducing gesture control becomes more tricky, whereas human gestures occur in the third dimension, while the interface is in the second dimension I have played around with the Leap motion and one of the add on application takes that into consideration, where essentially you move the 2d plane into the air. There is a window in front of your screen and that’s how you are able to transfer your motions into actions on the computer. It’s like controlling a touch screen laptop, without having to touch it.  As I own a laptop with touch screen capabilities, while using my Leap Motion to control my computer, I sometimes have to refrain myself from just touching the screen for whatever command I want.

The technology for capturing and recognizing hand gestures is not the problem. It is how to we make it intuitive and how does our interface represent that third dimensional space. The hype in tablets and multitoucch display provide has taught users how to interact with a touch screen. Users are now informed with how to use such devices. Does gesture control build upon that knowledge to create use of such technology? Or should the user interface be reveloutionalize and get a complete makeover for such interactions to become meaningful and worth the change?

Other things to take into consideration are whether the people are ready for such change? Is it worth putting forth the efforts of redefining computing and then not be adopted by the people. How would gesture control be implemented in the business place, for you will need more space to control a computer, and definitely more stamina.

/by

Network Topologies

Network topology is nothing more than just describing the layout of any given network. There are two types topologies we are interested in, the physical and logical. The physical topology refers to the physical distribution of the system, where each node is located physically. The logical topology refers to operational relationship between different nodes on the network, how the electricity flows. There are four major network topologies. The mesh network, bus topology, start topology and ring topology.

Mesh network

In theory the mesh network (full mesh to be exact) is the best network topology. It provides multiple pathways between the nodes. In a full mesh every node is connected to each and every other node on the network.  The downside in a mesh network is the cost of instillations; a large number of connections are needed as the number of nodes increase. The formula for the number of connections need is described by the following:

# of connections = nodes x (nodes-1)/2

Below is a graph of the function above, given the x axis is the number of nodes and the y axis is the number of connections needed in a mesh network. It is visible that as you add more nodes the number of connections needed grows really fast. Once you reach 10 nodes you need more than 40 connections.

number of connection function

Bus topology

In this topology each node is connected to a single BUS. When a node transmits a message to another node it broadcasts it to every other node and only the node that the message is intended to will listen to the message. You might notice two problems with this topology. The first is security oriented; if a node decides to listen to all the messages on the network it would be able to pick up all messages even if it was not intended for that particular node.  The other problem that arises is what happens if two computers broadcast a message at the same time? The messages collide and thus it would be corrupted. To solve this problem a collision detection scheme is implemented in which if two messages collided each will resend the message at a random different time again. When the traffic increases collision increases and the bus topology is not useful as much. One of the main advantages of a bus topology is the low cost of installation.

Star topology

The star topology is that in which all nodes are connected to a central device. The device is responsible of connecting the two nodes together. This topology is mainly used in local area networks, sometimes in metropolitan area networks as well. The star topology is very useful in which it creates a full duplex connection between two nodes. The down side of this topology is if the central device fails the whole network fails.

Ring Topology

As the name suggests each node is connected to the next node, and the last node is reconnected to the first node creating a ring. The ring topology is inherently unidirectional, and the downside of that is if a node wants to connect to the node before it the message must go through the whole network to arrive. This topology was used in the past to enhance network performance. Now a days this topology is used in fiber optic backbone.

/by