The first part of this chapter focuses on user’s
thoughts and feelings when using a device. If this device is not designed well,
the normal reaction is to blame themselves or to give up and assume they are
not good at handling that particular device. Dr. Norman then explains the seven
stages of actions. This raises questions meant to help implement the principles
of good design mentioned in chapter one by bridging the gaps of execution and
evaluation. At first, I thought he was saying that it was never the user’s
fault but only the design. However, he then clarified that if an error can be
made a user will make it. So, it is the developer’s job to minimize the
possible amount of errors. These errors are executed because of a faulty theory
that a person made. Dr. Norman uses physics and a thermostat to illustrate
this. However, these examples did not have that great effect on me since my
theory correlates with how it actually works. Though, I can see how it can be
misunderstood. Whether a correct or incorrect theory, we use the idea to
interact with the world. This interaction can be divided up into seven areas:
forming the goal, forming the intention, specifying an action, executing the
actions, perceiving the state of the world, interpreting the state of the
world, and evaluating the outcome. This
process was interesting to try and apply to my own life by picking an action
and dividing it up into those seven stages. Overall, this chapter helped
explain people’s reaction and thought process towards a bad design of everyday
things and will change how I think when I encounter a device that seems to be
especially tricky.
Reaction
to Chapter 3: Knowledge in the Head and in the World
In this chapter, Dr. Norman focuses
on how people remember something. He does this by comparing knowledge in the
head with knowledge in the world. He understands that “precise behavior can
emerge from imprecise [head] knowledge”(55) because there is knowledge in the
world, great precision isn’t even required, and natural and social constraints
are present. Poems and songs are examples of constraints that he uses. Our
memory is created through arbitrary things(which is the hardest way to remember
things), meaningful relationships, and explanation(which makes remembering
easier). For most of the chapter, Dr. Norman dives into various aspects of
memory and knowledge. However, I couldn’t tell how it related to design until
the very last pages of the chapter. He recognizes when head knowledge for a
certain product isn’t there, world knowledge is created through labels. Labels
imply that natural mapping was not applied very well. Therefore, “wherever
labels seem necessary, consider another design”(78). I’m assuming he will build
on this chapter later in the book and will mention how the information about
memory is important for how you applicably need to go about designing a
product. In summary, I enjoyed learning about how our brain works and how to
make myself better remember tasks and data, but it was currently hard to see
how this was relevant to design.
Reaction to chapter 4: Knowing What to Do
Chapter four covers how people know
how to interact with certain objects. The answer: constraints. There are
physical, semantic, cultural, and logical constraints. When explaining the differences between
these, he goes back to the examples of doors, and all I can think of is, “more
doors?!?!”. However, it was a good example for each one of these constraints.
He continues on to another example of switches, which we have all had issues
with one time or another. I thought his idea to have them horizontal was
brilliant and should tried to be implemented more. When these examples focused
on constraints and mapping, he wants to cover visibility and feedback through an
example over airplane controls. He also touches on using sound for visibility,
but concludes that there needs to be more research done to make sure it isn’t
annoying or invasive. The constraints topic was interesting to note. It
basically put common sense into words and categories. Other than that, I
thought he just went into more details that the previous chapters touched on.
Reaction to Ch. 5: To Err is Human
This lovely chapter talked about
all the lovely errors that us humans can make. This actually created a lot of
humor for me at the beginning when Dr. Norman was writing about slips. The
different types include capture, description, data-driven, associative
activation, loss-of-activation, and mode errors. I actually made a description
error when I went to leave my apartment. I had just stopped reading the book
and was thinking about it when I put on one maroon flip flop, and then walked
over to the other side of the room and put on my brown flip flop. I almost left
my room before I realized what I did.
Reading the examples that he gave of each of these types of errors made
me laugh and smile because I could relate to them. It was nice to know that
other people made the exact same silly errors. Not only were the examples
amusing, but they also were very efficient in explaining the different types of
errors. Dr. Norman then begins to explain why our brain works like this and how
we structure tasks. The topic of conscious and subconscious behavior was also
included in this. With all the errors that are made, many are explained away,
which could lead to disaster. I’m glad that he also incorporated social
pressure as a factor in explaining away error. I agree that it is an important
factor – one that we cannot truly understand since we can not remove ourselves
from our present culture. With all this information, the import concepts to
take away is that designers should deal with errors by making them easier to
discover and to make correction possible. One way of designing this system is
through forcing functions, which is a form of a physical constraint that can be
divided into interlock, lockin, and lockout forces.
Reaction to Ch. 6: The Design Challenge
In this chapter, Dr. Norman focuses on how a design
evolves over time. There are many different factors that work against normal
evolutionary design. These factors can consist of complexity, a competitive
market, time constraints, and the curse of individuality. I thought it was
interesting that there is actually a faster and more efficient keyboard layout
that has been produced. It is sad that this better model will never actually
succeed in today’s world because the current keyboard is “good enough”. When
designing these types of products, it is good for the designer to consider
cost, ease of manufacture, attractiveness, functionality, and durability. The
problem occurs when one of these factors dominates all the others. This can
happen because the designer does not understand the perspective of the user. I
liked how Dr. Norman put it this way: designers are experts with the device,
and users are experts with the task. (156) In order to design for the user,
every type of person has to be taken into account. This is possible through
flexibility. Two aspects that designers have to watch out for are increased
features, and worshiping complexity. What made me really sad was when he said
that computers are an area “where all the major difficulties of design can be
found in profusion” (177). Thankfully computers and programmers have come a
long way since he first wrote this book. We can continue to disprove his
statement by taking what we learn from this class and applying it to our future
jobs.
Reaction
to Ch. 7: User-Centered Design
This last chapter seemed to just be a conclusion of
what the book had previously talked about. It tied all the main topics together
and referenced many of the previously mentioned examples. It touched on
knowledge in the world and knowledge in the head, design model and the user’s
model and the system image, manuals, mental aids, simplifying tasks, bridging
the gulfs of execution and evaluation, visibility, mapping, constraints,
errors, and standardization. One new example that he used in this chapter was
when he began to describe the “home of the future” (213). Probably the most
amusing part was when he was talking about the information world of the future.
He couldn’t understand how we would be able to sift through so much data, but look
at Google – simple, easy to use, and yet contains billions of documents worth
of information. I am taking a class right now that is evaluating how the Google
machine works. It is really neat to understand how the complex background
creates an easy to use interface for the users.
Reaction
to the Book: The Design of Everyday Things
Dr. Norman’s writing style was very relaxed and
conversational making this book a relatively easy read. He uses the third
person and many rhetorical questions to engage the reader to make them think.
He explains the points of good design in a concise and clear way by using great
examples and tying in each previous design factor with the new one that he is
trying to explain.
The concepts provided
in this book were very beneficial to learn. They will help me to create better
products in the future and to evaluate the design of different types of
products. Not only does Dr. Norman bring many good ideas to light, but he also
supports each one of his claims with a couple of real world examples. Because
of this, I will never look at a door the same way ever again. In the past, I
could recognize design errors with some doors and could propose a solution, but
I did not have a complete understanding of what I was witnessing. For example,
I could not recognize that the blender was a bad design – I just assumed I
could not use it correctly. Now with my better understanding of design, I can
apply it to my everyday life and will hopefully be able to make life easier for
myself and my surrounding friends. One other topic out of many that I learned about
was how our memory works. These concepts will help me to study better in the
future and help me to recognize if information is actually located in my short
term memory or long term memory. This obviously also applies to design since
you cannot expect the user to memorize the whole manual. There has to be
knowledge in the outside world. This can be composed of lights, labels, or good
mapping. Mapping is another concept that I had never thought of before. This
natural mapping in combination with visibility and feedback make for a basic
good design. Then constraints can be added along with flexibility and error
control.
Throughout the book, it
is amusing to read about the various examples that he uses because most of them
are completely outdated. He referenced VCRs, tape projectors, type writers, and
really old home phones. On the flip side, he presents multiple ideas for future
technology, which has already been created like an iPhone or the search engine,
Google. It is interesting to see how some complaints he made about a certain
design have now been fixed. This makes me wonder if the improvement was caused
by his book or just the normal evolution of certain products.
Overall, I thoroughly
enjoyed reading this book. At first I was really sad and confused that we had
to read a book in a computer science class. However, it was a lot different
than I was expecting it to be. It actually served a purpose unlike some English
books that I have had to read in the past. I learned from it, it kept my
attention, and hopefully I will be able to use this knowledge in the future.
Everyday
Examples of Good Design:
Car
Light:
This handy dandy car light is helpful as you’re
driving down the street at night and need some light. It is easy enough to use
that it can be turned on/off without looking plus, an error really can’t be
made while using it because it is so simple. Let’s look at a more detailed
explanation of why this light is such a good design.
Mapping
and Visibility:
First of all, there is one action mapped to one
possible result. You want to turn it on, you push it. You want to turn it off,
you push it. Simple. How do you know how to push it? Well, there are slight
indentations on the plastic which gives an affordance of pushing. These can
also be discovered through touch if the user is unable to look for the light.
It is also not too far back to make it uncomfortable to reach and provides
visibility through the user’s peripheral vision.
Constraints:
There are obviously physical/logical constraints in
regards to this design. It cannot be pulled, twisted, or have anything inserted
in it. The design really only gives the user the option of pushing the button.
It also is understood that if you push the left button, the left light will
come on and if you push the right button the right light will come on. This is
the beauty of constraints and simplicity.
Water
Fountain:
Mapping
and Visibility:
For this case, the fountain strategically places the
“on button” near the same spot that a human would grab onto the device. This
visibility is added by only having one object that the user can interact with. However, I
would argue that this would not even be needed because the way that the “button”
is angled affords pushing it. This action is directly mapped with having water
come out of the fountain.
Constraints:
This device has physical and social constraints.
First, there is no way to obtain a grasp that would allow the user to pull the
lever. Also, most of us have grown up with water fountains in school since we
were all really little and thus have been using them and seen other people
using them for our whole life. This standardization helps immensely.
Call
Elevator:
Mapping:
This specific design has one button for one action.
If you want the elevator to come down to you so you can go up, press this
button. This avoids any complexities that could occur. In some instances, you
have an option of pushing a lower button or a higher button. In this case, the
lower button is mapped to the desire to go down, and the higher button is
mapped to the desire to go up. It is still simple and clean.
Visibility
and feedback:
This design also is very visible because it is
placed about the same height as you hand – the object of which you would
perform the action. In addition to this, once you hit the button a light
appears behind the button to let the user know that their action has been received
so that people won’t repeatedly push the button. In addition to this great
feedback, there is a display that has what floor the elevator is currently on.
From this users can see which way the elevator is moving and can tell that
their request is being executed. This is a great confirming factor.
Constraints:
Some constrains that help are physical constraints
since only one action seems to be possible. This is displayed by the affordance
of the button to be pushed.
Help to Cross:
This device helps us with our day to day life and has been continuously modified to perform better and better by making it easier for the user.
Mapping:
In this specific example, like the previous ones, there is only one action for the user to perform and that action is mapped directly to one result. It is also designed in a way that the indention affords placing your finger there.
Visibility and Feedback:
The product is great at giving feedback since it displays a light and makes a loud beep when it registers that the user pressed the button.
Constraints:
There are obviously physical constraints since it does not allow any other options.
Hole Puncher:
Mapping:
For this case, the handle creates a natural mapping because when you pull the handle down, you assume that the blade is also moving down to cut the paper. It is also an easy movement to duplicate. There is also a tray that holds all the unwanted pieces of paper. This is designed in a way that it affords pulling it out. This is the case because there is a lip in which you can grab onto.
Visibility:
With the clear plastic around it, the user can see what is actually going on inside the machine. This is really helpful to understand the model better and to confirm that it actually is working in the way that the user intended it.
Everyday
Examples of Bad Design:
Washing
Machine and Dryer:
Without fail, my roommates and I always seem to operate this device incorrectly
because the dryer is started by pushing the knob whereas the washer is
started by pulling the knob out. Thankfully the other functions work better.
Mapping and Visibility:
In this instance, there is no natural mapping as to whether you need to push
or pull the knob. Therefore, you either have to guess or search for the tiny
label in the bottom right hand corner. A better mapping would just be an on/off
switch. However, if this isn't possible, the designer should at least be able
to make the start command map to the same action - either both pull, or both
push to start.
Feedback and Error Correction:
There is no feedback if it is pulled/pushed the wrong direction, but
thankfully, the system gives immediate feedback when it does start. It also has
error correction since you can “redo” the action that you performed by
executing the opposite actions.
Overall, this system works well, but when you specifically look at the
mechanism to start the washer and dryer bad design is encountered. This could
definitely be improved to reduce the level of frustration that users have.
Blender:
Mapping:
This blender can be made to work, but it is not the
easiest device to figure out. First of all, it has two actions mapped to each
single button except for the crush ice button on the bottom. This creates a lot
of confusion because you don’t know how to pick the desired action. For
example, liquefy and chop are mapped to the same button. Now, there is a button
to the right that says high and low. If you push it down and push the
liquefy/chop button as well, I believe the blender is now set to liquefy. The
other option is that the high/low button is not pressed but the liquefy/chop
button is. This should mean that the blender is set to chop. This setup could
not be discovered without the manual. What if a non-pressed high/low button
actually means liquefy? – that won’t be good for your recipe. Also, I have no
clue how to stop the blender from blending. I think you can just push any
button half way and then it stops, but that is not clear at all. Overall, the
mapping would have been a lot better if the designers added more buttons.
Feedback:
The errors created by the mapping above could have
been avoided if the blender provided some sort of feedback. A light above liquefy
or above chop would have been sufficient to display what speed the blender was
running on. However, as it was, there was no notification to tell the user
which speed was active; thus, making them guess by the different pitch that the
blender made.
Constraints:
One other bad design feature is that there are no
forcing functions. Many times in the manual, there was a warning that told the
user not to hold down the “crush ice” button for too long because it might fry
the motor. They instead told the user to periodically hold down the button in a
sort of pulse pattern. What happens if the user pushed it and then started
talking to a friend and forgot to stop it? Does the blender recognize that a
particular mode has been on too long? Nope. It will continue to run and fry
itself out. What would be really beneficial in this case would be to add a
lockout device to stop the blender after a certain time or have the pulse
function built into the “crush ice” method.
Fan:
Mapping
and Memory:
You have three settings for one cord, which can be
confusing. Does it go from high, medium, low or low, medium, high? Maybe there
are only two settings. The point is you can’t tell until you experiment with
it. This is not necessarily the best case scenario. I think a better design
would be to replace the wall switch (which just turns the fan on) with some
sort of dial or range that can map to the speed or setting of the fan. If the
user wants to turn the light off while the fan is still on, they have to guess
which one of the two cords map to that. Normally I always pick the wrong one
and then have to reset the setting before I turn the lights off. Most fans try
to differentiate the two by having different length cords, but this fan has
them the exact same length. So, you cannot tell them apart except for a
different style handle on the cords. Thus you have to contain all the
information in your head. A better design would be to portray the information
in the world. This can be accomplished by
having one shaped like a light bulb and the other one like a fan blade. I know
this would definitely help me to keep them separate.
Earrings:
User Model vs. System Image:
This is not a common device that I would think of, but it made me realize that I had a different conceptual model compared to the system image. For some reason, I look at these clip on earrings and see that the one on the left has a bigger gap between the front and back. Therefore, I assumed that this means that it is loser and would not hurt as much. However, I also noticed that when you turn the screw clockwise, which is normally mapped to tightening something, the earring looked like the one on the left. So, was my first idea right or the second one? In actuality, the system model is more like the second idea. Even though the front and back are farther apart, there is a spring that causes more force as the back is forced away.
Emergency Help:
For this object, I wouldn't think that it would be considered bad design. However, after I really looked at it, I noticed some changes that could be made to drastically improve it.
Mapping:
First of all, the mapping is not all one for one. In order to start a call you would have to push 1 first. This fact is only known through reading the tiny read sentence on the top right. If I was in danger and needed to call someone, I don't think I would be calm enough to read and comprehend why the phone wasn't working.
Feedback:
However, there is good feedback with the light in the bottom left hand corner and of course the big blue on one the very top. However, I don't know if any of the buttons make a noise or if you could accurately read everything on it if it is dark outside. Therefore, lights behind the buttons would be very beneficial. It would also be good if these lights changed color once the button was pressed.
Usability:
One other factor that I thought of was to have larger size buttons to account for scared individuals who are apt to make a mistake.
His
thought experiment was set up to where a person was locked in a room with only
a filing cabinet, a lot of paper, pencils, and erasers. This individual would
be given a manual in English, which is a duplicate of the computer program. The
human, who only knows English, is given Chinese characters through a slit in a
door. Using the computer program, it is assumed that the human can then reply
to the input with his own Chinese sentence; thus, being able to hold a
conversation in Chinese. However, the human is only following directions and
does not actually understand any Chinese. This process simulates what is going
on in the computer, which emphasizes the point that computers don’t actually
have a mind, but that they are only simulating one. This defines “weak AI”
whereas actually having a mind would be a “strong AI”. 
