R2: Design and Usability from DFAB

I agree with all Armando that the principles are timeless–principles are meant to be However, I think the more specific standards are constantly evolving.

The principle of familiarity particularly stands out to me. The text mentions that familiarity can enhance a new user’s ability to effectively interact with the object, since the user is able to bring in existing knowledge into this interaction. When I watched the assigned video last week, I felt that the new grocery carts were sure to fail from the marketing standpoint because they were too different from the present designs. Overall I loved many elements of the concept, but felt that users and potential purchasers may not receive it well because the designers were trying to introduce too much at once (a walkie talkie, a new check-out system, use of small baskets only). This seems overwhelming to me–way too much for people to process at once. This brings me to what I felt the text was missing–the affective component of familiarity. The authors do a great job of describing cognitive components (e.g., being able to bring in past experiences into the new interaction), but I think we cannot ignore the affective aspect of human-computer interaction. I think that a huge benefit of familiarity is that it reduces fear and apprehension in users. As much as I love crazy creative divergent ideas, I think that they often have to be “reeled in” a bit to be embraced by the public. Incremental changes help a lot; I’m thinking of the Vibram 5-finger shoes as an example. They were successful with a niche population. Now many have taken the barefoot shoe concept and made it more traditional by getting rid of the 5-fingers, but keeping the other novel aspects (e.g., lightness, breathability)–and these seem to be even more successful because the average person is open to it–it’s not TOO different from what they are used to and thus helps to ease fear.

I have to echo Armando’s comment and say writing out personas and scenarios is an amazing tool to start scrubbing out edge cases that one may have never thought of. It also leads into the ideas of constraints – you can possibly separate what will work for enthusiasts and what will work for first time users, and what will work for one group and not the other. While trade-offs may emerge in the design process, I think that having real and tangible decisions will at least drive the design process in a productive direction, even if its wrong.

Human behaviour can present an almost unlimited amount of possibilities for interaction with the product and design, so I find these principles as a way to set definite guidelines to make reasonable design decisions around. In a creative process, it can be easier to have some constraints to guide the outcome rather than be given a blank canvas and be told to just “go”

It is undoubted that the goal of interaction system design is to maximize the usability of the software product. There are three usability principles listed in chapter 7, which are learnability, flexibility and robustness. It makes perfectly sense that no one wants to use a system with a great aesthetic interface but slow or error-prone. In order to achieve the goal, we can follow the design principles to increase the usability. I agree that we should always stay focused on the user and understand our users. In requirements stage, we should think about users’ current situation and what they want to achieve, by interviewing, videotaping, talking to, or directly watching them, because we can’t take for granted that users would have the same interests or abilities as us. Besides these observational approaches, we might need to use imagination due to cost or some limitations. It doesn’t mean to imagine what we would do in that case, but what the user would do.
I think a good way to know your users is to combine persona and scenarios. Persona helps to describe the scope of intended users while scenarios, mimicking a real event, can force designers to think about the design in more detail and notice potential design problems. But scenarios are incomplete, because only one cannot describe all different aspects of interactions between users and computers. Therefore, we might need several scenarios to analyze all possible interactions.
When in design stage, designers still need the attitude of users focus. Now we should think about the interaction between users and the computer screen. Consider what information is required, what kinds of layout, color, order, alignment or decoration are needed. These design concerns lead to reduce the likelihood of human errors and increase the usability. Just like Apple’s multi-touch product, we don’t have to take much effort to learn how to use them, because they are accordant with our understanding of physical world.
Besides understanding the materials – computers, humans and their interactions, we still need to think about the economic benefits, because the final software product should produce profits. In some cases, you have to balance the trade-off between usability and a perfect system.

I think the design and usability principles in these chapters are very sound, which can help us have clear procedures and evaluation methods of design. Especially, I love the “customizability” part. In my own experience, many softwares lack of adaptivity. They treat expert users as same as novice users. For example, there are no shortcuts for expert users to accelerate their tasks, which lead to inefficiency.

However, with the popularity of mobile devices, new challenges for design and evaluation methodology have emerged. The design and usability principles in this book may become insufficient for our design and evaluation. For instance, people use mobile devices almost anywhere and anytime, so usage situations are complicated. People’s privacy is easy to be revealed, and people are easy to be interrupted. In this sense, how to secure people’s privacy and how to deal with interruptions should be taken into account. Also, people usually need to connect their mobile devices to other devices, to share information with others or transfer data to their own devices. Therefore, the convenience of connecting other devices also needs to be considered in our design and evaluation process. Another case involves touchscreen-based mobile devices. This kind of devices has physical user interface and other special characteristics, consequently, new principles should be generated.

Usability of a system is core to user-centered design. However usability metrics or usability specification is not enough to define “usability” of a system, as it is hard to visualize different aspects of usability at an early stage of software development. Actual measure of usability can only be made when the system is in the user’s hands or at least at user acceptance testing stage. For this we need iterative interactive system development via rapid prototyping. In my view storyboards are inefficient or outdated when we need to design multi level or complex software systems. High-level programming or Wizard of Oz concept can be used instead.

Usability can be enhanced with right principles and guidelines. Though these principles are abstract they can still help the design engineer quantify usability. Learnability, Flexibility, and Robustness are main principles with several sub principles. Having existing knowledge of computers I find synthesizability, recoverability, and customizability most intriguing.

Synthesizability- lets the user know of changes made upon his instruction. Example: If a user wants to proofread a word document he created; he gives a command to replace all ‘the the’ string with a single word ‘the’. Here the system should notify the user about every change made or the theorem would be replaced by theorem.
Customizability – allows the user to customize a layout on his will. This functionality gives a leeway to make changes to existing organization. Example: Compact, cozy, background color etc.
Recoverability- is undoing an action that was accidentally done. Features like Ctrl Z, error message before ctrl+shift+delete, recycle etc. facilitate this functionality.

As per DFAB, design and usability principles are divided into three broad categories namely learnability, flexibility and robustness. These main categories are further subdivided into specific principle.

In my opinion, a well designed system will let user interact with the system without referring to any manual. Operation of any user interface element should be obvious to user. If the interface performs in way that user think it does, it delights the user otherwise it is frustrating for the user. Hence predictability and synthesizability of interface are important for a successful design.

Responsiveness of system is also very important for the usability of the system. Not all user interaction take same amount of time. Some interaction are quick and some are slow due to system constraints. However, a good design will keep users informed about the state of a system. For example, during a file copy operation, a copy animation along with a estimated time left is far better than just a copy animation.

Customizability of a system is another important design principle. For example, a web application where user base is wide, it is hard to develop interface which will address every users need. Hence any kind of customizability which fits suits needs is a sign of good design.

Indeed scenarios are important tools of design; they involve the end-user in the design, they serve as an example of how the “artifact” that is designed will be used in the real world settings. But in order to use these tools it is very necessary to bring in the actual components of design first, an important way of doing so is Metaphorical Design many great systems that have been made till date have used some metaphor or the other, be it the desktop metaphor or the classroom metaphor, real world objects serve as the best example for designing new objects. Metaphors not only enhance learnability but they also trigger the creative cells of the designers. They also serve as a tool for the stakeholders to critique on the current design, give ideas for the future designs, enhance the overall user experience and are much more engaging.

An important principle for usability apart from the very apt principles broadly classified as learnability, flexibility and robustness in the text is “reusability”, designs should be such that they efficiently use existing components as well as become usable by other designs. Novelty is essential, but recreating a thing that can be reused can be a waste of time and effort. For example, you may not want to start building from scratch a “calendar widget” for a website even when there are tons of customizable widgets available.

Additionally involvement of the beta users, has become an important part of the design process these days, earlier users came into picture only when there was a ready prototype for them, now they play important part since throughout the process.

Usability of a system is core to user-centered design. However usability metrics or usability specification is not enough to define “usability” of a system, as it is hard to visualize different aspects of usability at an early stage of software development. Actual measure of usability can only be made when the system is in the user’s hands or at least at user acceptance testing stage. For this we need iterative interactive system development via rapid prototyping. In my view storyboards are inefficient or outdated when we need to design multi level or complex software systems. High-level programming or Wizard of Oz concept can be used instead.

Usability can be enhanced with right principles and guidelines.. Though these principles are abstract they can still help the design engineer quantify usability. Learnability, Flexibility, and Robustness are main principles with several sub principles. Having existing knowledge of computers I find synthesizability, recoverability, and customizability most intriguing.

Synthesizability- lets the user know of changes made upon his instruction. Example: If a user wants to proofread a word document he created; he gives a command to replace all ‘the the’ string with a single word ‘the’. Here the system should notify the user about every change made or the theorem would be replaced by theorem.
Customizability – allows the user to customize a layout on his will. This functionality gives a leeway to make changes to existing organization. Example: Compact, cozy, background color etc.
Recoverability- is undoing an action that was accidentally done. Features like Ctrl Z, error message before ctrl+shift+delete, recycle etc. facilitate this functionality.

I think prototyping is the most crucial part in the design process, since it provides us a quick method to test if our ideas are applicable or not. From the DFAB charters, the book provides three ways of prototyping: Throw-away, Incremental and Evolutionary. Personally I love the Throw-away method best since I believe by discarding and redesigning prototype is the best way to come up with good designs, but this method is time-consuming. The other two methods might be more applicable for companies or teams to design since they consider the balance between time and design. I wonder if there is another prototyping approach that could give us a fast way to generate good prototypes.

In some Test-Driven development approaches, like the Cucumber and Capybara in Ruby on Rails development, the principle of design is that “Write tests before you write code”. Customers can describe their requirements in simple English sentences by the language features from Cucumber and Capybara, and how programmers evaluate the system is to check whether it “pass” all tests or not. If the system has passed all tests, then the design is complete. If in the future the customer wants to add or remove some features, he can simply add or remove related tests, and designers just have to go through the prototype-test process again to provide a new version. I think the approaches given in the book all separate the processes of prototyping and evaluating, but the Test-Driven development method cleverly combines the two processes: we design in order to pass the evaluation. It has the feature of the Evolutionary approach, and we can discard or remove some parts of our design easily as in the Throw-away approach. Although “passing tests” might not be equivalent to “designing the right thing”, but it might be a helpful and productive way for us to come up with designs with good usability.

Principles are the rules that never obsolete. In the book, there are three categories of principles to support usability. Learnability, the ease with which new users can begin effective interaction and achieve maximal performance; Flexibility, the multiplicity of ways the user and system exchange information; robustness, the level of support provided the user in determining successful achievement and assessment of goals.

But from my experience, learnability, which represents the ease expectation of design, is always opposite to flexibility, which represents the complication of design. For example, customizability is one principle affecting flexibility. When system-initiated modification happens, it intends to improve the flexibility. But users have the preference to adhere the way it used to be. Think about the Facebook timeline as an example, Facebook initiates that modification to highlight the user’s recent activities and to emphasize the events time, which mean to provide higher flexibility, while millions of users dislike this change. For users, familiarity ensures learnability. What users are used to is the best design. The designer should always think about it before making any modification decisions.

Sustainability is also an important one. I am thinking of Android’s early dates upgrading strategy. Android team upgrade 4 times a year introducing huge modifications every time, which makes developers be afraid to contribute to Android. While Apple iOS upgrade once a year giving the sustainability for that one year so that users and App developers have time to get used to those new features. Recently Android team considers the sustainability principle as they slow down their upgrading pace and attract more users and developers into their market share.

There are many case studies that reflect the essentialness of these design principles. As designers, we should ask ourselves through all those principles before making any design decisions.

User-oriented design requires designing systems/products to be efficiently used by as many people as possible in as many situations as possible. It is crucial to keep in mind the different kinds of users when designing interactive systems, as every user possesses a different range of abilities and knowledge. Therefore, such design and usability principles provide a framework to facilitate the development of interactive systems and maximize the usability of systems.

Learnability/familiarity is one of the most important principles to consider as it is important for users to be able to easily adopt a new system and begin interacting with the system effectively. However, users are often intimidated to interact with systems/products that are entirely new and unfamiliar to them. In order to achieve maximal performance, users should be able to apply prior knowledge and past experiences to a novel system. For example, products that do not deviate far from the norm, but still provide innovative features over existing products tend to be more successful; this can be substantiated by Apple’s yearly iPhone releases which provide incremental updates over the previous years models, but still remain simple and familiar to those who have already used the product.

Flexibility is another important principle. Users should readily be able to extend a specific interaction to novel situations; flexibility also provides understanding of the various ways users and systems can exchange information. With the increased penetration of computing devices and multitude of devices owned by a single user, it is significant to understand the capability of systems to support user interactions for multiple tasks at a time.

When I read the design rules, I feel all the concepts are so adjacent to each other and even overlapping. For example, when the author mention Generalizability said:” the generalizability of… more complete predictive model of the…. We can apply generalization to… in some way similar. Generalizability can be… a form of consistency.” This includes Predictability, Familiarity, Generalizability, and Consistency. This kind of overlapping is not rare example, and they happen everywhere. Such as Dialog initiative includes a part of Multi-threading and Recoverability is linked to reachability. But these did not influence the comprehensive understanding the author’s main idea.

When I finish the reading I got a little bit confusion that the author gave us a generally agreeable rule and implied that do not obey the rule all the time. Like the author talk about Consistency, Consistency is one of the principles, but consistency is not always the best. The conflictions are so obvious that between consistency and moderate flexibility. However, the concept is so cool we can make our products suit diverse customer levels. For instance concerning Customizability, for lower Level customers can simply move the position of the soft button to get flexibility and for upper Level customers are able to use script language.

But even we do this, the conflictions are still outstanding, because of = to restrict the capability of the computer and = More powerful but more mistakes made by customers. The designer needs to figure out the balance. As is known, the software interfaces development dependents on hardware development. Maybe one day, when the operational 3D Holographic Projection come out, we can solve all the problems we are facing. But for now, personally, the most essential principle for us is who your customer is.

Learnability: As a young budding technology enthusiast, it does not take much effort to understand and quickly adapt to new design panels and screens. But a classic example is watching our parents trying to use these same gadgets. The very thing that is so simple for us seems like a hard task for them. Technology will no doubt keep evolving, but it should always be easy-to-use for all demographics. Xinlu said that the design should be close to intuition, which is something I wholeheartedly agree with. Also, with respect to the consistency factor, the arrangement of alphabets of the keyboard might never change because of familiarity with this standard. Some things are best left as they are.

Karen mentioned that the IDEO cart would not really sell. Sometimes, I think too many features on the cart seemed like a personalized cart, not really public store material.
While designing, the main point in our mind should be changing a situation for the better. New versions of the windows OS keeps getting developed, but quite frankly sometimes the previous version is just as good as the new one and need not be replaced. Matthew mentioned that it’s easy to spot bad design, but not as easy to spot good design until much later. One factor while considering robustness at an extreme but necessary level is the fact that lots of new products could enter the market every day after the entire iterative design process. However these fail to attract investors because of their huge capital investments and low returns. Due to this, the product’s ultimate goal of reaching the market fails.

All of the design principles stated in DFAB are important and, as Armando said – they are timeless. However I think we need more specific ones to meet the demands of present and future.

Today, for a user, a device is more about the homogeneous experience rather than just its functionality. Three principles that I think are very important to achieve this are affordance, predictability and responsiveness. By affordance what I mean here is, interfaces should be as self-explanatory as possible. Nobody wants to remember the
steps to carry out a task on a device but would rather prefer familiar icons and easy navigation hints on the interface.

Responsiveness can be explained in various contexts. By responsiveness, you could also say how responsive the system is to user actions. For instance, responsive web designs which is an excellent usability feature can make transitions between devices smoother and portable. Depending upon the user’s screen resolution and browser size, the web page breaks down completely and re-invents it layout. This again adds to the overall experience. In today’s world where there’s a plethora of smart devices, I think it is really important thing to consider the portability aspect.

Predictability is about embedding intelligence in the device, thus giving more room to human errors but still knowing what the user really wants to do. BlindType is a perfect example of this. It allows the users to be sloppy but it still somehow knows what the user wants to type. Another striking feature is the easy reversal of actions. You can erase words just by rubbing over it with your finger, just like using a traditional eraser. BlindType is also thoroughly robust and responsive in the sense that it re-aligns the position the keyboard according to user’s orientation and comfort.

As technology evolves, principles also need to evolve along with it. If not evolve, atleast new principles need to be added to the existing list.