eChunks 2

“In the ideal process, you'd first conduct competitive testing to get deep insights into user needs and behaviors with the class of functionality you're designing. Next, you'd proceed to parallel design to explore a wide range of solutions to this design problem. Finally, you'd go through many rounds of iterative design to polish your chosen solution to a high level of user experience quality. And, at each step, you should be sure to judge the designs based on empirical observations of real user behavior instead of your own preferences. (Repeat after me: "I am not the Audience.")”----Jakob Nielsen

        In this article, Nielsen illustrates a new way to increase usability: “Parallel & Iterative Design + Competitive Testing = High Usability”.  Before reading this article, I seldom think of conducting competitive testing to get to know the users’ needs. Or say, I have little knowledge about competitive testing. Questionnaire, observation, and user testing are the methodology we often use to gain feedbacks. But these are the testing for a design prototype. Competitive testing is conducted before we actually design the product. I can see the advantage by using this method to get deep insights into user needs. That is, by analyzing existing software/products, you pick up useful information among them, no need to spend resources by creating a new one only to find out the users actually do not need that.  And this method is essential because it may save large amounts of human and capital resources to develop initial models in the first place. When design a new product or apps for customers, every step in design should be considered and supported by strong research evidence. Never take things as granted. Every decision should be supported by research.

        After competitive testing, conduct parallel design to expand the ideas. In a parallel design process, either a person or a team can quickly scratch a few version of the initial interface. Nielsen mentions that the interface in this process needs not to be detailed. It can be a rough prototype. The idea for this design method is to explore many possible interfaces for the software. Notice that Nielsen also mentions 5 is probably the maximum.  After testing these different interfaces, merge the good features into one instead of identifying a winner.

        Finally, proceed with iterative design to polish the merged design.

        With Nielsen’s methodology for increasing usability, we can avoid being trapped in a “best” idea. 

eChunks 1

“As machines start to take over more and more, however, they need to be socialized; they need to improve the way they communicate and interact and to recognize their limitations.” (P.9)---Norman

        Norman’s thought of socializing machines is quite interesting to me. As I read an article before, which talked about human beings are socialized animals that need to be constantly involve in social context in order to fully develop our mental intelligence. And now Norman is talking about machines need to be socialized so that to better communicate and interact with people.  That seems a whole new era for technology since machines from nowadays only do what they can do as human program them and that is mostly obedience.

        When talking about socializing, self-awareness first came into my mind. Socializing, to some degree, is expressing your individual views on certain issues. In most cases, it refers to having
conversation and interaction among one species of which they have the same living background and use the same language system.  Maybe there is cross-species socializing like human interact with cats and dogs or animal trainers, but both sides of these socializing are self-awareness creatures who have their own will. How about machines? Do they have self-awareness that can in fact having real conversation with human beings? A real conversation that is generate through their mental process instead of programming. If so, is that a machine as we known anymore?

        If a certain machines have the ability of communication and interaction, that would be an artificial intelligence product for sure. And if that were the way, even if the AI knows exactly what their limitations are, would they obey to those limitations? Or they are so intelligent to know that they have the power to disobey human and may even harm them?

        I am not saying that I am totally disagree Norman’s point of making machines being socialized. It is just my concern that if machines socialize and communicate they will evolve at a rate we cannot imagine and cannot control.