Reading Seminar 2- Denny Lekic

Notes for reading seminar two

The first chapter we read, chapter 13 explains key concepts about evaluation and why it is done. It goes on to motivate that evaluation is a major key due to the fact that user experience is the most important factor in term of market value. Next, we have different types of evaluation which basically is in what environment the evaluations take place. On a website it is easy to record data, and as such, companies have a Natural Setting involving users. There are other types of setting such as a controlled one etc. but the natural setting avoids interference with the evaluation process and often means the company is data logging, observing and using focus groups.

Chapter 14 then goes on to explain usability testing is something that a lot of computer programmers should be somewhat familiar with. It is a test where one measures heavily quantitatively. These are often things like: time to complete a task, typical errors, number of errors per unit of time and so on forth. The chapter also goes on to explain what remote testing systems are, that are cheaper and often more usable for profit-driven companies Moreover field studies are a way of letting the evaluation be involved in the world very practically and the aim of field studies is to discover how people interact with technology in the real world.

Chapter 15 introduces heuristic evaluation and what that means is that the evaluation is done on people other than future users(they can be but they don't  have to be) and rather than that they might be experts that know how the users will interact etc. A good example are game testers that strive to discover as many bugs as possible and this is used as a compliment to regular evaluation and not a substitute. There are factors that go into play in heuristic evaluation and they are many but a few are: user control and freedom, error prevention and visibility of system status.

Lastly the chapter discusses analytics and basically means logging user data to see how they react to what,where and when. This is usually done after the fact but there are some methods such as Fitts' Law that try to predict what will happen and it is a test that measures time taken to "click" your way through a webpage, and what is interesting about this is that this test somewhat evaluates how good and user-friendly the design is because it is affected by visibility of interface, size of buttons, how far apart things are etc.

Question for the seminar:

In which industries is heuristic testing not a viable option, if there are any?