Block 3, Studio 1 Smart Homes

1. After doing a Google search for ‘smart homes’ I found this website http://www.smartkontrols.co.uk/ on this website the about in the different things that can be controlled remotely in your house, these mainly being the Heating, Air-Conditioning, and Lights.

They also have a section on user interfaces, the interfaces included are:

• Sensors – sensors are used to record the current temperature in rooms of the house
  

• Wallplates – A collection of buttons that user can use to turn the heating or air-conditioning up or down and to turn the heating off in rooms not being used. Turn lights on and off, have preset schemes where certain lights will come on at a press of a button and/or could be used to turn on lights for certain route i.e. living room to the bathroom. Wallplates can also be used to open curtains, blinds. These can also be touch screen devices which can control a larger variety of things in the house.
  
• Remote controls – Wireless devices with buttons on them, These can control lights, curtains, TVs, Radios etc depending on the type of controller and how it is set up.
  
  Mobile phone – New technology allows any standard mobile phone to “text” the house, obviously restricted by the home owner. This type interface will allow the user to turn lights on or off, and set modes for when they holiday mode for an heating ventilation or air condition system they use. The house will text the user back to tell them what action it is going to take.

2. If I was going to evaluate one of these systems for usability. I would test the wallplate or touch screen devices. Then I would probably use Usability Metrics, this is because a system like this that is fully functional is set up for this type of evaluation. As you can set a number of goals that the user needs to complete, and while the users is completing the task you can record metrics like time to complete a task, percent of tasks completed, number of explicit actions. This could be used in as part of usability engineering as the metrics can be used to prove the system has what it takes to get the job done.

I personally wouldn’t go for a heuristic evaluation on a product that is already on sale, because this sort of testing picks up more problems to be sorted early on in the design phase. I’m also not sure if I would use the talk aloud protocol because I think that the metrics would find where the errors are happening and that you would not need the user to tell you what’s going on as well.

3. I would first of all I would establish what it is I was testing for. Then I would set a list of tasks or goals that I would want the user to complete. While creating these goals I would also have a list of metrics I wanted to record on each task to give me my quantitative data. These metrics would include:

Time to complete each task

• Number of button presses (actions)
• Percentage of tasks completed
• Number of tasks uncompleted and which tasks they were
• Number of errors
• Number of times asked for help

I feel that these metrics would give me enough data to find correlations and similarities behind the tests. I would go through the tasks I’ve set myself, or get another expert user to do so. During this stage I would be measuring the metrics that I wanted use, or if I was running through the system I would get some one else to do it for. This will gave me the baseline of what to compare the users to. I would then find 5 users over the age of 30 as I feel that after that age more people will buy these sort of things, or move into a smart home.

Then I would let the users run through the test while video recording them, so I can refer to this video later to get correct timing and counts. I may also set up a small camera on the top of the wallplate to record facial expression as this would allow me took look at users’ satisfaction with the wall plate.

When I have finished the testing with the users I would try and find similarities within the data as these would be the key issues that I would recommend for improvements