Sound-Based Interaction

From CS160 Spring 2014
Jump to: navigation, search

Slides

Videos and Papers Covered in Slides

  • Voice as Sound. Igarashi and Hughes. (webpage)
  • VoiceDraw. Harada et al. (webpage)
  • Scratch Input. Harrison and Hudson. (webpage)
  • Touch and Activate. Ono et al.(webpage)
  • Acoustic Barcodes. Harrison et al. (webpage)

Readings

Reading Responses

Jamie Cai - 1/27/2014 14:31:44

An example of a physical device with bad design is a lamp, and not just any lamp, but lamps in general. Just as Norman speaks about doors not having uniform signifiers of how to open them, neither do lamps. Some you turn a switch attached on the wire, some you turn a round switch near the bulb, others you have to press that same switch instead of turning it, sometimes you press a switch on the base of the lamp, other times have that switch attached to the wire, etc. A lamp affords light, therefore generally you know that there must be some way to turn that light on, but the location and operation of that switch may be difficult to figure out. This violates Norman's design principles of having appropriate signifiers and having one general conceptual model. I would re-design this problem by working with others to set a standard of how to switch lamps on. I believe lamps should work like ceiling lights that have a switch on the wall - have one general area where the switch is located that is easy and intuitive to find (near the entrance to a room in the ceiling light's case), and have one uniform way of operating the switch (flicking it up/down to turn on/off ceiling lights). Similarly, lamps should have a switch near their base to flick up and down to turn them on and off. Some lamps are already adopting this concept, but a lot still have this form of switch: lamp-socket.jpg

The difference between an affordance and a signifier is that an affordance is what something does, and a signifier shows how to do what that something does. They differ for physical products and software UI because each one requires a different set of skills to operate. Physical products require a more diverse amount of physical movements and motor skills to operate, and thus they have different signifiers that need to cater to body movements. For example, handles on something generally mean pull, circular knobs mean turn, etc. Additionally, the affordances/signifiers of physical products exist in the physical world whereas those of software usually do not. Most software affordances are on the cloud, such as Google affording searching, Instagram affording picture sharing, etc. Therefore most software signifiers must be symbolic of computer functions and the cloud, such as Google's search bar and button to symbolize searching, Instagram's camera button to signify uploading photos, etc.


Gregory Quan - 1/27/2014 22:12:24

The multi-light switches in my house are designed badly because they do not map accurately to the lights that they control, and they also provide no feedback about whether some of the lights they control are on or off.

For example, in the kitchen there is a panel with two switches. The one closest to the garage controls the garage light. In the entry area, there is also a panel with two switches. However, the switch closest to the porch does not control the porch light. Furthermore, in both of these cases, it is difficult to tell whether the garage or porch light is on from inside the house, so people will often unknowingly leave these lights on. In the case of the panel in the kitchen, they will often hit both switches at the same time in order to turn on the kitchen light, not knowing they have also turned on the garage light. Complicating everything is the fact that the "on" position can change because there are multiple switches that can toggle the same kitchen lights, so the "on" position for the kitchen light is not necessarily the same as the "on" position for the garage light.

A simple way to solve this problem would be to arrange the switches so that the location of the switches maps to the location of the lights they control. For example, the porch switch should be the one closest to the porch. Also, an on/off label for the garage and porch lights would be helpful. Alternatively, changing the design of the switch itself so that the position of the switch signifies the state of the lights would also work.

Affordances are the possible interactions between a device and its user. Signifiers signal things, in particular what actions are possible and how they should be done. For physical products, affordances and signifiers are usually manifested in three dimensions, while software user interfaces are usually restricted to a two dimensional screen. Physical products can use different types of tactile feedback as signifiers, but software interfaces are usually limited to vibrating. Also, physical products can sometimes have affordances that were not intended by the designer, such as using binder clips to keep potato chip bags closed. Software interfaces do not usually have as many unintended uses, again because they are usually restricted to two dimensional screens.


Jeffrey DeFond - 1/28/2014 11:25:27

1) When I lived in the Units, I had a rigid plastic tray/carrier with a handle in the center. It came from Ikea and was designed to carry ones toiletries. I hated it and never used it for a few reasons. It had no way for water to vent out of, so after a shower I would have to to take everything back out of it, turn it over to drain the water, then replace all my items. A draining affordance, no signifier required, would have been nice, if it had had vents, simply picking it up by its handle could have made this affordance possible and easy. It was also not well sized for bulk shampoo and conditioner (I guess that might be on me, but still). I wound up just carrying my toiletries to the shower by hand because there were less things to fiddle with, solving the paradox of technology with staunch luddite-ism. I would suggest to anyone creating a toiletry carrying device to style it after a fishnet. A large mesh bag for bulk shampoo, with smaller external mesh compartments for things like razors, toothbrushes and tooth paste.

2) An affordance is something that a product allows a particular user to do. It is inherently between a user and a device/interface. A signifier is something that alerts a user to an affordance, and must be perceived. An affordance can be inherently signified by design, for example the handle of a hammer does a fairly good job of both providing an affordance (a grasping spot for hammering) and signifying that the affordance is there (grasp here). I believe signifiers are more important in software, which can provide a huge number of affordances that unlike a physical device are not necessarily apparent from the appearance of the device. Good signifiers to ensure that users know what is going on to help them create a conceptual model of the software (while not drowning them in Clippy suggestions) are more important because of the amount of “behind the scenes” operations occurring in a piece of software.


Myra Haqqi - 1/28/2014 12:29:56

1) Give an example of a physical device (an "everyday thing" as Norman would call it) with bad design that you have had to use. Do not think about software! Think about household appliances, sports equipment, cars, public transportation, etc.) Which of Norman's design principles did this device violate? How would you re-design it to solve the problem?

An example of a physical device with bad design that I have had to use is a mechanical pencil with a cap over the eraser. The cap that covers the eraser causes a lot of irritation when I have to remove the cap every single time that I want to erase something, and then I have to put the cap back on the eraser when I am done so that I do not have to keep holding it as I write. This is very poor design because it is annoying to have to remove and replace the cap. It is also inconvenient to have to hold the cap as I use the eraser of the mechanical pencil to erase something.

Another issue that the cap over the eraser of the mechanical pencil causes is that whenever I put the cap back on to cover the eraser of the mechanical pencil, I have to push the cap hard to ensure that it stays in place and is properly closed. However, this causes the mechanical pencil’s lead to come out, as the mechanical pencil is designed to extract more lead by clicking the eraser. This gets frustrating when I do not want the lead to enlarge. I am then forced to hold down the eraser and push the lead back in to get it back to the desired length. This design is extremely inefficient and bothersome.

One of the design principles described by Norman that the lid over the pencil eraser violates is that it does not even offer a visible affordance. By looking at the cap, I as a user do not see the use or value in the lid of the eraser. The lid does not possess a proper signifier to communicate to the users what the purpose of the lid is.

To solve these problems, I would redesign the mechanical pencil to not even include a cap to cover the eraser. I do not find much merit or utility in a lid over the eraser of a pencil, and I therefore think that all mechanical pencils should simply never include a lid over the eraser.

Another example of a physical device with bad design that I have had to use is a food blender which contains a button to “puree” and a button to “blend.” This has always troubled and confused me when I am attempting to make myself a smoothie or follow a recipe when cooking, because I am not sure what the difference between “puree” and “blend” are. I believe that the average person would not necessarily understand the distinction and could therefore possibly accidentally choose the wrong option when trying to mash together ingredients.

The design principle of Norman that these two confusing buttons violate is understandability. In particular, it is difficult to discern between the meanings of the words “puree” and “blend.” Users will not know what the words mean, and will therefore not understand how the settings on the blender work.

To overcome this issue, I would redesign the food blender to use more specific descriptions in order to provide succor for users to understand what the settings mean so that they can perform the appropriate and necessary feature with their ingredients. “Puree” means that the food blender mixes together the ingredients until they become very smooth, so I would label it to include more informative words such as “grind” or “mash” “until thick and thoroughly smooth” in order to allow people to understand what exactly pureeing their food will do. On the other hand, blending simply mixes ingredients until smooth, so I would use better words to describe this setting such as “mix until smooth.”



2) Give a concise definition of the difference between an "affordance" and a "signifier". How do affordances and signifiers differ for physical products and software user interfaces?

A concise definition of the difference between an “affordance” and a “signifier” is that an affordance is a relationship that discerns the actions that are possible and also symbolizes ways in which interactions between agents and an environment transpire, while a signifier is the signal that communicates where the action should take place, what actions are possible, and indicates the appropriate ways to interact and make the appropriate actions.

A more in-depth analysis of the differences between affordances and signifiers involves the relationship between affordances and signifiers. While affordances are a relationship rather than an actual property, signifiers are what signify the affordances. Also, affordances may be either perceivable or not perceivable. However, signifiers are always perceivable; otherwise, they would fail to fulfill their purpose, which is to communicate to users about how they can accomplish their actions in the proper manner.

For physical products, affordances are important in that physical objects possess details that enable users to gain an understanding of its uses and features. They contain information that allows people to understand how to behave and perform the appropriate actions to them. Signifiers are actual physical signals on or around physical objects that provide a signal to users as to how to go about interacting with the object. For example, a signifier for a physical object such as a door might be a physical handle that allows people to see that they are meant to open the door and where specifically to perform the action of opening the door. Signifiers for physical objects are often physical cues that give users the opportunity to understand how exactly they are meant to execute tasks in order to successfully utilize the object.

On the other hand, for software user interfaces, affordances are features of the interface that allow the user the perform some actions. Signifiers for these affordances are often less prominent because space is a particularly precious resource for software user interfaces given a fixed screen size. Affordances for software user interfaces are still relationships that portray the possible features, functions, uses, and applications of some software user interface; however, when contrasted with affordances for physical objects, affordances for software user interfaces are not physical in nature. While physical objects convey a lot of information for people about its use, software user interfaces do not necessarily demonstrate the same information. Signifiers for physical objects can be bold and obvious, such as signs. But for software user interfaces, signifiers are often more subtle, but just noticeable enough to allow a user to see that they are meant to use some feature, perhaps with an indicator such as an arrow that shows that one may swipe the touch screen of a phone to view a different screen or the use of images and icons.


Daniel Haas - 1/28/2014 13:08:03

1) The cabinets in my bathroom are an example of a physical device with bad design. The problem is as follows: every cabinet looks the same (a flat panel with a knob on it), and all are the same size. However, some panels correspond to drawers, others to cabinets with shelving, and still others are dummy drawers for decoration that cannot be opened at all! In Norman's terms, the signifiers are very misleading, there is a false mapping from the door knob to the expected storage space behind it, and it is therefore impossible (without memorizing the layout of the drawers/cabinets/fakes) to build a conceptual model of the unit's operation. To fix it, I would build different handles for the different types of storage space and remove the handles on the dummy panels entirely.

2) An affordance is an action that can be carried out on an object by a human given both the properties of the object and the abilities of the human: e.g., a chair affords lifting. A signifier, on the other hand, is a signal that indicates where and how an action can take place on an object: e.g., a "push" sign on a door. Software user interfaces, unlike physical products, tend to be affordance-dense: that is, they provide many capabilities in a small amount of space. This means there is less space for signifiers, and it becomes much more difficult to provide a system image that correctly communicates the interface's conceptual model.


Anju Thomas - 1/28/2014 14:16:08

1) A physical device that would fit Norman's definition of bad design would be the stove counter top in my home, which has four controls for the four burners. Though the burners are located in a square model, one on the top left, one on the top right, the other on the bottom left and another on the bottom right, the controls are arranged in a vertical line in the side. This causes me to forget which of the two top controls belongs to the left top burner and which one belongs to the right top burner and the same with the bottom two controls. It violates the need of visual signifiers and mappings, without which the user fails to map the controls with the corresponding burners. I would have redesigned the model to make it more easy to remember by arranging the controls instead in the same square design as the burners, with the first two controls side by side and the other two controls beneath them also side by side. This would also fit the mental model that many users have.

Another example of a physical device with bad design includes the car brake and accelerator pedal. Since both the pedals look similar and close together, the user is forced to remember which one is used to brake and which one to accelerate unless they try it, which can be dangerous in some situations. For instance, the user might forget and might step on the accelerator instead of the break during a red light, possibly causing it to crash with the vehicle ahead. The design also violates the need for signifiers and mapping, forcing the user to remember the controls. I would redesign this by having a signifier on the milage screen that shows a lighted text “Brake” on the left and “Accelerator” on the right to help the users map the side of the signifiers with the position of the pedals, before doing trial and error.

2) An affordance is a relationship that conveys the different possibilities of interaction with an object. They hint at the way a user can use an object without any labels, signs or drawings. The design communicates naturally with the user, rather than logically. A signifier is a visible cue that defines how an object should be used or the possible actions. It relies more on visible marks to inform the user of a specific mode of interaction, without which the user might become confused. Perceived affordances are usually signfiers, though they can be misconceived. In physical products affordance is the relationship between a person and how the possible ways they could interact with an object. For instance the affordances of a plate include lifting, eating from it, washing it, etc.. Signifiers for physical products can be percieved affordances such as the handle of a door or it can be labels, signs or drawings on a physical object that indicate to the user how to interact with the object. In software interfaces, affordances and signifiers play similar roles. For instance, an affordance in a software user interface would be the possible interactions conveyed of a human with the desktop, as saving files/ shortcuts and setting background images. A signifier would usually indicate the signs or drawing visible to a user in the screen that indicate the action that they want to do. For instance, some signifiers include the icons on the desktop, that tells the users its tasks or the progress bar of a download that indicates the amount of downloading finished, like the physical bookmark. However, possible interactions with user interfaces are usually based on previous experiences than natural experience. Affordances are more limited in user interfaces than in real life. For instance food in real life can be eaten, washed, etc.. however images of food on a computer screen disable the user from affordances that they would have in real life and limit them to viewing. Also at times signifiers are simpler than those for physical objects, like pushing a button to send mail rather than follow all the rules post a mail physically. Though recently software user interfaces provide more affordances by enabling interactions with the objects on screen simply by touching – more natural than using a mouse.


Ziran Shang - 1/28/2014 14:44:55

A physical device that has poor design are the gear shifters on some bikes I have ridden. On many such designs there is a lack of signifiers that show the user which motion shifts to an easier or harder gear. You only figure it out by trial and error, and practice. To an experienced rider this is not as much of a problem, but it could be especially confusing to a new rider because applying the same action to the front and rear shifters causes opposite effects (that is, a harder gear on the front and an easier one on the rear, or vice versa). I would redesign the system to be more user friendly by having two shift levers, one above the other, on each side. The top would shift to a harder gear, and the bottom to an easier one. This design utilizes natural mapping.

The difference between an affordance and a signifier is that an affordance is simply the possible actions on an object, while a signifier is an indication to the user of what actions can be done and how to do them. A difference between physical products and software user interfaces is that in software affordances and signifiers are mostly limited to what is programmed. On the other hand, physical products have a greater chance of anti-affordances, and people can make their own signifiers, such as in the staircase example that Norman mentions.


Emily Sheng - 1/28/2014 16:11:58

1) Some of the faucets I've encountered have taken me a while to figure out how to turn them on. For the faucets with knobs, I usually assume I have to pull up, push down, or turn to a side to work. Sometimes the shape of the knob informs me which action I should try (ex: a long, horizontal, bar-like knob tells me I should try pulling upwards), but sometimes the knobs do have a horizontal, bar-like shape and work by being pulled forwards instead. This sort of bad design violates the principle of signifier visibility and conceptual model. A horizontal, bar-like knob has the affordance of being able to be pulled upwards (at least it does in the conceptual model created from my experiences), but if it really should be pulled forwards, there should be a signifier saying so. I would redesign this type of faucet by adding an arrow to the knob pointing in the direction to be pulled, along with adding the words "pull."

2) Whereas affordances are the relationships between object properties and abilities of people to use the object, signifiers are indicators of information that tell people how to interact with the objects. Affordances could be perceivable or not, but signifiers need to be perceivable to communicate with people how to use objects. Affordances and signifiers could differ for physical products and software user interfaces in that in physical products, affordances and signifiers can often be the same thing (ex: a door knob). This may be partly because physical products are often less constrained by dimensional limits and can utilize their shapes as both affordances and signifiers. Software user interfaces are much more limited in terms of dimensions. They are usually on a screen, so although they can be given depth and shape through lines and shadows, the interaction is still done through a flat surface. In these interfaces, we often have to be more explicit with the signifiers, because the affordances (ex: swipe up) may not be as clear.


Maya Rosecrance - 1/28/2014 16:34:52

Our hallway at home has two light switches on either end of the hallway. In order to turn the light on, both the light switches must be turned on. Since you occasionally have to navigate the length of the hallway in the dark to turn the light on, and it is very difficult to figure this out the first time someone interacts with it, it violates the principle of discoverability, specifically because the mapping is confusing and there is no obvious feedback about what is going on. A possible solution would be as long as the switches differ (eg one is off and one is on) the light will stay on, so that a user can change the light state from either end of the hall. It is still confusing since you might have to flick the switch “off” in order to turn the light on but it is more usable. A better solution would be switchs that synch so if at either end the switch is turned “on”, they both are moved to the “on” position.

   Affordance is all the possible/practical interactions between the object and the user. Signifiers communicate how to do those interactions or indicate that an interaction is possible. For physical products affordances can be almost anything and signifiers do their best to communicate whatever the designer wishes the user to know about. For example, most shopping carts have a concept model indicating how to safely have a child interact with the device, a model that is probably not used anywhere else. You can use the shopping cart as a skateboard, a homeless cart, a food basket, a child holder, etc. However for software user interfaces, the affordances are more limited as the designer has only the screen and interactions read using the camera, microphone or moving the device itself to create affordances. The signifiers are typically icons that the user is familiar with, a microphone, arrows for swipe, a menu icon, etc as the possible interactions are limited.



Michelle Nguyen - 1/28/2014 17:14:18

1) My friend and I rode the Muni metro light rail in San Francisco for the first time this December. As a nonlocal of San Francisco, I found my experience as a first time user revealed a major design flaw. The first problem occurred at the faregate--I had thought it would be similar to BART's, but the faregate had both a mechanism to slide your card and a light sensor and no instructions. I had to try both the slide and the sensor until I figured out the correct one to use. When we boarded the Muni, we realized that we did not know how to use our ticket to pay on our way back from an above-ground station. The stickers on the Muni's door indicated that it was possible to use our tickets to do so in any of the cars--but where and how? We attempted to watch others who boarded at other stops, but none used a ticket as we had. The stops the driver announced were also difficult to hear over the noise for the train, and as nonlocals, it was hard to determine where we were to decide when to get off. This was especially difficult since the Muni halted at unofficial stops, so we could not just count the number of stops we had left. Although we had paid for a roundtrip ticket, we opted instead to use the bus on the way back to avoid the light rail's confusing operations. The design of the light rail was catered to locals and violated Norman's principle of signifiers, despite the fact that San Francisco is a major tourist city. There were many signifiers, but few were helpful to those who needed it. I would first redesign the faregate by having a sign to indicate where and how to use the ticket. Better yet, since the sliding mechanism of the faregate was not actually used for tickets or a clipper card, it should be removed to avoid confusion. There also needs to be instructions on each bus near the entrance of the door (maybe near where the clipper card is scanned) to explain how to actually use your ticket. To address the problem of determining what station we were at, I would put big signs with the station's name at every major station, similar to how BART does.

2) An affordance is the relationship between the object and interacting agent that determines what actions are possible, while a signifier shows how or where those actions can be done. For physical products, the affordances are much more obvious than in software user interfaces. For instance, Norman uses the example of the scissors. He states that it is easy to see that the holes are for your fingers, and the implications of its uses are clear (we can see the blades that open and close). From the structures of physical products and experience in our daily lives, we can usually determine the affordances of that object through just the properties the object has in order to perform its functionalities. Thus, explicit signifiers such as labels are often unnecessary and signifiers can just be perceivable affordances. For software user interfaces, it is not so easy, since these interfaces are all just displayed on a screen. Without any labels, the affordances can be unperceivable. What we see before us on the screen has no relation to how our object is functioning, and therefore we MUST include signifiers in the interface. This is done through labels or by mimicking a physical object we know how to use in real life.


Allison Leong - 1/28/2014 17:57:06

1) The shower control in my parents' bathroom was designed to look sleek and artsy, but I feel confused every time I use it. It is a skinny metal rod that is mounted to the wall and points vertically in the off position. To turn the water on, the rod/handle is rotated clockwise. There is a small “h” symbol to the upper right of the handle where 10 o’clock might be on a clock. There is a small “c” symbol to the lower left of the handle where 8 o’clock might be on a clock. These symbols imply that to get hot water, the handle should point upwards and for cold water, the handle should point downwards. However, it is unclear whether it is the base of the handle or the “top” of the handle that should point up/down. Since the water temperature is slow to react to adjustments made to the shower control, I often let the water run for several minutes before realizing that the handle is in the wrong position. The shower control violates the principle of good mapping between the position of the handle, and the change that should take place. The user has no sense of a natural mapping between hot temperature and positioning the rod at one angle vs. cold and positioning the rod at another angle. To redesign this particular shower control, I would place an arrow pointing out of the base of the rod (this would be an arrow pointing down when the shower is in the off position) because it turns out that it is the base of the handle that must point at the little “h” to get hot water and “c” to get cold water.

2) An affordance for a product is any potential interaction that can take place between a subject and the product itself. A signifier signals an affordance to the subject. In certain cases, affordances are themselves signifiers. In physical products, the physical characteristics of the product often serve as signifiers. For example, the shape of a chair is often enough to signal to a person where he/she should sit. However, with software user interfaces, the interface often exists within a rectangular screen. Signifiers in a software user interface must be explicitly designed because the screen itself will not give the user any indication of how the software should be used.


Sang Ho Lee - 1/28/2014 18:24:13

1.I have had the pleasure of using a particularly bad car interface, of which I don't remember the exact brand, except that it was a mid-high range luxury vehicle. It was overloaded with buttons and controls, a prime example of Norman's paradox of technology. While there were controls for even minute details, things as simple as turning the AC on or changing the radio station was difficult because of the low discoverability caused by the overabundance of signifiers. In this case, the overload of signifiers, text and symbols each representing some control and tuning button actually made it dfficult to find the signifiers that I usually looked for in a car's control system. In addition, feedback was not plentiful enough for me to figure out instantly what the results of pressing each button were. While there were obviously many things to be afforded by this control system, my mental model for a simple audio or climate control system did not coincide with the car's control dashboard, and I could not adjust right away to its confusing complexities. If I were to redesign the control dashboard for that specific car, I would cut the number of control interfaces dramatically, or integrate a large, central touchscreen hub with essential control surfaces surrounding. The larger display would provide immediate feedback to the user. Sound feedback would be very useful as well. By reducing the number of buttons and knobs and leaving only the most essential and helpful signifiers, the user (even a new passenger) would be able to navigate and use the interface with ease.

2. An affordance is a relationship between the object and its user that allows or disallows the user from doing a certain action(s) as a result of using or being manipulated by the object in the environment. Such actions are often physical, but may also be emotional, mental, or environmental. A signifier is a perceivable object that signals things-- such as the possible interactions or affordances between an object and a user. For physical products, affordances and signifiers can rely on actual physical shape to communicate with the user while software user interfaces are often delivered through a fixed platform which is often a physical object. Physical products can also be subject to some senses that software user interfaces are not, such as the sense of touch (texture, temperature, etc) which offer immediate feedback and further insight for the user for the interactions afforded by the product. Software user interfaces can have more helpful and arbitrarily placed signifiers compared to actual physical objects. While adding a signifier to a physical object may disturb the overall design of the product, signifiers are often essential UI elements in software user interfaces that are actually to be expected to be present in software universally.


Haley Rowland - 1/28/2014 18:41:55

1) The window in my room has poor design: it has two locks- one at the side of the pane and one in the bottom corner of the pane. The lock at the side of the pane is not labeled, but contains a lever that affords flipping the lever in order to lock the window. This lock intuitively makes sense and is easy to use. However, the lock at the bottom corner is a rectangular block jutting off the side of the pane and is labeled “Push down and hold to release lock.” This second lock is unnecessary given the presence of the first lock and makes locking the window more complicated, thus adding difficulty and frustration where it shouldn’t be. Secondly, it is ambiguous what is meant on the label by “down”: should I press towards the window (since the block looks like a button that can be pressed in this way) or towards the ground? Furthermore, neither of these actions (pushing toward the window or ground) actually does what the label suggests: in order to open the window (what I assume to be “releasing the lock”), the block must be lifted toward the ceiling. This is an example of a poor signifier: the shape of the lock does not lend itself to any obvious action and the label further confuses the user in what action should be taken.


2) An affordance is a relationship between an object and a user that determines what actions are possible, while a signifier is an indicator that communicates to a user what actions are available and how to perform them. For physical products, affordances are determined by their physical properties- its weight, size, structure, etc. In a software user interface, affordances are determined by the user’s interaction with the technological device- the computer, phone, or tablet. In a software setting, the affordances are determined by a user’s interaction through mouse clicks, keyboard entries, gestures, or other properties of the device involved.

In physical products, signifiers may be more implicit in the object’s physical design, while in software user interfaces, signifiers may need to be more explicit to direct users in their interaction with the application. Because hardware imposes a restriction on the possible actions available in using the software (mouse clicks, keyboard entries, gestures, etc.), these same actions are used differently across many applications, and thus signifiers have a more prominent role. They must teach users which actions are possible and convey their effects and thus the signifiers have a more direct role in defining the affordances in a software setting.


Max Dougherty - 1/28/2014 19:00:32

7:55AM and still no bus. Ten minutes of waiting with no result and no indication of an imminent arrival. If it comes in the next 5 minutes, I’ll be moments away from a warm seat in class. If not, I’ll not only be late, but I will also have had to run to class just to be less late. So do I wait? Or do I leave now to avoid the frustrating exercise? In a time crunch, a few minutes can make the difference, and without feedback, any decision I make is a gamble. There are apps that predict and schedules that say, but the 51B has its own way. Without real time information, the Berkeley bus system fails Norman’s design principle of feedback. The user is left waiting, but has no indication of the next arrival. The simple solutions is already used in many cities. A live indication of a bus’ distance and predicted arrival time would provide the prospective rider with the information required to make an informed decision.

Donald Norman uses affordance to refer to the “relationship between physical object and a person”, this extends to the user’s own ability to understand how an object can be used or manipulated. A signifier also extends the user’s ability to understand how an object can be used or manipulated, but requires explicit identification by the designer through the signifier. Hence the difference between an affordance and a signifier. Affordances are innate to the object, often incurred through previous experience with similar objects or systems. But when a function of an object is not inherently clear to a user, a signifier may be required to illustrate the possible interaction and where it can occur. It may be easy to say that software requires a greater number of signifiers and maintains fewer affordances, but this could easily be a result of continuous radical change in UI design over the last decade, which have made recognizing or “affording” a function of program difficult. With uniform design principles as in the “swipe left to delete” function in iOS, our expectations are satisfied and we afford the ability even in programs we have never used before. If software UI design converges, the number of signifiers required will decline as certain functionality becomes expected.


Nicholas Dueber - 1/28/2014 19:15:35

1: Mechanical pencils have bad design. The process of reloading lead and changing out erasers on pencils is entirely too difficult. When the eraser gets worn down, the difficulty of the task is exponentiated. This violates Norman’s design principle of visibility. If I had not seen another person take of the eraser to insert more lead. I would be at a loss for how to do it. Taking the eraser off of the pencil is not intuitive. Indeed there should be another way. I would propose that a method to fix this is that there should be a sliding door on the pencil. At the top of the pencil you could insert the lead through the sliding door. If you imagine the design of a rifle, to unload the casing and reload the rifle, you open the door that has a sliding door. I think a similar design could be afforded for pencils. This would solve the problem of not knowing where to insert the lead. It would be intuitive. You would see the door and know that the only thing that goes in pencils is lead. This also reduces the time required to take of the eraser. 2: Affordance is the relationship between the object and how the agent will use the object. Affordance can be thought of as what you perceive an objects purpose to be. Often times, it is visible to the human eye, the agent can understand what the purpose of the object is based on the objects properties or relationships. This differs from a signifier. An affordance shows what is possible with the object. A signifier more directly communicates to the agent what should be done. A signifier tells the agent where an action should take place. Affordances can be understood without context (i.e. a chair is for support and affords sitting), and a signifier signals to the user to do an action (i.e. labeling one side of a door “push” and the other “pull”). In software, you can think of an iPad as affording the user to touch the screen. This is how the user will interact with the machine. The signifiers on an ipad are the buttons programmed into the UI. Naturally the user will want to interact with the machine so there have to be signifiers on how they can do that.


Andrew Fang - 1/28/2014 19:18:45

One device which has frustrated me on numerous occasions is the horizontal blinds on windows where there is only one string used for opening and closing. When I pull the string, the blinds go up and they stay wherever the I stop pulling. However, it is not clear on how to close the blinds. Sometimes, the solution is to pull the string to the side and this releases some mechanism, which allows the blinds to go down. But this doesn’t always work, and there is no clear indication that this is the right way to close them. This device lacks signifiers; there is nothing that says how to property close the device. The mapping is odd, it is not natural for me to have to pull the string to the side if I want to close the blinds. The design lacks a feedback mechanism (such a simple click) to let me know that I am pulling it in the right direction and that the blinds are ready to be lowered. If I were to redesign this, I would replace the string with three buttons: a right side up triangle, a circle with the word “stop”, and a upside down triangle below that. The top button would be to engage the motors to raise the blinds, and the bottom one would be to lower the blinds. When the blinds have been fully raised or fully lowered, there would be a small click to let the user know. Pressing the up or down triangle would automatically raise or lower the blinds and by default they go all the way up or all the way down. If the user wants to put it only part-up, they would press the stop button to stop it where it is.

An affordance is a relationship between an interface and its user, while a signifier is a tool that demonstrates to a user one of the ways an interface can be used. Or, as Norman puts it, “Affordances determine what actions are possible. Signifiers communicate where the action should take place.” Often, physical products can have many more affordances than a software user interface. It is up to the user’s imagination to come up with different ways to use a physical product. For example, a chair affords sitting, it affords lifting, but it could also afford weighing down papers or afford stepping on to reach higher ground. The possibilities are potentially endless. On the other hand, a software user interface is coded to only allow so many affordances. It is up to the developer to imagine how the application can be used, and it will seldom be used for many other purposes other than what it was written to do. Signifiers for physical products usually are visibly displayed on the product as a constant reminder of proper usage. For a software user interface, however, signifiers can be toggled, perhaps even disabled. Often, signifiers can be used upon initial installation of a piece of software, and are pedagogical tools that allow the user to be familiarized with the product. Once the user is comfortable, these signifiers can be removed or hidden for a cleaner look.


Armando Mota - 1/28/2014 20:16:59

1) An object that immediately comes to mind when discussing a terribly designed physical device is the cable modem/router, specifically those used with tv/internet service in average homes. This product fails most spectacularly in Norman's "feedback" design principle, and anyone who's had one (and is perhaps not an IT or network person) will have dealt with this issue. Internet/television service is, for the most part, reliable, however when it does fail this device is extremely unhelpful in providing you with an explanation. There are commonly 1-5 different lights on the machine that indicate whether or not the machine is powered on, whether it's receiving signal, and various other metrics, however in a general view these aren't entirely helpful. I've had the cable company tell me I was getting full signal over the phone while directly looking at my unlit "signal" button. This might be by design, in an effort to get people to call the company's help line directly, however during extremely busy hours or after hours this is often not even an option, leaving you stranded with no idea of why your service is not working at all (bad weather?), is slow (the 1's and 0's started fighting eachother?), is patchy (James Brown has been inexplicably reincarnated as your router, and is doing what appears to be "the grit" on your data stream?), or is making pink unicorns appear on your screen (no conceivable explanation). In this case, it is not necessarily the modem's fault - the usage of modems is vast, and for technical companies that have support staff this is no issue. However, specifically for cable companies, the addition of a simple digital readout that displays error codes that are clearly explained on the product's side, or more specific lighting arrangement (in which, perhaps lights were very specific and could be directly referenced in phone conversations) would help alleviate the load on both average home user and tech support agent. The device itself might even have an "error check" button that, through some simple coded function, develops a specific error code that is sent to the company and causes one of their support reps to call you. This last feature is a feature that would be company specific, it borders on being a design upgrade to their support service along with the actual physical router, and would most likely have to be a collaboration between cable and router production companies, however it still applies.

2) A signifier is a direct cue that lets us know to use an object in a specific way, an affordance is a way in which is can possibly be used. In addition, an affordance is a relationship between user and object while a signifier it a direct cue.

  When thinking about the way in which affordances and signifiers differ for physical products and software user interfaces I believe a careful distinction is required.  In practice, software user interfaces must be situated in a physical medium (we must, after all, be able to interact with them through screens, etc.), so in this sense the affordances of physical products and software user interfaces are one in the same, they are both physical products.  A filing cabinet would afford the storage of data just as a hand-held data holding device would.  The signifiers of the physical object would be placed in 3-d space in your physical world, and the signifiers in a software-driven user interface would be situated on a screen or display (or command, if audio driven).  If you discount the physical factor of a user interface though, the affordances of a user interface essentially become the affordances of another human (that is to say you can interact with sophisticated enough software as you would with another human).  You could order your software you deliver you a pizza.  You could order it to complete a math problem.  You could possibly make it show you a picture, or a video, or tell you a funny story.  The possibilities are as varied as those that exist with an actual person.  Likewise, the signifiers for this concept of software user interface are varied.  You could have a pizza app that has a large "PRESS HERE TO ORDER" button in the center of the screen, you could have a small visual calculator that allows you to input data and it displays the answer in its screen.  In this sense, software user interfaces are like an aggregation of physical products, which explains their wide utility.


Christopher Echanique - 1/28/2014 21:34:37

One example of a poorly designed device is the cable machine at the RSF. This multipurpose machine is designed to be adjustable to a wide range of positions in order to accommodate for different types of strength training workouts. When I first tried to use it, I wanted to widen the arms of the machine to perform chest fly. I noticed there were holes with numbers at the base of each arm affording rotation. However it was not clear how to enable this rotation. A student who noticed my bewilderment informed me that the pedals below the machine were used to rotate the arms. This device clearly violated Norman’s design principle of signifiers since it lacked such a signifier to indicate this action. In addition it violated his mapping principle since the pedals were not located close enough to the arms to which they mapped. If I were to redesign this machine, I would create a knob near the base of the arm that can be used to adjust the arms. In addition I would color the knob yellow as most of the other machines do to signify an adjustable feature.

Affordances determine what actions are possible whereas signifiers communicate where the action should take place. In physical products, affordances could also be signifiers, such as a vertical plate on a door. In this case the plate affords pushing and also signifies to the user where to push the door. Software user interfaces on the other hand cannot add new affordances to the physical device to which they reside. Inside, designers must rely on signifiers to signify the affordances of the device to the user. An example for this is the use of arrows in a mobile application to signify the affordance of touch to move between app views.


Charles Park - 1/28/2014 21:05:13

An example that pops into mind is something simple that we all use everyday but still have issues using it all the time; this item is the shower knob. One of the most frustrating thing I’ve encountered over the years is the first few moments when I try to figure out how to use the shower knob at my friend’s house. (Not to mention how it’s impossible to get the water to the desired temperature. The experience become a lot worse when it’s early in the morning and you feel like you’re going to freeze to death.) While all of us have become accustomed to the shower knobs at our homes, it is a common practice for the owner to explain to the guest how to turn on the shower because this product is counter-intuitive. The product lacks signifiers which makes it rather difficult to intuitively finding the means of using the product. It lacks consistency as there are too many different variations of the product which makes it difficult to use. I think a nice design is already available in which the shower knob is replaced by a handle which makes the turning on and off extremely easy and intuitive, as well as changing the temperature.

Affordances determine what actions are possible while signifiers communicate where the action should take place. For example, a user can determine that a cup can hold things inside of it (affordance) and the handle one the cup signals to the user where to hold the cup (signifier). The affordances are what the product can do in relation to the user and the signifier gives clues concerning how one can use the product. These things are not limited to products. (The example from the reading about determining whether or not one missed the train through the presence or absence of people waiting.)


Ian Birnam - 1/28/2014 21:53:44

1) My car doesn't have a display for it's own temperature, choosing to just have a visible signal light up on the dashboard when the engine overheats. The signal is also accompanied with a sound for audio feedback. To me, this seems like poor design because I don't know about the problem until it happens. If I knew the car's temperature in real-time, I could possibly catch the overheating in advance, or even avoid it all together. This violate's Norman's feedback principle, since I as the user am not getting what I consider to be sufficient feedback from the car. To re-design it, I would simply put a gauge in the car that showed me the car's temperature in real time, and would allow me to spot overheating as soon as possible.

2) As stated in the reading, affordances are the possible interactions between people and the environment/object. Some are perceivable, others are not. They are not properties, but rather relationships. Signifiers signal things, such as what actions are possible and how they should be done. Signifiers have to be perceivable, otherwise they serve no purpose. They tell the user how to interact with the object.

Though affordances may be similar across physical products and software user interactions, their signifiers can be completely different. For example, a pen affords drawing and writing, and therefore affords communication. The squishy, gel area near the tip of some pens is a signifier, as it indicates where users should put their fingers. The push-top on most pens is another signifier, indicating how to reveal or hide the tip of the pen.

Let's look at a drawing app for a phone or tablet to compare. Much like the pen, the app affords drawing and writing, which therefore affords communication. Maybe the app affords sharing by allowing the user to upload their work to social networks or email. However, the signifiers for the app don't involve squishy gels or push-tops. Signifiers for the app could be arrows that show you where to draw. They could also be simulated examples of digital fingers doodling on the app, to show users that they need to use their fingers for drawing. An envelope icon could be a signifier to attach your piece to an email. What may be a great signifier for something physical may not necessarily work for something digital.


Andrew Dorsett - 1/28/2014 22:49:47

An example of a physical device is the shower switch in a bathtub shower. It's usually a little piece of metal sticking out of bathtub faucet that when pulled up causes the water to redirect to the shower head. This device lacks signifiers. If you had never seen it before you wouldn't know what it does. It looks like a drain stop on the sink but doesn't act the same way. The designs also vary between manufacturers. Some switches reset back to the tub when turned off while others don't. It's typically hard to tell what position the switch is in without taking a few showers. If I redesigned it I would make it dial of sorts with a symbol of a shower head above it and faucet on below. Like where they're located in the shower itself. There would also be an arrow that showed what it was currently set to.

Affordance is what something does for a particular user. Signifiers tell you have to operate something in order to achieve that particular use. For example a folding chair affords the user a place to sit. The bolts and joints signify that the seat must be rotated. The chair may also afford a person of a certain weight an object to stand one. The signifier would be its material. A chair made out of flimsy plastic would signify only a petite person would be able to stand. Physical products such as a chair offer their structure, size, and shape as signifiers for their affordances. Software uses symbols, words, animations, etc to signify it's affordances.


Vinit Nayak - 1/28/2014 23:08:28

The one device that has simultaneously captured excellent affordance yet has a very poor signifier is the USB cable/stick. The fact that it affords memory transfer and storage in such a small device provides very high convenience, yet the biggest pain by many of its users (including me) is not knowing which direction to put the USB drive into the machine reading it. There is a 50% chance of inserting it correctly, yet it feels as if we are inserting it incorrectly >50% of the time. This doesn't cause a large time delay or isn't fatal in any way, but it gets unnecessarily frustrating. It violates Norman's principles of signifiers, since the user does not immediately know how to use the device as designed (which orientation the machine is expecting the USB in). To solve this, we would need to create a universal convention for both USB cables as well as the devices themselves to put some kind of indicator, something as simple as a dot. The user would then align the side with the dot on the USB cable and on the machine and would successfully insert the USB on the first try! Another example is stove top dials to turn on the stove. The stove coils are arranged on the top as a 2x2 matrix but all the dials are arranged in a horizontal line, making it difficult to know which dial maps to which stove top. There is a poor signifier which shows the picture of the stove the dial is associated with, but the need for a signifier could be eliminated altogether if the dials were arranged in the same format as the stove coils. So each dial, in a similar 2x2 matrix, would correspond to the coil it controls.


Norman states that affordances determine what actions are possible, whereas signifiers communicate where the action should take place. Signifiers can be thought of as instructions, in some sense, that must be perceived by the user to instruct them how to use the object. Affordance can implicitly or explicitly show what the object is used for (the book states, what the object can "afford" the user). Affordances are more implicit with physical products as their visuals and the way they are designed can provide many indications to a user of what the product can do. For software products, affordances might need to be more thoroughly stated since the medium used to interact with the product (mobile device, website, etc) is the same medium that is used for many other products with a wide array of affordances. Therefore, whenever one downloads and application or goes to a new product website they usually have a quick blurb about what the service provides. For signifiers, they need to be more explicitly stated for software products as compared to physical products. This is because a wide demographic of people view the exact same webpage, or mobile app and their different backgrounds could make them unfamiliar with how to interact with the product. That's why software products are usually accompanied with tutorials or tool tips that follow a user the first them they are using the software. For physical products, the user most likely has seen something similar in nature before or might have an easier time figuring out how to use it by adding pieces of its physical design and the products affordance.


Luke Song - 1/28/2014 23:24:57

Probably the one frequent thing that annoys me the most is figuring out how to adjust chairs. There are an almost uncountable different chair designs, all with their own knobs, levers, and the like to adjust things like height, tilt, flexibility, and more. Sometimes, these things have signifiers, and sometimes they don't. In the latter case, it is up to I, the user, to do a bit of experimentation to support my hypotheses about what each control affects. In my opinion, the best chair designs have symbols engraved on each control which clearly illustrates its function.

From the author's definition, an affordance is a function of the product, while a signifier allows the user to know about that function. For a physical tool, such as a cabinet, the signifier is the handle, which points out the cabinet's function as a storage device. Similarly, a USB drive might open its own folder on the screen of the computer it's plugged into which signifies its own function as a storage device. There are many different things that can be called signifiers and affordances, both in software and hardware, they always help the user complete a task more quickly and easily.


Conan Cai - 1/28/2014 23:36:13

USB plugs are something that are used everyday, but they always leads to frustration. You try to insert the USB plug. It doesn't fit. Flip it over. Try again and it still doesn't fit. Flip it once more and oh, look at that. Now it fits. It just seems like I never manage to insert the USB plug the right way the first time. I believe that part of this is caused by violating Norman's design principle of discoverability. Discoverability allows a user to intuitively and quickly use a product. The physical shape of a USB plug is a rectangle. This symmetry suggests that a USB plug can be put in whichever way, because after all, the rectangular plug should be able to fit into the rectangular port. This is not the case; the plug needs to be put in "right side up." The symmetry makes it hard for a user to quickly ascertain the proper orientation of the plug. A solution would be to change the pinout of the plug/port to be able to support any orientation. In this way there is no "right side up" and a user will be able to just plug a USB in without fumbling with which is the right way.

An affordance is like an ability that is possible when an object and user interact. An affordance is a relationship, and for an ability of an object to be used, a user must know an ability exists and how to use it. This is where a signifier comes into play. A signifier notifies the user and makes it clear that such an affordance exists. A physical object may need less signifiers than a software product. A physical object inherently lends some clue to a user on how it operates. Basic properties of physics lead to intuition on the users part on how an object moves or otherwise interacts. Software on the other hand is completely virtual - completely defined by the programmer. Because of this, there is no basis from which a user can built intuition. Signifiers are important to teach a user the "rules" and what affordances exist in the virtual software world.


Tien Chang - 1/28/2014 23:40:33

1) A physical device with bad design is a stationary spinning bike. The ones I use in the RSF lack in discoverability. Without the help of experienced riders and the group exercise teacher, I would not have been able to figure out what each red knob, button, and lever did. Without discoverability, understanding was also lacking. This violates the design principles Norman stated as the most important characteristics of good design. While the machine had a display with minimal instructions, perhaps a redesign to solve the problem would be to label each knob, button, and lever with its respective functionality. It would allow users to operate the machine without needing help. Norman was right in that good design should not only include happy paths where users correctly use a product, but also sad paths where users will incorrectly use a product.

2) Affordances are potential interactions between a user and a product, while signifiers are signals of what actions are possible. Anti-affordances can also exist, in which interactions between a user and a product is prevented. Affordances of physical products are interactions a user can have with that product in an environment, while affordances of software user interfaces are interactions a user can have with that software on a device. Signifiers of physical products would include additional physical signs of how to use a product, while signifiers of software user interfaces would have how-to signs within an application.


Zack Mayeda - 1/28/2014 23:41:30

1) The kitchen stove is a device that I often find has bad design. In particular, they often violate Norman's principle of mapping: the connection between knob and burner is often confusing. Stoves also violate his principle of feedback: if an incorrect burner is turned on, the user may not notice until several minutes have elapsed. Many stoves have different types of labels and arrangements of knobs, which forces the user to learn a new interaction on many occasions. I would re-design the kitchen stove to have uniform knobs, regardless of stove type (electric, gas, induction). I would assign each knob a color and include a line on the cooktop between the knob and the burner. In addition to, or in place of the color pairings, a circular light around the burner could flash on when activated to provide visual feedback to the user about which burner they are using. 2) An affordance is a relationship between two objects - a way that they can interact. A signifier is a signal that alerts people to an affordance, hinting at the existence of a possible interaction. Affordances exist for both physical products and software user interfaces, but are much more diverse for physical products. For software user interfaces, interaction is largely limited to touching a screen or using a mouse and keyboard while physical products have countless types of possible interactions. Signifiers can, and are, included in physical products, but they are much more deliberate for software user interfaces because the user has less given context/information. Like Norman's example, a book provides clues about exactly how much of the text you've read just by holding it, but a digital textbook must deliberately show the number of pages remaining on screen.


Erik Bartlett - 1/28/2014 23:45:04

1) The heart rate monitor I just purchased has bad design. It has three functions: keeping the time of day, displaying your heart rate, and displaying the amount of time you've been monitoring your heart rate. The problem is you cannot switch between the displays to see how long you've been working out unless you pause the monitoring, and you cannot see what time it is unless you completely exit the monitoring portion. Also the button you use to pause is also how you unpause, while the starting button exits the monitoring portion. This design has poor mappings (the pause-unpause-restarting buttons) along with a poor mental model. I would design it so there is a button that goes switches between Time of day, heart rate, and stopwatch. Also I would make the timer/heartrate pause and reset be done by the same button (tapping once to pause, twice to reset like normal stopwatches). This better fits my mental model of the watch having three functions which can be switched at anytime.

2) An affordance is a use/act of a product that is determined by the relationship of the object and the user - i.e. a chair can be sat in by everyone, but not necessarily lifted by every user. A signifier is something that helps the user understand an affordance of a product - i.e. "slide to unlock" on iPhone lock screens or the word "power" next to a power button. Affordances in physical products generally differ more by user than software UI's do - for example everyone has the ability to use every function of microsoft word, while not every person would be able to use a sword in the same way. Signifiers in physical products and software UI's main difference is that many signifiers in software are hidden and must be found by clicking a tab or navigation, where as in physical products generally all signifiers are visible all the time.


Hao-Wei Lin - 1/28/2014 23:51:20

1. Taking BART can be a pain at times. One of the things I don’t enjoy about BART is that during rush hours, people swarm up frenetically to the edge of the platform, before the train arrives and it becomes really crowded. The worst thing is that there is no specific line formation during this kind of rush hour; people are pushing, trying to get in the train while cutting in line. It was a chaos because often time the people start swarming into the train even before the passengers can get off completely. What the BART lacks is a good signifier of where people should line up. A possible solution to this is to have actual lines on the floor that point to the door and diverges to both sides of the door, indicating where people should line up to get onto the train, while having another set of lines coming straight out of the door, signifying the direction to which the passenger should get off (maybe with arrows)

2. Affordance is the connection between a design, or a part of the design, and its function, or capacity when a user interacts with the design. For instance, a desk lamp affords the illumination of the surface of the book one is reading, and also affords portability. Signifier, on the other hand, is the part of a design that signifies a particular affordance. A successful signifier should tell the user where to look for the affordance, if not explicitly telling what the affordance is. For instance, a symbol of a house on an electronic device is the signifier that tells the user if he/she hits the button, it will bring him/her back to the home screen.


Shana Hu - 1/29/2014 0:22:27

Staircases in homes often have two light-switches which control the same light source located in the middle of the stairwell. One switch is at the foot of the stairs, whereas the other is at the top. Both switches work fine, but they offer an inconsistent interface which users interact with. If the light starts off, a user at the top of the staircase might flip the light-switch upward to turn the light on. Later, when the user descends the staircase, he or she might turn off the light--by flipping the light-switch at the foot of the stairs up. The user interacts in the same way with both light-switches by flipping them upward, but the same action creates two different results, light-on and light-off. This design violates Norman's principle of consistency, and makes users question their actions when the task should be simple and straightforward. To re-design the light-switches, it's important to move away from an interface which relies on manual configurations to denote the status of the light. Since up is usually associated with "on" and down is associated with "off," the use of the light-switches as literal switches is unhelpful. It would better benefit users to have a solution such as a button which also just changes the light from on to off, or vice versa depending on the current state. The button offers no confusion about the state of the light. Although the button itself may not directly offer feedback, the user can just observe the light itself, and there are always options to incorporate a factor of glowing in the button, which is automated based on the light's current state.

Signifiers are perceivable denotations of potential affordances. Affordances are experiences that users have with interfaces. Affordances are more limited for software user interfaces because the interface is typically a flat screen. Signifiers for software are usually designed as buttons and other clickable widgets. For physical products, there is much more potential for rich interaction. There is more variety in signifier designs, but also, there is more potential for signifiers to be misleading due to the creativity of human interaction.


Andrew Chen - 1/29/2014 0:40:17

1) I have an electronic toothbrush that is started by pressing the only button on the toothbrush, located on the handle, right below where the brush head is supposed to be inserted. The bad part of the design is that when the button is pressed once, the brush vibrates at full power; if it is pressed again, it vibrates at a lower amplitude; and if pressed a third time, it turns off. This design violates Norman’s signifier and conceptual models design principles. Firstly, it violates the signifier principle, because there is no real signifier that tells the user what power the toothbrush is vibrating at (besides the vibration itself). Secondly, it violates the conceptual models design principle, because from what we expect, the only button on an electronic device should simply switch the device on and off, not control the strength of the device. A possible way to redesign the toothbrush involves using a vertical slider with a small knob instead of a button, so that the device can be switched on and off as well as finely adjusted.

2) An affordance is an aspect of the design that allows the user a certain action, whereas a signifier is an aspect of the design that suggests the user a certain action or condition. Affordances and signifiers differ for physical products and software user interfaces in that physical ones are limited by the physical specifications of the product, as well as the laws of physics, whereas software ones do not have these constraints. In addition, in some ways, physical affordances and signifiers are predecessors to their software counterparts, because software affordances and signifiers are in many cases designed based off of successful physical models, because users have become accustomed to using the physical ones.


Andrea Campos - 1/29/2014 0:44:39

1) A device with really bad design I've encountered is the push bar device found in SF Muni trains, used when someone wants to open the doors to get off at a stop. The main problem with it in Norman's terms is its "discoverability," due to a lack of signifiers and bad mapping. The push bar on observation looks just like a railing at the entrance to the train, useful for people standing on the street to lift themselves onto the raised step. People already on the train have to stoop to be able to touch the push bar, and so it is not very useful as a railing until the last step. Because of the awkwardness of having to stoop, and the appearance of it being a mere railing, people often don't even see it or think of using it as a means of opening the door! Who would guess that pushing the railing down should open the doors? Such an illogical mapping. Plus, there are not very many signifiers like signs or instructions present, at least not that I can remember from my frequent use of the trains, and that in itself indicates their poor visibility or absence. San Franciscans may have gotten used to this horrible device, but the many tourists that use the trains often get extremely confused. I would redesign the device to open doors by making it a visible button of some sort rather than a push bar on the side. I would include large, clear instructions on the door themselves, and place the button in an easy to reach location that doesn't require bending or stooping, and that is in a logical place--perhaps like on the door itself!

2) An affordance is the relationship between an object and a person, defined by not only what the object itself can do, but also by the capabilities of the person in using the object. Meanwhile, a signifier is something that indicates and brings attention to an affordance that may not be immediately visible. So, an affordance is a relationship, and a signifier lets us know that this relationship exists. I think often affordances are often more intuitive for physical objects. You can walk around something, explore its shape, examine its details to help ascertain how you can use it. The signifiers don't always have to be so purposeful, because the affordances can many times be perceived through the object itself. For software UI, you mostly only have the flat screens, that often can't tell you much without the help of purposeful signifiers. An icon can suggest many things, but unlike a framed poster on a wall, can have multiple functions we would not be aware of without explicit signifiers. Can we merely click it, can we drag it, can we swipe it off the screen, can we remove it? Unlike a framed poster, which we can surmise how we can act on based on its size, weight, and the environment itself, software UI exists in a (mostly) 2D world in which the affordances are often hidden to us, and so we must rely more heavily on signifiers.


Doug Cook - 1/29/2014 0:51:05

One design that often bothers me is the coin-operated laundry machine in my apartment unit. After loading the coin slider and pushing the slider in to deposit the coins, the machine automatically begins filling with water and follows whatever options were previously set on the dials. The matching dryer next to it, by contrast, requires a button press after inserting the coins in order to start the cycle. It seems that Norman’s principle of providing signifiers for invisible affordances isn’t being used in the machine’s design. There is no indication that inserting coins starts the machine, and the dryer behaves oppositely, which confuses my conceptual model of what to expect. To solve this problem I would first place a text-based signifier that reads, “insert coins to start cycle” on the machine so that it’s obvious what happens next. I would do the same to the dryer (removing the button-press afterwards) so that both appliances are consistent in their design.

An “affordance” is a relationship between a product and the person who will be using it, based on what functionality the object provides to the user. A “signifier,” unlike an affordance, is not a relationship but some perceivable indicator of how a person should act. These concepts differ for software user interfaces in that the domain of potential affordances is more constrained. While an interface could still have some of the accidental signifiers that Norman observed, it’s confined to a digital environment that designers have significantly more influence over. The range of physical interaction is also restricted to what the machine’s IO and display permit. Due to these restrictions, clear signifiers are critical to a software interface’s effectiveness.


Rico Ardisyah - 1/29/2014 0:52:01

Some physical devices in our daily life, indeed, have bad design. A device considered as a bad designed device is when the users of the device keep making mistakes, even though the users have used the device for couple times. One of the examples that I have used it is a stove. The one that I used has 4 burners and 4 knobs; I believe you have the same one. However, the designer arranged all knobs next to each other, and the reading is front left, front right, rear left, and rear right. At first, I am confused between rear and front. Moreover, the reading is quite small so I cannot read it clearly. Based on Norman’s principles, the stove does not have a good mapping. The knobs does not clearly map to which burners. The way I will re-design the stove is by re-arranging the position of the knobs so that the position of the knobs maps the actual burners.

An affordance is the property of the device that can be used by the user. Hence an affordance depends on both the device and the user. For instance, iPhone has an affordance browsing for me, but it does not for my grandmother who does not know how to use iPhone. While, signifier shows users what to do with the device, or it can show the affordances of a device. In physical products, both affordance and signifier has important roles. However, without signifier, the physical product may still work even though it is hard to use. In software user interfaces, the principles are the same. Both of them are important parts. Nevertheless, signifier has to be presented; otherwise the user will not use the affordance since without signifier the software will not work properly.


Christina Guo - 1/29/2014 0:59:20

One example of a physical device with bad design was a elliptical at a gym in my hometown. It offered different types of workouts, each with a specific incline and resistance. Each workout had a different button on the interface. However, there was no power button, and if the first thing you did was press one of these buttons, nothing would happen. The machine turned on once you started moving, and then a workout could be selected. However, the first time I used this, I stood for a few minutes pressing random buttons, unaware that I needed to move in order to turn it on. This violates the signifier design principle, because there was no indication for how to turn on the machine. In order to fix this, I would either allow the workout to be selected before the user started moving, or have text on the interface telling the user that the machine would automatically turn on after moving.

Affordances are based on the relationship between an object and it's user. It is essentially the ways an object can be used as a result of properties of the object and the capabilities of the agent, or the person using the object. For example, a chair always affords sitting (by virtue of being a chair), but can or cannot afford lifting, depending on the weight of the chair and what it is made of. Signifiers, on the other hand, communicate to the user how the object should be used, as in they communicate the existence of affordances.


Ryan Yu - 1/29/2014 1:40:16

One example of something with bad design would be the water bottle holder in my backpack. The design in question involves a sort of mesh pouch that lies on both the left and right side of the backpack, near the bottom. However, on my backpack, the mesh pouch is only about four inches tall or so, which makes any water bottle that is larger than a standard plastic bottle *not fit* into the pouch. In fact, some slightly oversized water bottles fit into the pouch, but fall out when the backpack is picked up at a slight angle, which is obviously a very big inconvenience for the user of the water bottle.

The water bottle holder for the backpack is good in Norman's principles of *discoverability* and *understanding* (it is easy to figure out what the pouch is supposed to be used for, and how it is supposed to be used). However, the holder is lacking in the fields of industrial design and experience design. The product is not optimized for the mutual benefit of both the user and the manufacturer, in that the user might still find it inconvenient. Furthermore, if the user's water bottle is continuously having trouble fitting within the confines of the holder, then obviously his/her experience is being damaged, and the whole purpose of the holder is defeated. Overall, the idea of "human-centered design" is not really satisfied by this feature of the backpack -- I feel as though the designers should have looked at what users really want, and should have done some research as to typical sizes of water bottles users carry.

To improve the design of the water bottle pouch, I would make it both wider and taller, maybe to about six or seven inches in height. I would then add a zipper to the top of the pouch, so if a user put a regularly-sized bottle in the pouch, he/she could zip it up to prevent it from ever falling out. Obviously, this would not prevent larger bottles from falling out if the backpack was tipped upside down, but at least it would alleviate some problems with the most common bottle size.

---

Essentially, an affordance represents the *possibilities* in an environment for how someone (or something) can interact with something else. On the other hand, a signifier represents a *signal* that directly informs a user how to do something. A great example, I think, was the description of a row of vertical pipes across a service road in a public park. The pipes visibly blocked cars from driving on the road, and were *anti-affordances* in that they eliminated the *possibility* for a user to drive across the road. On the other hand, an example for a signifier could be something as simple as a word next to an up arrow that signifies what happens if a user scrolls up on a page.

For physical products, an affordance can act more or less like a signifier -- for instance, in the example described above, the rubber pipes sticking out in the middle of the park acted as an anti-affordance, but also acted as a good *signifier* (to the normal person) that road was blocked to vehicles. In other words, perceived affordances often can act as signifiers (but can be ambiguous, sometimes.) On the other hand, for software user interfaces, affordances actually tend to *afford something new* to the application -- affordances actually adds possibilities for the user to do different and new things within the scope of the user interface. Signifiers, on the other hand, merely signal the user how to operate or use something that *already exists* within the scope of the application.


Jay Kong - 1/29/2014 1:44:39

Dishes in dishwashers collecting water after wash. This problem arises because the layout of the dishwasher requires the dishes to be put upside down. This device violates Norman's design principles of good mapping, as the interior of the dishwashers can only hold dishes that are fitted a certain way. I would redesign the dishwasher to have "angled" slits so that dishes sit at 45 degree inclines, which prevents water from being collected.

An affordance is a relationship between the properties of an object and the capabilities of the agent that determines just how the object could be used. A signifier helps clarify that relationship. Affordances for physical products can be conveyed through the physical properties of the product; however, for software, there are no physical properties. Affordances come in the form of familiarity -- for example, an agent would expect to be able to ctrl+c on any selectable text area. Apart from the digital/phsyical sense, I believe there are fewer differences for signifiers in physical products and software UI. In terms of physical product, there might be a knob or a button. In terms of software UI, there might be a scrollbar, an arrow for a dropdown menu, or even a button. The goal of these signifiers are the same, but they just come in different forms.



Jay Kong - 1/29/2014 1:47:15

Dishes in dishwashers collecting water after wash. This problem arises because the layout of the dishwasher requires the dishes to be put upside down. This device violates Norman's design principles of good mapping, as the interior of the dishwashers can only hold dishes that are fitted a certain way. I would redesign the dishwasher to have "angled" slits so that dishes sit at 45 degree inclines, which prevents water from being collected.

An affordance is a relationship between the properties of an object and the capabilities of the agent that determines just how the object could be used. A signifier helps clarify that relationship. Affordances for physical products can be conveyed through the physical properties of the product; however, for software, there are no physical properties. Affordances come in the form of familiarity -- for example, an agent would expect to be able to ctrl+c on any selectable text area. Apart from the digital/phsyical sense, I believe there are fewer differences for signifiers in physical products and software UI. In terms of physical product, there might be a knob or a button. In terms of software UI, there might be a scrollbar, an arrow for a dropdown menu, or even a button. The goal of these signifiers are the same, but they just come in different forms.


Sangeetha Alagappan - 1/29/2014 1:47:43

1) A physical device that is often badly designed is a shower dial. Having used a variety of shower dials while at hotels, I’m usually frustrated with figuring out how the dial works. Often dials have different maneuvers to turn the water on (twist either clockwise or anti-clockwise, lift and then twist, etc.) and once the water is on, it’s difficult to regulate the water pressure and temperature and to divert the water through the shower head rather than fill the tub. I once used a dial that had no markings except for a blue and a red dot on either end of the dial. According to Norman, good design requires good signifiers in order to communicate the operation of the device to the people who use it. Shower dials often have poor signifiers and require some trial-and-error to figure out how they work - something Norman despises claiming that machines should be designed for humans and should not have to be painstakingly 'figured out'. Often times the ‘H’ and ‘C’ stickers meant to be signifiers for hot and cold water do not correlate with how the dial works (many times cold water is turned on when the dial is at H) and thus, its mapping does not use spatial correspondence between the layout of the controls and the devices being controlled. This violates Norman’s design principle of good, natural mapping. The device provides no feedback as well and has the option of temperatures/heat settings that are probably never used and only are a cause of hazard. I have a mental model of old school shower controls when I think of regulating water in the shower. Two separate knobs; one for cold water, one for hot. Each tap controls water pressure. However, one shower dial is more convenient than two taps. I would redesign the dial to control just water pressure (that sweeps counter-clockwise, with markings along the circumference of the dial indicating levels of pressure - low, medium, high - and an arrow indicating how to turn the dial). I would also opt for a temperature display of the water at the centre of the dial and an up and down button to adjust the temperature. This would require more work/button pressing but will provide water of the desired temperature without dial fiddling. Also, I would opt for the stopper on top of the tap that has to be pulled up for water through the shower as it mimics natural signifiers; pulling up would mean water from the top.

2) An affordance is a relationship between an object’s properties and an agent’s capabilities that determine how a product is going to be used. A signifier is a perceivable indicator that communicates appropriate behaviour when a person uses a device. Therefore, affordances determine what actions are possible while signifiers show where actions should be taken. In physical products, affordances come about as a result of certain properties of the object (a chair affords sitting as it affords support) while in software user interfaces, affordances are actions that are programmed to be done (the program’s functionality). In physical products, signifiers can come about by a variety of ways - there is a possibility for natural or unintentional signifiers. These are rarer in software user interfaces as they are not ‘natural’ objects that can mimic things like amount left to read as with bookmarks or trails previously travelled by. Most signifiers need to be designed when it comes to interfaces. Signifiers are often different when it comes to software; it involves movements like swiping and finger sliding rather than more natural human movements of pushing and pulling, and thus software user interfaces require more instruction and detailed signifiers.


Seth Anderson - 1/29/2014 1:47:48

1) One physical device that I use everyday which has less than optimal design are the laces on my shoes. Not only do I have to bend over multiple times a day to tie them when they become undone, I often go a long time without knowing they are untied, putting me at risk of injury until I look down or someone gives me the infamous tip, "Your shoes are untied." This violates two of Norman's design principles. One violation is for signifiers: there is no clear way to mark how the shoes should be tied, the user is expected to know how to tie the shoes beforehand. Secondly, there is a lack of feedback in the shoes: when they come untied, there is no clear way of letting the user know that they have done so.

I would redesign the shoelaces to fix this problem by first adding some sort of feedback on the shoe to notify the users the laces have become untied. I would add small sensors at the end of each lace, that when bounced off of the ground after being untied would send a quick vibration through the shoe, letting the user know the shoe has become untied. Secondly, to aid in the tying of the shoe for beginners, I would install a flap on the top of the shoe that when flipped up shows visual instructions for how to tie a good knot.

2) Affordances are relationships that determine what is possible with relation to an object, while signifiers are visual, aural, or tactile stimuli that show to the user where affordances are possible. In physical objects, affordances are limited by physics and the capabilities of the users as well as the product, and thus are fairly open ended. In software, however, the affordances of software are generally strictly set in place by the designer, and their physical affordances are limited by the device they function in. Signifiers in physical products and software are more similar to one another, in that both are simply ways the designer has implemented symbols or guides to direct the user to the affordances.


Steven Pham - 1/29/2014 1:53:13

When I stay at other homes or in hotel rooms, some shower knobs are very difficult to work. There are some that are very easy like two knobs that are labeled with red and blue. Then there are the hard ones where it’s a single knob that you turn to control the temperature but some you have to push in to turn on the water or some you have to pull out. Some even have you push the knob to the temperature you want it. In this particular case I’ll use the single knob push in version: 012028-concord-single-handle-shower-faucet.jpg. First it violates consistency since not all shower control mechanisms are the same. There is a mapping for the temperature however there is no clear indication, signifier, to push in the knob to start the water. The knob in the above picture seems to be attached to a metal cylinder that sticks out of the wall. I would be it so that there will be marks providing a mapping on the cylinder that will be labeled with the on or off. So that when you push the knob it hides the off so you clearly see it is on and when it is all the way out you see the off. According to Norman, affordances determine what actions are possible and signifiers communicate where the action should take place. Physical interfaces are pretty much set and stone after the manufacturer builds it so its complexity if it had any would not get any worse. With software there are so many possibilities of interfaces and they can easily be modified and updated. You can constantly add new affordances or modify signifiers in the software causing confusion. With physical interfaces, it is not as easy or malleable.


Brenton Dano - 1/29/2014 2:01:09

1) Norman mentioned two principles in good design, discoverability and understanding. My microwave (which I bought used from some people who were moving out of my apartment on the day I moved in) lacks in the "understanding" area of good design. It has buttons on it like "Auto defrost", "Popcorn", "beverage", "Pizza", "potato." Since I don't have the manual I have no idea what they mean. I've tried pressing some before and sometimes it tries to make you enter a weight or a time, with no clear labels or idea of whats going on! Frustrating to say the least, and I'm not going to get started on the unpressable buttons! I could redesign this microwave by having a scale built in so users won't have to manually enter the weight. How are users suppose to know the weight of their food anyways? Perhaps, a minicomputer inside the microwave will help it compute what settings should be used. Also, I will have voice command so all the user needs to do is open the microwave, insert the food, say what the food is, and the computer will do all the work of weighing and calculating how long to heat it.

2) An "affordance" is a relationship between an object of some kind and the person interacting with it. The object by itself doesn't "afford" anything without the person interacting with it. A good example that Norman mentioned was that glass affords being see-through. This interaction of seeing-through glass requires both the glass and the observer. A "signifier" is a term used to describe something that brings attention to an "affordance." Simply put, if some object has an affordance with a user and its not obvious, it is necessary for it to have a "signifier" which alerts the user of this relationship.

In physical products, an affordance and signifiers will be something physical like a sign or some sort of physical thing that sticks out and makes it obvious what to do. For example, a pull handle on a door points you to the relationship between the door and your hand and hints that you can pull it. Obviously, software UIs can't have things that physically stick out but in a sense you can have it stick out to the user by highlighting certain areas, or having arrows, indicating swiping actions or tapping actions.


Will Tang - 1/29/2014 2:41:33

1) While many car companies have probably solved this problem already, I often have problems remembering which latch in my 1997 Toyota opens the fuel cap, and which latch opens the hood. The two latches are placed right next to each other, and are located on the left side of the driver's seat on the floor of the vehicle. While they are clearly labeled and their purpose is clearly communicated, they are hidden from view unless the driver side door is opened. This is rarely a problem as I'm usually going to get out of the car if I need to pull either of these latches, but I feel that the design could be better. The only real design principle from Norman's writing that this violates is the mapping principle. If I can't remember which one opens what, then the mapping has failed in my eyes. To solve this problem, I would probably put the latches on somewhere low on the dashboard, and perhaps near the latch that opens the trunk. With a clear line of sight to the labels indicating trunk and fuel cap, mix-ups would probably be less frequent. 2) An affordance is something that a physical object can do in its interaction with a user, and a signifier is something that can indicate to the user what actions should take place. For physical products, affordances are likely some sort of physical capability like a shovel moving dirt. For software user interfaces, affordances would be tasks that the software can perform, such as how a Microsoft Word affords the entering of text. Signifiers are more similar, with a difference being that unintentional signifiers such as a game trail are much more common in the physical world.


Nahush Bhanage - 1/29/2014 3:18:59

1) I can totally relate to the part of text where Norman describes frustration caused due to a poorly designed product. I've had an experience of using a pen that had a really bad design. Appearance-wise it looked like a typical ballpoint pen, with a push button at its rear end. When I pressed it, expecting (like anyone would) a ballpoint at the front end, I was literally blinded by a shockingly bright blue light - this button lit up the entire body of the pen. Regaining my composure, I looked for other possible ways. I tried rotating the front half while holding its rear end still. This literally opened it up. On reassembling it, I finally managed to draw out the ballpoint by pressing an inconspicuous notch on the side.

I think this pen violated at least four of Norman's design principles:

a) Signifiers - Lack of signifiers, to be precise. If the push button is not to draw out the ballpoint (unlike all other pens), there should have been an indicator specifying how to use it. Also, there was nothing on it to indicate that the push button was to light it up.

b) Mapping - I'm sure every person using this pen would try the push button first. If the designer wanted to add a new functionality in the pen, he/she should have mapped it with a new button/notch instead of the push button. This kind of mapping was the major reason behind the confusion.

c) Feedback - Even if I somehow knew that the push button would light up the pen, the intensity and brightness of this light would have startled me. Having a light in the pen turned out to be a frustration instead of a useful feature.

d) Conceptual Model - Based on the reasons mentioned above, this design would definitely not be aligned with the conceptual model of a pen in the user's mind.

If I have to redesign this pen, I would map the ballpoint with the push button. This would ensure that the users won't be confused with its basic functionality. If this pen has to have a light, then I would map it with a new button near the rear end along with a small indicator (for instance, an image of a bulb). Assuming that its purpose is to enable the user to write in low ambient light conditions, I would place the LEDs close to the ballpoint rather than lighting up the entire body. I would also ensure that the light isn't intense enough to hurt the eyes.

2) An affordance represents a possible way in which a person or a machine can interact with an object, whereas a signifier signals the presence of the possible action. In other words, affordances determine what actions are possible whereas signifiers communicate where these actions should take place. For instance, the touch feature is an affordance of a mobile screen and the "Gmail" icon is a signifier which communicates to the user that Gmail app can be launched by touching that particular area on the screen.

In case of software user interfaces, the range of possible actions is limited. For instance, pressing keys on the keyboard, pointing and clicking with the mouse. These actions are abstract as compared to the physical manipulation of objects which is the case for physical product interfaces. Another difference would be that physical affordances are more or less obvious and intuitive in most cases whereas affordances of a software UI might be confusing for a new user. Hence the software UI designer has to rely upon conventional interpretations of different symbols and their placements.


Jimmy Bao - 1/29/2014 4:08:26

1) At my apartment, all the sinks have turning knobs, as opposed to the easier (in my opinion at least) handle that you can just lift up and down. These sinks in particular are bad designs because there aren't any external signifiers that tell the users which way to turn the knobs so that water will come out of the faucet. The knobs are marked with the letters "C" and "H" to signify cold and hot water, but fail to indicate whether the user should turn the knob clockwise or counterclockwise so that water would be dispensed.

Norman stressed that a good design requires good communication by indicating what actions are possible, what is happening, and what is about to happen. I feel this principle was violated because there isn't good communication between the user and the device. I definitely didn't and couldn't figure it out by just looking at it. It took trial-and-error for me to figure out which way I should turn the knobs if I wanted water. To solve the problem, I would also just include clear indicators (perhaps arrows pointing clockwise or counterclockwise) so that the users know which direction to turn. This may already exist as I'm not a sink expert, but it's a problem in this household.

2) Affordance is the relationship between the properties of an object and the capabilities of the agent that determines how said object could be utilized. Signifier is the presence of some indicator (any marks or sounds) that can successfully communicate the appropriate behavior to the agent.

Affordances differs for physical products and software user interfaces in that there can be both physical or perceived affordances for physical products, whereas there are mostly just perceived affordances for software user interfaces since the agent is interacting with mainly software. As an example, a user could still physically touch a phone (physical affordance) but I would say that many of the important features that today's phone offers are all perceived affordances. Moreover, in many cases, I find that signifiers don't differ too much for physical products and software user interfaces. Signaling happens in both cases; the only difference is that some signifiers for physical products are actually tangible. I think in most cases, signifiers for software user interfaces are not. You can touch the screen or something, but you can't really physically touch the signifier as you would a door handle.


Romi Phadte - 1/29/2014 4:08:45

1) One example of an everyday device that has bad design is the USB plug. The plug is internally asymmetrical (there is only one way to plug it in). However, the plug is externally symmetrical. This means that the user often takes multiple tries to plug in a USB correctly. There is no signifier to easily indicate what side is the "top." One way to fix the problem is to indicate (signify) the top of the USB. Another way to fix the problem is to actually make the plug internally symmetric and afford it the ability to plug it in multiple directions.

2) The book does a fantastic job of giving the difference. "Affordance describes what actions are possible. Signifiers describe where the action should take place." For physical devices such as scissors, the shape required for the affordance usually are signifiers. For example the holes are naturally used for fingers. However, for software, the affordance and signifiers are not inherently related and require much more forethought.


Shaina Krevat - 1/29/2014 10:05:20

1. I was once staying in a hotel with my sister, and when she woke up she wanted to take a shower before we left. I was packing up my suitcase when she came back into the main room, unable to change the flow of the water to take a shower instead of a bath. We looked all over the shower nozzles to try to find the switch, with no luck. There was a shower spout, so clearly the shower/bath afforded to turn into a shower, but there was no signifier. Finally, I called the front desk and asked them the trick. It turned out that on the back of the long handle that turned on the water was a switch that wasn’t easy to see, but it turned the shower on. (On another note, to try to find a picture of the shower so I could remember how to describe it, I found a Yahoo! Answers page with the same question, so it wasn’t just my sister and I who had trouble with this design).

The design principle violated was the signifier. Having the switch hidden out of sight, in order to make the tub look better, made it harder to know how to get the functionality that the user wanted. In this case, there was a front desk to call for help, but if we hadn’t had access to someone who knew how the device worked, we never would have been able to get all the utility the device afforded. If I was redesigning the system, I would move the switch to somewhere more visible, for example on the top of the bath or shower nozzle, because that signifies that the switch would change the water direction in some way.

2. An affordance is something that an object is capable of. A power button is an affordance which enables? a television to be turned on. A signifier shows how something can be used. A power button with the power symbol on it signifies this button will turn on the television. Affordances and signifiers differ in that an affordance creates a capability and a signifier draws attention to that capability. With physical objects, affordances and signifiers often go hand in hand, like in the scissor example in the reading. The handle of scissors afford fingers going through them to use the device and signify where the fingers go. In software, actions can be afforded without being signified. It’s possible to have a pop up window afforded by pressing in a certain spot on the touch screen, but without a button or a signifier there, no one would know. Because many affordances are virtual in software, signifiers are not inherent (like they would be with the handle). It’s important when designing affordances to keep in mind signifiers that will clue in a user to the ability.


Andrew Lee - 1/29/2014 5:44:43

1) My floor lamp, which is one of the basic ones where you twist a thin rod to click the bulb on and off. Signifiers is the main design principle that's violated. There's no indication which direction the rod has to be twisted to switch it on and off. I know that it's clockwise only because from experience. It technically affords to twist the other way, but that does nothing. In fact, feedback is violated here because it does does click when it's twisted in the wrong direction, which gives the user a false hope that something will happen. Also, one has to have dealt with similar lights to realize that the rod is the on/off switch.

Additionally, that on/off rod is plain uncomfortable. It has a ribbed texture, meant to increase the grip on the user's fingers while twisting. However, the amount of torque that's required to twist it is remarkably high, and often, my fingers slip, which on its rough surface, can actually hurt. I find it interesting that had they reduced the necessary twisting torque, the rough surface would probably be unnecessary, reducing the finger pain that it would cause from slipping.

I would redesign the on/off mechanism as a vertical strip of touchpad, where touching the top would turn it on to full brightness, and touching the bottom would turn it off. Placing icons denoting an unlit lightbulb and a lit lightbulb on the top and bottom respectively would be a nice signifier. Additionally, touching the touchpad in intermediate locations can adjust the lightbulb brightness accordingly, so the user can even slide their finger up and down the touchpad to adjust the brightness to their liking. As a final touch, the touchpad can incorporate strips of light along its length that represent the current lighting level, which will move according to the brightness (AKA, right underneath where the user last touched it). This design has clearer signifiers on how to operate it. Also, the strips of light will follow the user's finger, which is a nice feedback mechanism. Because this is connected to the brightness, all of these map together into a simple conceptual model.

2) Affordances refer to the ways the user can interact with the product, while signifiers refer to features that hint to the user how they can use the product. Oftentimes, signifiers "signal" affordances.

For a physical product, affordances generally refer to how the user can use their body (typically their hands and fingers) to touch and manipulate it. Things like handles, buttons, and wheels can give the user an idea of ways to use it. For software UI, the product is usually not interacted with directly with a part of the user's body. For PCs, usually the interaction is through a peripheral, like mice and keyboards (though with touch-enabled devices, this is less the case). So here, affordances often refer to what is clickable, draggable, or where the user can type. For touch-enabled devices, it can refer to what is tappable, pinch-zoomable, and dragable.

For a physical product, signifiers generally refer to icons or labels that indicate the function of a particular part. For software UI, this isn't that much different. The set of icons and actions that are possible can be pretty different, but they're often sprinkled on the display to help the user understand what they can do. Something that software UI can do much more easily than physical products is to manipulate the display to throw a signifier in their faces, such as popping up dialogs. So because the display is easily manipulable, it's easier for software UI to change the visible signifiers than physical products, which are usually stuck with whatever it came with.


Ravi Punj - 1/29/2014 10:12:10

1) Machines that accept cards (credit, debit, prepaid, gift, etc.) with magnetic strips. More specifically the entry slot for the card. There are four possible ways to insert a card, and if I'm not paying a ton of visual attention to how I'm inserting the card, it takes me an average of 2-3 tries to get it right. Norman would say that the affordances are not made clear enough, either by the actual card, the machine that accepts the card or by oftentimes cryptic markings on either the card or the machine. Basically, a credit card affords to be inserted in multiple ways, only one of which is desirable, which causes human error. To redesign this experience, I have two ideas. Firstly, you could taper the thickness of the card and the slot, thus reducing the number of ways the card can be inserted, or by having the magnetic strip reader on both sides of the input slot of the machine. Secondly, post a sign right above the slop that marks which side of the card should go up (for a horizontal slot; I'm assuming that vertical slots don't exist). Let's say we start the convention that the side with the name or credit card number should be facing upward, which leaves us with two possible orientations the card can be inserted into the slot. Looking back at the first option, we see that tapering the thickness reduces the possible orientations to just one, or makes it so both orientations work.

2) An affordance gives the relationship between an object and an interacting agent, according to Norman. In other words, it specifies the possibility and type of interaction between the two. A signifier, on the other hand, points to an an affordance or anti-affordance. For eg. a road affords crossing it, but a crosswalk signifies where you can cross it. For software use interfaces affordances are often perceived by users through principles or spatial analogues, proximity and grouping, and possible signifiers are labels, arrows, tooltips, etc.


Lauren Speers - 1/29/2014 10:23:32

I own a poorly designed toaster. The toaster’s design has two main issues. First, though it is possible to turn the temperature dial to a “Cancel” setting, I have yet to figure out what the “Cancel” setting does. There is a lack of feedback – the bread does not pop up out of the toaster as I would expect it to, nor does the glowing metal appear to cool down inside the toaster as the bread stays inside. As far as I can tell, the “Cancel” setting is useless, so I resort to popping the toast up by hand. I assume the toaster does have some cancelling functionality (after all, there is a “Cancel” setting), but the signifier for this affordance is ambiguous with regards to actually using the “Cancel” functionality. The toaster’s design would be much better if the “Cancel” setting caused the toast to pop up, which would provide clear feedback and would provide the functionality expected by the user because of the signifier.

Second, the direction of the dial is only indicated by a small divot in the white plastic. When the dial is directed towards the lower settings, the divot gets completely lost in shadows. I once overheard my two roommates speculating that they had broken the toaster since their bread was not toasted at all, even though they had been toasting it on a high setting for a few minutes. I walked into the kitchen only to realize they were confused about the direction of the dial (they thought it was pointing to 4 when it was really pointing to 1) because they did not see the divot hidden in the shadows. The necessary feedback provided by the dial – looking at it should tell you the temperature setting – is very easily misinterpreted. To ensure accurate feedback, the user has to bend down and look at the dial straight-on to see the divot in the shadows. This design issue would easily be fixed by filling the indicator divot with black paint that would stand out even against the shadows, providing easily-obtained and accurate feedback.

An affordance refers to a possible interaction, either perceived or not, that a user can have with a device, while a signifier suggests what possible interactions exist and what actions the user should take in such an interaction. Affordances and signifiers for physical products can rely on some senses, like touch and smell, which cannot be used in software user interfaces. Furthermore, signifiers for software user interfaces can be hidden when they are not immediately necessary, for instance the names of all the fonts in Microsoft Word are only displayed if the user is searching through font drop-down menu. It is harder, though not impossible, to temporarily hide signifiers on physical devices.


Albert Luo - 1/29/2014 10:32:51

One example of bad design would be the plastic utensils found on some campus restaurants, or other places as well. They all have the exact same handle, and are all usually put in buckets by type (fork, spoon, or knife), with the handle sticking out. This creates a design problem because there is no easy way for customers to tell which bucket contains each type of utensil. Sometimes there are labels on the buckets, but a lot of times the labels are either missing, damaged, or the text is so small that it seems faster to just venture a guess and take a utensil instead of looking for the label. The design doesn't have good signifiers, to tell the customer which utensil they need to take. If I were to redesign it, I would small logos on the back of the handle that would make the utensil easily identifiable. That way, whether there are labels or not, the customers can always identify what type a utensil is just based the handle.

An affordance is the relationship between a person and an object, or what a person can do to the object. A signifier is what indicates an affordance, or how a person knows what ways to interact with an object. These two differ for physical products and software user interfaces in that physical products often have very different kinds of affordances than software user interfaces. What I mean is that for physical products, we have affordances that often involve more forceful motion such as pushing, grabbing, twisting, pulling, etc, but for software, it's small finger gestures or rotating something or clicking something with a mouse. Because of this difference, the signifiers for physical products must be similar to other products the user has used before, so that the user can quickly and intuitively understand the affordances. On the other hand, I think software signifiers are pretty much the same - buttons and arrows that signify clicking swiping or tapping.


Gavin Chu - 1/29/2014 12:28:24

1) I have this TV at home that came with a very unusual remote control. It has a pretty simple design, but the buttons are placed in a unfamiliar way. I believe most remotes have the channel switch on the left side and the volume switch on the right. At least most name brands like Sony and Samsung follow this convention. For some reason the company that manufactured my remote decided to switch the two, so sometimes when I try to adjust the the volume, I would accidentally change the channel. I dont use that TV very often, so the control hasn’t stick to my head and I would still make the same mistake every now and then. The designers for that remote fail to consider what the user already know. Switching the order of the channel button and volume button only confuses new users.

Another thing about remotes in general is that they contain so many buttons. Remote controls now days are fairly complicated with all the new TV features, especially with the introduction of smart TV. I would bet that many users have never pressed on half of those buttons before because many people just wanted to simply relax and watch tv rather than learning how to use every feature that came with the TV. TV designs do offer many capabilities, but it has low discoverability. There’s usually no need to press on a new button unless you are looking for some new feature. It takes a lot of curiosity for someone to explore fully what their remote is capable of. Also a problem with many buttons is that it is easy to accidentally press the wrong button, which would lead to unexpected and unwanted behaviors.

My suggestion for redesign of remote controls is to keep it clean and simple. First off the buttons need to be arranged in a way that is familiar to most people. A user shouldn’t have to spend more than a couple of seconds figuring out how to turn the power on, or simple task like changing the channel or volume. Designers need to follow conventions to avoid confusion. Second of all, designers really need to figure out what the user actually needs, like the essential features of a TV. Many buttons can be removed to reduce clutterness. Instead of having the control straight from the remote, have a button that opens a menu on the screen and make adjustments there with arrows, which could also be combined with the num pad buttons to further reduce the number of buttons on a remote.

2) An affordance is the relation an object has with its agent, for example a button is pressible. A signifier specify how to take advantage of an affordance. Using the same button example, the power symbol or volume symbol on top of a button is a signifier because the symbol tells the user what will happen if they press that button.

Affordances and signifiers are much more limited for software user interfaces than for physical products because software user interfaces are limited to 2D interactions. Anything visual can only be displayed on a flat screen whereas physical products are 3D and can have infinite ways to interact with them. Software devices like phones could have affordances like voice recognition or motion sensor, but they are not malleable like rubber, which can be squeezed or twisted. Signifiers for software user interfaces are also limited because again we are in a 2D world. The bookmark as a signifier that Norman used as an example would not apply directly to softwares because ebooks have no physical depth. However, software designers have found an alternative by displaying scrollbars. Clearly there are ways to simulate affordance and signifiers in software user interfaces; designers just have to be more creative about translating 3D objects onto a 2D device.


Zhiyuan Xu - 1/29/2014 12:46:33

An example of a physical device with terrible design is the door lock located in my apartment. The device looks like a door hinge, but works in a manner where it clamps the door shut as it swings against the door, thereby locking the door. To unlock it, one must pull up on the metal plate to un-clamp the door. Although it shares many of the same affordances with other door locks, there are no signifiers that highlight these affordances. Mainly, it is difficult to unlock the door because it is not very clear what should be done to the lock to unclamp the door. To fix this, it may be wise to include an indicator to lift upwards before swinging the hinge (ie, a small handle designed to look similar to a door knob).

An affordance is a function that the current product has. It may not always be visible, but if the product can be used in a certain way, then it is an affordance. However, a signifier is a sign or signal that highlights an affordance of an object. This may be unintentional or intentional. For example, an external signifier can be a sign that lets the user know how to use a product, which is intentional. It may also be unintentional, such as when an object has a glaringly obvious affordance that is not the same as the designer's original purpose for the object (ie. Norman talks about the safety feature installed next to a flight of stairs being used as an area to discard waste).


Matthew Deng - 1/29/2014 12:55:16

The first device that I thought of that I believe has a bad design is a single dial combination lock. The lock has a ring of numbers, and its user must enter a series of numbers by rotating the dial three times clockwise to the first number, twice counterclockwise to the second number, and once again clockwise to the third number. This design affords great security, but maybe even too much. Many times, its user does not know exactly how many more times they must rotate it until going on to the next number. This means that is lacking in a signifier to indicate when to move on to the next number. Furthermore, the design has a strange mapping pattern as well, as its first number requires 3 clockwise circles, then 2 counterclockwise, and then 1 clockwise. Its feedback ability, however, involves the lock opening, which is great because it is serving its purpose. Unfortunately, these bad design qualities are very important to keep the lock secure and therefore serve its purpose. If I could redesign the lock, I would try to add an indicator that could show which direction to spin and how many times.

An affordance is an action or purpose that a product provides. A signifier indicates when actions or purposes that a product is capable of can be utilized. Therefore, a signifier indicates affordance. I believe that in physical products, there are many more accidental signifiers to show affordance. For example, the bookmark showing how much of a book is left or the can showing a trash heap. On the other hand, UI seem to mostly be deliberate. For example, most actions that can be taken are clearly labelled.


Daphne Hsu - 1/29/2014 13:05:25

1) A household appliance that I've had a problem with is my coffeemaker. To make a cup of coffee, you pour in coffee beans at the top of the device. Then, the machine will grind up some beans, and as some beans go through the grinder, the other beans will fall down and lay on top of the grinder. The problem I had with this was that even if there were still a lot of beans left on top of the grinder, they would somehow not be ground unless you moved them around and had them sink into the grinder. This was hard for me to figure out at first because there was no directions whatsoever for me to figure out how to fix it, so I had to just play around with things. This violated interaction design, and the ability to understand how to use the product without trial and error. A really simple way to fix this would just be to place a sticker somewhere near the grinder that tells the user to move the beans around if the coffee maker doesn't seem to be working.

2) An "affordance" is the relationship between an object and an agent and the ability of the agent to use the object. A "signifier" is used to determine what and where an action between the object and the agent should take place. For physical products, affordances and signifiers will be able to tell you what you can do with the product and how you can use it. For software user interfaces, I think that signifiers are more important that affordances, because the interface should be able to tell the user how to use it correctly.


Tristan Jones - 1/29/2014 13:07:43

1) For an example of a everyday physical device with bad design, I really have to nominate public toilets. It's not that they are unintuitive to use, it's just that if you need to go #2 there's a good chance you're going to have a bad experience. Before you walk into a toilet stall, there's no good way to tell if the toilet is clean or messy. If it's really messy, then you just hold your discomfort and walk away and try to find another. That's bad UX. Furthermore, even if you look at a toilet and it looks decently clean, how do you know someone peed all over the toilet seat and wiped it up with a tissue? There's no way to see all the residue that's left on the toilet seat. There's no feedback to see if a toilet is really clean. Also, what do you do when there's no toilet paper left or the place doesn't have the seat covers? You're kinda screwed if you started #2 and there's no tp left. There are also people who seem to really enjoy messing up public bathrooms. If one person can mess up the system for everyone, well, that probably isn't a very good system then right? I'm not sure what Norman would complain about this bc it's partly a people issue and not a design issue, but it's kind of an externality issue of "what happens when people use my product and don't use it properly. does that ruin it for everyone else?". I guess that's something designers have to think about. Maybe Norman would write something about how there's little feedback on if a toilet seat is clean and people have a conceptual model of "well i've seen quite a few dirty toilets before so how do i know this one is clean". To fix this I guess you could go the technology approach and use those fancy water jets and blow driers instead of tissues and use automatic plastic seat coverings. I think the japanese are doing this but hey this adds cost to the toilet and it's another change that people have to figure out. I'm not sure how to solve the problem of people being assholes but that's human nature and hard to control I guess someone else can figure that out.

2) Affordance: the set of all interactions between a person and the object they are using Signifiers: a set of hints that guide people to figure out a preferred affordance They're different for physical products because in real life you can see (in 3D), touch, hear, taste, eat, and smell physical things. For computer software, most of it is seeing (in 2D), hearing (sound only comes from one place: the speakers), and if you're on a mobile device touching (but touching a glass screen is an incomplete form of touching). Basically, with software, your set of ways to interact with the user are more limited than with physical devices. This makes it harder to communicate signifiers that people might figure out intuitively when using a physical device.



Cory McDowell - 1/29/2014 13:11:27

1) Many sinks are designed poorly and violate Norman’s design principles. For example, the kitchen sink in my apartment has two handles: one for hot water, and one for cold water. If one does not want the hottest water or the coldest water our system provides, one must use some allotment of hot water and cold water. However, I always find myself arbitrarily guessing how much hot water and cold water to use in order to receive my desired temperature. This constitutes poor interaction design. I do not intuitively know how to get the desired temperature I want, and I end up guessing until I achieve it. Consequently, this results in a lesser-quality emotional experience, meaning we have a poor experience design as well. I would redesign sinks with a hot and cold handle to instead have one single handle with a temperature gauge on it. Then, I could always easily turn the one sink valve to the desired temperature quickly and easily. This would be much better interactive and experience design.


2) Signifiers are concrete signals of what is possible and how it can be done, where affordances are implied capabilities or preventions that come as a result of what is possible and how it can be done. Signifiers, in physical and software products, are quite similar, as they state what can be done. While the gestures of opening a door and scrolling through a webpage are very different, the signifiers of a “push” sign and a scroll bar both concretely describe how to use the door or scroll functionality. Signifiers in physical products are pretty obvious to see. For example, a window affords visibility while preventing transportation. Software affordances are more subtle. For example, a menu bar may offer less-options in order to afford a quicker user experience. Affording simplicity-of-use is much more subtle, but a necessary software affordance.


Brian Yin - 1/29/2014 13:17:31

One device that I believe has bad design was our laundry dryer at home. The dryer had a knob to set the time. However, the time indications were fairly haphazard and not in constant increments, which led to unclear times being set. Moreover, the knob which set the time could be turned in the same direction indefinitely, which made it unclear if it would run for an additional hour per rotation (like a kitchen timer) or if it would run for as long as the value the knob pointed to. Additionally, there were no signifiers as to how to start the machine such as a 'Start' button. Apparently, to start the dryer, you first had to lift the time knob, move it to the appropriate time, and then push it down. There was no indication of this on the physical device, which made people who don't use the device often confused.

Ways I would redesign it would be to remove some of the ambiguity in the timing. This could be done via digital time display which displays the amount of time the laundry is set. Additionally, I would have made the timer and the starting of the device separate. By having a 'Start' button distinct from all the other features, it provides a clear signifier of how to start the machine. This helps remove some confusion caused by obscure procedures required to activate the machine.

Affordance is a relationship between an object and some actor which defines what actions are possible on this object for that actor whereas a signifier is something which reveals what these actions are. The affordance of software user interfaces is a lot more limited than physical devices because users generally must interact with them via computers, mobile phones, or tablets. As a result, input will generally be limited to clicks, keyboard inputs, etc. Moreover, because software user interfaces are designed to do a specific thing and inputs are restricted, there are less affordances between the user and the software interface compared to the user and some physical device. For example, a web page that simply displays latest stock information cannot be repurposed by a user. However, a physical object like a table can be repurposed to many things (such as a surface for work, surface for eating, barricade, etc.)


Aayush Dawra - 1/29/2014 13:45:21

An example of a physical device with bad design that I've had to use is definitely the stereo in my car. Firstly, the main dial for volume lacks a label and is used to adjust the radio frequency as opposed to the volume, as you'd normally expect. I think this is an example of a perceived Affordance gone wrong and also a lack of a Signifier. Another issue I have with the radio is that the labels on most of the buttons are marked '1', '2', '3' and so on. It took me a while, and several hit and trial attempts, in order to figure out that those buttons were to store specific radio stations so my favorite stations are accessible on the fly. In this case, the signifier exists but there is clearly a Mapping issue, in Norman's terminology. The way I would fix these problems is by labeling the main circular dial to say 'Adjust Frequency', so there is no ambiguity in its use. As for the poorly labelled buttons, I would use the stereo display to say 'Hold this button for 3 sec to save a radio station' and once 3 seconds pass, it should say 'Station saved!' thereby removing any confusion regarding the utility of the buttons.

Affordances are the possible interactions between people and the environment. Some affordances are perceivable, others are not. These perceived affordances often act as signifiers, but they can be ambiguous for instance the use of the color red to signify danger/do not enter can depend largely on the person's experience. Signifiers, on the other hand, signal things, in particular what actions are possible and how they should be done. Contrary to certain affordances, all signifiers are perceivable, else they fail to function. The idea of an affordance and a signifier for physical objects and software has a very closely knit relationship, and are in fact very similar to each other, in my opininon. Consider for instance the example Norman cites in the chapter, where he says that an embossed plate on the door immediately suggests that the plate is a door handle and needs to be pushed in order to open the door, and thus is an effective affordance. A very similar device used in software is the button, where the button is slightly embossed on a page so as to suggest to the user that it can be pushed in order to open a link. As far as signifiers go, an example that demonstrates that physical objects and software are closely aligned to each other as far as design principles go is the use of the 'Exit' label. When we go to a movie theater to watch a movie, the doors are labelled as 'Entry' and 'Exit' in order to create a distinction between the two. A similar 'Exit' label is used in computer games where the main menu nearly always contains an option to 'Exit' the game. Both these examples highlight the close parallels that exist in the affordances and signifiers between software and physical objects.


Dalton Stout - 1/29/2014 13:51:22

1) Similar to the 'Sink that would not drain' example from the text, I have to deal with a frustrating design choice for one of the appliances in my bathroom. Like most bathrooms in a small apartment, it has a bathtub that doubles as a shower, depending on which nozzle the water is flowing through. My first time trying to take a shower in my apartment was an odd challenge. When I turned knob for hot water, water began flowing out of the bath nozzle. There is only one control switch under the water knob so I flipped it (assuming it would cause water to come out of the shower nozzle). This didn't happen. Instead, it causes the drain to stop and the bath begins to fill. I flip the switch back and begin fumbling around with the shower nozzle instead. I turn all the knobs that it has, I even almost break it by unscrewing it too much. I knew the solution was simple and I was trying not to be dense, but for the life of me I could not change it to the shower setting. I learned later that day from my roommate that if you bend down and get eye level with bath nozzle, you can see a thin silver ring under the spigot that you have to tug slightly while the water is turned on, causing a bunch of cold shower to fall on your head. I think it is clear here that my shower design has failed with it's signifiers. The affordances are pretty straight forward. I knew I could take a shower or a bath because I could see both nozzles. The problem came from me trying to access the affordance and being unable to. My redesign of this product would be simple and straightforward. Much confusion could be avoided if we add a signifier to the aid with the 'Bath to shower' affordance. To do this, I would add a small amount of text to the bath nozzle, that said something like "Pull Down For Shower" or perhaps a shower icon with a downward arrow.

2) An affordance is a service or ability that is offered from a specific product to a specific user. A product's affordances are not static, but rather defined by which user is using them. Affordances are like a relationship between a user and a product that defines how the product can be used. A signifier is related to an affordance in that signifiers make user aware of affordances that they may not have known about or how to use. A signifier need not be intentionally placed by the designers. If an ability on your product is not clear, a good signifier teaches the user how to use it in a straightforward way. Affordances and signifiers differ drastically between physical products and software interfaces. For one, physical products cannot be updated remotely. If there is a vague or ineffective signifier before a product goes to ship, it cannot be fixed. Software however relies on this update feature to fix bugs. Often times apps are released early on purpose (beta) so that the developers can learn about affordance/signifier problems straight from the users.


Diana Lu - 1/29/2014 14:08:25

An example of an everyday household item that has poor design is lamps. Much like the issues that Norman brings up with doors and how users can have difficulty knowing whether the door is meant to be pushed or pulled, lamps can often have the same problems. There isn't any consistency with light switches; some are buttons within the lampshade, some lie at the base of the lamp, and some are even attached to the cord. Without having the familiarity or prior knowledge, it can be difficult to find the switch and how to operate it. This device violates Norman's design principles that specifies that signifiers should indicate where an action takes place. Lamps often lack the proper signifiers to let the user know how to operate the switch. In terms of redesigning the device, I think that the most important thing would to be emphasize consistency of the switch and placement, where it can be easily found without needing to search extensively. The primary difference between an "affordance" and a "signifier" is that an affordance defines the relationship between a user and an object, or the service that object can "afford" the user, or vice versa. As Norman describes in his article, an affordance generally refers to the service an object provides a user, while a signifier is defined as an indication of how that device should be used. Affordances can be fairly similar across physical products and software user interfaces, while signifiers tend to me more differentiated. A prime example of this can be found in Norman's words when he describes a bookmark as being not only a signifier for what page a reader is on, but also as a sense for how far into a book someone is. While an ebook can utilize the same sort of signifier, a bookmark, the same indication for how much of the book is left cannot be determined.


Jeffrey Butterfield - 1/29/2014 14:12:58

Q1) At the start of my sophomore year, I moved out of the dorms and into an apartment. Remembering how space-efficient the loft beds were in Unit 2, I bought a metal loft bed frame off Target.com for $199. Unfortunately, both the bed and the instruction manual were very poorly designed.

First, the instruction manual listed over 100 steps for assembling the frame. Norman says that for complex products the design characteristic “understanding” may only be possible with a guide, but the large number of instructions (relative to similar products from, say, Ikea) indicate that the design of assembly procedure itself was more complex than necessary. Second, while the design manual included signifiers in the illustrations as to which parts were which (each part was assigned a letter), the physical pieces of the loft bed did not have corresponding stickers on them, making the signifier in the manual useless and the discrimination between similar pieces frustrating. To fix these problems, I would have reduced the number of assembly instructions to less than 100 steps and placed letter stickers on each of the parts, so that each of the parts’ affordances would be clearer to the assembler.

The actual frame’s poor design showed in its ladder. The ladder was, quite simply, painful to use. The rungs of the ladder used to ascend to the bed were skinny and cylindrical, rather than wide and flat like those on the ladder of my Unit 2 bed. I was reminded of the difference every time I climbed into bed at night. This infuriating design flaw illustrates Norman’s paradox of technology: the loft bed technology should allow me to have the same experience of sleeping in a bed with the added benefit of fitting a desk underneath. However, the added complexity of getting to the top of the loft bed introduced discomfort because of a poor design. I would use the Unit 2 design instead (use flat ladder rungs).

Q2) As defined by Norman, affordances are relationships between an object and an agent using the object. The affordance, as an abstract concept, describes what the object enables, allows, or facilitates for the agent. Signifiers, though related, are subtly different. Signifiers are perceivable clues or indicators that inform the user about an affordance within the design. Signifiers can be directly perceived (see, hear, etc.), unlike affordances.

With physical objects, signifiers can be physical signs or even physical qualities of an object (like Norman’s door hinges signifying which side of the door swings), whereas signifiers in software usually are graphic cues on a screen (like arrow or animations) that notify the user about the software’s capabilities. As for affordances, different physical parts of an object might afford different things (a chair’s legs afford support and its cushion affords comfort). Even though the different parts of a software UI are virtual and all part of the software’s conceptual model, individual UI components can also be said to afford certain results, just like objects.


Liliana (Yuki) Chavez - 1/29/2014 14:34:36

1) There is this mug in my shared house that has the shape of mugs that are microwavable/heat resistant. This mug has a handle, which is supposedly useful for holding the cup when it contains hot things so as not to burn oneself. It also has that typical cozy, rounded shape that is characteristic of a lot of mugs that are heat resistant. However, while this particular mug is made out of a material the resembles the material that regular heat resistant mugs are made of, the material itself heats up relatively fast, and does not actually resist heat in any way, so when you use it to put hot stuff in, you burn yourself. The mug itself has an anti-affordance of not being able to hold hot things. There is no clear signifier that it is unable to do that (as its design mimics mugs that have an affordance to do so). To fix this problem, you could either have a clear signifier at the bottom of the mug that it is not microwavable (in order to maintain the same exact design of the mug) or one could actually re-design the mug to not have a handle and rounded shape, and/or be made out of different material that doesn't seem to afford heat (glass,metal,plastic).

2) An affordance is something that allows an action to take place. A signifier is something that tells or hints to the user that that action is possible course of action. In physical products, affordances and signifiers can be the same thing and are usually the shape or the structure. In the article, Norman discusses how the shape of the handles of a pair of scissors both are an affordance for putting your fingers in, and are signifiers for putting your fingers in since they have holes that are the size of your fingers. In software user interfaces, the affordance and signifiers usually can not be the same thing. There is typically more of a separation between the too since both aspects have to be specifically coded. For example, in an email app, there is the affordance to send an email, but without the clearly labeled 'send' button acting as signifier, you do not know that you can send this block of text to someone.


Namkyu Chang - 1/29/2014 14:42:15

1. Good designs require discoverability (possible to figure out what actions are possible and where and how to perform them) and understanding (How is the product supposed to be used?) A physical device that I had to use with bad design was my car’s radio. The 2004 Acura TSX has a radio that looks like this (49246637.jpg). Norman’s principle of “discoverability” is well preserved because it’s easy to see what actions are possible (AM/FM button for the radio, seek/skip button with arrows to change channels, volume knob with “VOL” to change the volume, etc.).

However, the “understanding” principle was lost when trying to change the radio’s time. I attempted to change the hour on the radio for daylight savings time, but could not figure out how. (i.e. How is the radio supposed to be used to change the time?) You will notice the little “H”, “M”, and “D” physically written on top of the “4”, “5” and “6” buttons, which stood for “hour”, “minute” and “day”. Even then, how are we supposed to use these buttons? I tried holding the buttons, which changed the quick-select radio stations. I tried repeatedly pushing the buttons, which just changed the radio station to whatever was saved to “4”, “5” and “6”. Only after reading the manual was I able to figure it out.

In order to solve the problem, I would have a separate “Time” button which would make the display’s to start blinking, adjust with the “H”, “M”, and “D” as necessary, and push the “Time” button again to set it. However, a problem I see with my new design is that it would take up a lot of physical space for a feature that would rarely be used.

On a side note, on my way to work on this reading response I ran into a design very related to this lecture’s reading. I visited a library I haven’t been to since 2 years ago (Music Library) to finish the reading response, when I came across the main entrance which looks like this. Reading_response_2_door1.jpg Reading response 2 door1.jpg What would you do to open this door? Push on the left side? Pull on the right? Slide it to left or right?

When I saw the door, I didn’t want to spend too much time thinking about such a trivial issue and did what any normal person would do. I saw the horizontal metal bar and assumed “push”, but instead of figuring out which side to push, I pushed both. My theory was that one side would start opening, and I focus my efforts on that side until I could enter. However, the door refused to budge. To make matters worse, there was another person that approached the other side of the door trying to leave the library. I started panicking, and tried to slide the door, but to no avail.

At this point, the girl on the other side looked at me as if I was a little challenged. However, as my CS education developed my logical thinking, I quickly processed that a) there were people on the other side of the door so this door must be openable, b) I tried pushing and the door didn’t move, and c) I tried sliding the door and it didn’t move. Through logical deduction, I was able to gather that I should try pulling the door (again on both sides to avoid any further embarrassment). Sure enough, I was able to open the door, let the girl pass first, and enter the building.

Reading_response_2_door2.jpg Reading response 2 door2.jpg

The part that tripped me up the most about this door was that there was a horizontal bar that popped out, which signaled to me that I should either pull or slide. I didn’t include this as my prime example of a physical device with bad example since it is identical to the author’s example.

2.Norman defines affordance as the relationship between properties of an object and capabilities of the agent that determines how the object could be used. In addition, he defines signifier as some means of signaling the presence of an affordance or anti-affordance. The difference between the two is that one describes what action is possible while the other describes where the action should take place. In my above example of trying to change the radio’s time, the “H”, “M”, and “D” were signifiers. Affordances and signifiers differ for physical products and software UIs in that the former could have additional accidental affordances/signifiers whereas software UI requires the designer to deliberately provide a clue. Take, for example, the author’s example about the bookmark. Its main purpose is to allow the reader to go back to the spot in the book where they left off, but also acts as an accidental signifier. If an e-book designer wanted a bookmark feature, it could do it in many ways. However, to only have the main feature (allowing the user to go back to the exact spot in the book), it could simply be a button that jumps to that page. This leaves out the accidental signifier of showing the reading how much of the book is left. To have this additional “accidental” signifier, the designer could deliberately tell them how many pages are left, show which page the reader is on out of total pages, etc. However, these would all have to be deliberate.


Anthony Sutardja - 1/29/2014 15:13:11

I have a horrible time with the circular valve that controls the flow of water through water hoses. I know, some people say "right tighty, lefty loosy" as a way to remember which way is open and which way is close, but singing this phrase to myself is never the first thing that I do when opening/closing the valve to a water hose. I'll try to tug in both directions and it seems I never get it on the first try. If the water valves adhered to Norman's "signifier" principle, then I would have a better idea which way I should turn the valve. Some water valves already solve this problem by adding little engravings to indicate opening and closing the valve, which is exactly what I would have done. However, most of them still don't and it's frustrating!

"Affordances determine which actions are possible. Signifiers communicate where the actions should take place." Although both affordances and signifiers involve actions, identifying affordances is like a different level of viewing a product. The primary goal of identifying affordances is to help the designer identify actions, whereas the primary goal of signifiers is to help the user perform an action.

Affordances and signifiers in the physical realm are more geared towards manipulating the physical environment. The feedback for many of these interactions are physical as well. For example, a switch turns lights on, a blade cuts things, and door opens passage ways. However, affordances and signifiers in the software user interface realm encapsulate both the physical and the virtual environment. This means a lot of the feedback we receive are visual.

Since so much of the environment is visual, software user interfaces also introduces new kinds of affordances. Sure, we have the computer mouse and keyboard that enables us to perform actions on the computer, but we also have virtual affordances now as well. We recognize that the pointer on the screen is an extension of our hand and it has abilities to interact with the system further. There are virtual buttons that allow us to perform actions, actions that need signifiers to tell us what they do.


Emon Motamedi - 1/29/2014 15:13:20

1) An example of a physical device with bad design that I have had to use is a shower faucet, and more specifically, shower faucets that only turn counterclockwise. These faucets require one to turn all the way through the cold half (right side) of the faucet to get to the hot half. However, it is difficult to tell where the hottest and coldest temperatures are and hence where to position the faucet. Imagining a clock, the cold side is represented by 12 through 6 and the hot side by 6 through 12. As the faucet begins at 6 and moves counterclockwise, does the temperature start off coldest and begin warming up as the faucet nears 12, making 12 the middle temperature and 7 the hottest temperature? Or does the faucet get colder as it nears 12 (which in my opinion is the more intuitive understanding as one is turning the faucet higher on the cold side)? In this situation, once 12 is passed, does the water suddenly become lukewarm as the faucet is moving counterclockwise and does it continue get hotter or is it at its hottest point after passing 12 and continues getting less warm as it moves counterclockwise.

This shower faucet violates a number of Norman principles. One is its system of feedback. The shower does offer feedback, but this feedback comes from one feeling the water temperature, which can oftentimes be too late and extremely painful if the water is scolding hot or ice cold. Another violation comes from mapping. The system does not offer an intuitive approach to mapping faucet position to water temperature. Finally, related to mapping, is a lack of signifiers to signal where the hottest and coldest temperatures are.

To redesign this problem, I would start by placing four signals at 7, 11, 1, and 5. These signals would either say "warm", "hottest", "cool", or "coldest" depending on the faucet structure, and would clearly signal to the user where to position the faucet dependent on his needs. I would further attempt to eliminate this issue altogether by adding a marker to faucet which would allow the user to mark his ideal faucet position. This way, rather than constantly adjusting the temperature every time one enters the shower, the user can just move the faucet to the market which reflects his perfect temperature.


2) An affordance is the relationship between what an object is able to do and how capable a specific user is at utilizing those properties of the object. Hence, affordances designate what actions are possible for a specific user of an object. However, sometimes these actions are not readily evident to a user. This is where signifiers come into play. Signifiers are signals that communicate to a user what possible actions he can take on an object and where/how to take these actions. Thus, if an object has a very obvious affordance, this affordance also serves as a signifier because it readily and obviously indicates what behaviors are possible.

Affordances and signifiers differ for physical products and software user interfaces in that it is much easier to create a perceivable affordance for a physical product than it is for a software user interface. Physical objects are 3 dimensional, come in all shapes, sizes, and colors, and allow for a user to touch and investigate the object. Software user interfaces on the other hand are mostly restricted to a screen and hence do not offer the user the same opportunity for investigation. As a result, affordances can be more readily noticed and created for a physical object than a software user interface. This means that software user interfaces require even more signifiers to ensure that a user understands the various behaviors that are possible.


Prashan Dharmasena - 1/29/2014 15:35:58

1) Certain microwaves are a very good example. Many new microwaves come with many different settings that aren't obvious to a new user, especially without a manual. For example, in order to maximize the functionality while keeping the number of buttons down, you often have to push buttons multiple times, even if it just says "Start" or "Cook." This breaks Norman's principles of discoverability and understanding as it prevents the user from easily using even the most basic functionality. In addition, by having the function of the buttons being so vague, they break the principle of mapping.

2) An affordance is essentially an interaction that the designed object allows for, while a signifier is a piece of the designed object that allows for a user to easily identify affordances (or anti-affordances) and how to interact with them. Norman uses the example of a window. It has an affordance of letting light through it, but an anti-affordance of not letting physical objects through it. Thus, many people injure themselves trying to walk through glass doors. A signifier for this situation could be a window pane. It allows the user to see that the glass is indeed a solid object, while still letting a majority of the light in through the window. Affordances tend to be more straightforward for physical devices than in software. Physical objects only have a limited amount of interaction. As such, signifiers on physical objects are also much more subtle, since with software, you can render anything you want on the screen.


Aman Sufi - 1/29/2014 15:39:49

My former shower head at home always puzzled me due to its poor design. It seems to have an inner ring of smaller holes and an outer ring of larger holes, but water only ever came out of the larger ring. Additionally, the front of the showerhead seemed to have three grips around its perimeter (which itself had was labeled with the word ‘Adjust’, with arrows pointing from it in both directions) which seemed to indicate that twisting it would cause it to change the rings the water came out of or at least change the water flow somehow. Instead it didn’t do anything except spin the showerhead around and maybe loosen it from the pipe a little, which didn’t really change the shower experience at all. After asking the manager, it turned out that there was a small adjustable ring on the back of the showerhead to adjust the flow, but it was only visible if you were looking for it specifically. In the end, it turned out the shower head was defective anyways as turning that ring caused the water flow to shut off instead of switching the ring in use, so I got it replaced due to that and a poor flow level.

The showerhead clearly doesn’t suggest a very useful or accurate system image to the user according to Norman’s terminology. Although the designer may have thought that by having grips along the showerhead’s side would cause the user to perceive the word ‘Adjust’ to mean that the shower grip could be swiveled around on and rotated on its axis, it seems clearly misleading to me as a user because seeing the unused holes on the showerhead and only one thing labeled as an adjustment on the showerhead lead me to think it could be used to adjust the flow and use of the additional holes. This resulted in the user’s conceptual model being markedly different from the designer’s conceptual model. In a more general sense, this issue was caused by the manufacturer failing to include a useful signifier to access the adjustable functionality of the showerhead and a misleading perceived affordance in the way the front of the showerhead could be adjusted.

As a side note, I also had some design complaints about the printing station at Moffitt and started writing about it until I remembered that the was basically poor software design and not really a physical issue.

For the second part of the response, an affordance is a possible interaction between the user and the product, which may or may not be obvious, or perceivable, to the user, whereas a signifier is a perceivable indicator which lets the user know of the presence of an affordance. A perceived affordance often may act as a signifier for itself, although it may be ambiguous.

For physical products, affordances are mostly concrete features which can be perceived physically, and can often be made intuitive through the physical design, and by extensions signifiers, of the product. The feedback provided by physical affordances is often an intrinsic nature of the affordance, such as turning a tap, where you will see water flow out immediately, although it may not always be the case, such as if you pushed the button for an elevator and it did not light up to let you know that it was coming to your floor. Comparatively, software user interfaces often rely on crucially designed feedback much more because without any meaningful feedback, one cannot really navigate a software interface and it becomes meaningless. Imagine using a computer without a monitor and typing up an essay. Even if a word processor is open, you will be unable to see whether you made any spelling errors or easily navigate the menus except by having memorized keyboard shortcuts or similarly unintuitive methods, and the essay you have just written will have no value to you as you cannot access it or navigate the interface successfully to do anything with the essay.

On the other hand, signifiers on physical products may appear naturally through perceived affordances as a nature of the appearance of the affordance. In software interfaces, however, affordances can rarely be ‘discovered’ without going through an intermediary signifier alerting the user to the presence of the affordance as most features/affordances cannot be directly accessed while in the main interface (think of the desktop of a computer or the home screen of a smartphone) and must be launched by following an appropriate signifier. In summary, signifiers tend to arise much more naturally in physical products and the effort is in making sure that the signifiers are clear to the user, whereas in software signifiers must often be implemented from the ground up to allow the user to be aware of the different affordances the interface has to offer.


Justin Chan - 1/29/2014 15:40:40

Literally everyone who has ever ridden in my car has complained about how “stupid” the manual door lock is. This is an assessment that I have begrudgingly acknowledged over the years (the other complaint everyone else has is that I’m stubborn) due to the honest fact that the manual door lock in my car works completely opposite compared to the de facto standard in cars. In most cars, unlocking a door manually involves some sort of “pulling” – whether it involves pulling a latch toward you or pulling a stick up. This I would attribute as an example of a signifier. With the latch, there is usually a little red sticker behind (key word: behind) it, which intuitively tells us that we need to pull the latch to see it (and unlock the car). With the sticks, since they are shorter in “lock mode,” we can safely assume that we should pull them in order to unlock the car. Additionally, most people will link the actions of unlocking a door and opening it. Since the latter involves pulling a latch, 99% of the time people will try a similar pulling method to achieve the former.

The “pull” method in my opinion is a rather intuitive design choice that most people should be able to grasp relatively quickly. However, in my car, you need to actually push the latch in order to reveal the red sticker and unlock the car. You can then imagine how frustrated people are when they ride in my car the first time and try to get out. They first have to push the latch to unlock the door and then pull another one to open it. It creates a sense of disjointedness that I don’t think Honda intends for its customers. This problem is easily solved by re-designing the door lock back to how it is in every other car – pull to unlock, push to lock. Perhaps Honda wanted to try something new when they decided to make the change in the first place, but as they say, if it ain’t broke, don’t fix it.

An affordance tells us what actions are possible with the given design, while a signifier tells us where and how those actions should take place. In a nutshell, an affordance is like a general principle (“be nice to others”), while a signifier tries to be a little bit more specific (“hold the door when people are behind you”). For physical products, affordances and signifiers have a different dimension because we are literally dealing with more dimensions. Physical products are 3-D objects, so we are afforded a lot more possible actions that we can take. Given a larger slate of actions, it’s important for signifiers to be good because there is a larger chance for people to get it wrong. With software interfaces, there are only so many things you can do on a screen – click, drag, scroll, pinch, pull, etc. Given the lesser slate of actions possible, signifiers are “easier” to implement because 1) you have less actions to get confused about and 2) chances are that someone else has already done the work for you. (ex. clicking X will close the window, right-click gives you options, etc).


Steven Wu - 1/29/2014 15:56:44

I thought the add-on PS3 Keypad was one of the worst designs for video game peripherals. It had a keypad clipped at the top of the regular video game control buttons and joysticks and it felt discomforting overlapping my palms on the main controller to reach the physical keypad. Here's the device attached on a PS3 controller. Since there are people with large hards and large finger tips, then there is no affordance for these users. Another cruical violation was the affordance it made with the size of each individual button. If you examine the buttons without the keypad, you may notice those inputs are large and very few of them are needed for a gamer to fully emerse themselves into a game. With the add-on keypad, the product designers compromised the familiarity with the computer keyboard with the limited real estate on the PS3 controller. In addition, the three signifers under the SONY label don't convey anything useful to the hardcore gamer. If I think about it long enough, I can gather that the center one is the selection (click) button, that the left one is the menu key, and that the right most is the emoticon key. But that took me a while to figure out, especially since menus are more familiar with computer users as being right-click and the selection button as being the left-click. A quick fix to those three appalling buttons would be to reduce it down to two buttons as if it were two buttons on a mouse. But if I were to re-design it to solve the problem, I wouldn't even bother creating a physical keypad for the PS3 controller. Even comparing it to the competitor, the XBOX 360 keypad, by relocating the keypad below the main controller buttons, I find it still intrusive to where I place my hands when I am typing. But the XBOX layout would be more familiar with mobile phone users who use a slide-out physical keyboard. But ultimately, I think it would have been more beneficial to the end user by allowing them to pair the game console with a bluetooth enabled keyboard. The features and the form factor of the full sized keyboard is generally still used today. You have people using a full sized keyboard to surf the web on their television sets. Keyboards have generally remained the same size for decades for desktops. This leads me to believe that there shouldn't be a need to compromise something that comfortably fits into your hand (such as a video game controller) with a terribly scaled keypad. It just doesn't feel right, literally and figuratively. As for the positives, the design of the product does a well job at following the principles. There, the key layout is mapped from the keys on a keyboard and there is onscreen feedback when a key is pressed.

2. Affordances are the compromises that a product has that accommdates some users. Whereas signifiers are symbolic representations of what a user is to do with the product and the placement of where an interaction should be performed. For physical products, affordances and signifiers can be invisible. But it is crucial for a software user interface, to provide representative visuals to convey the message of the product. Software products are usually representations of physical products, so there is this mapping loop a user draws back on to remember how to interact with a 'virtual' interface like software user product with the familiarity it would hold with something that already exists in the tangible reality. Physical products however bounce off a preexisting products and into newer directions, allowing for a user to eventually to unravel the functionality of the physical product with a number of trial and errors.

Sol Han - 1/29/2014 15:56:52

One experience I've had interacting with a badly designed physical device was when I was using a metal pot to boil water. The pot's handles were also metal and thus became very hot, meaning I had to use oven mitts to lift and carry the pot. This design demonstrates both an affordance (transportation of the pot) and an anti-affordance (prevention of carrying the pot). As with Norman's sink stopper example, these contradictory relationships reflect poor design. Furthermore, if the user of the pot were to be unaware that metal conducts heat (or if they were to assume that the handles were safe to grab, based on their previous experiences with other cooking pots), they might accidentally burn themselves because my pot had no other signifiers to indicate that the handles could get very hot. I would, of course, design the handles to use some other material that doesn't conduct heat as well, so that they can be safely used while cooking. The handles could also protrude further out to help prevent me from accidentally touching the metal side of the pot.

An affordance is the relationship between the product and the user, whether perceived or invisible. A signifier, on the other hand, is a signal that indicates what to interact with and how; signifiers can sometimes be perceived affordances. Affordances and signifiers can be more limited in software UIs, though more sophisticated methods of interacting with UIs are being developed. Many modern software UIs mimic physical systems to make interacting with UI easier and more intuitive for users. For example, computer desktops mimic physical desktops. Also, swiping in mobile apps to move items mimics the physical motions we make when moving real-life objects.


Alexander Chen - 1/29/2014 16:02:50

I have always had a great deal of trouble with the garage openers at my house. Sometimes when I try to operate the remote to get the door to open or close, nothing happens. At this point, I don't know why that is. Perhaps my remote control is out of battery, but there is no feedback to let me know that a signal is being transmitted from it. Or maybe the receiver is out of range, there is a power outage, something is wrong with the motor, or there is a safety trigger preventing operation. But how would I know? The only feedback from the entire system is whether the garage door opens or closes.

In terms of affordances and signifiers, I don't believe there is much to be improved. The sole functionality of the system is to allow me, an entity with a registered remote, to control a motor from a distance and raise or lower a door. The remote control is quite simple also, it has 3 buttons, each to operate a separate garage door.

However, feedback is where the system begins to lack. First of all, I don't know if the remote is transmitting. When a button on the remote is pressed, I hear a mechanical click, but there is no other feedback to signify whether the remote is trying to send out a radio signal. Additionally, the communication between the remote and the motor's receiver is only designed to operate one way. I cannot get the status of the motor on my remote, such as whether it is in range or whether is has accepted the input and is closing or opening the door.

Without completely redesigning the system to use the internet as a means of communication, I think the best solution is to add some LEDs to the remote. It should have a grid of LED backlight words to indicate the status of each door. The words should include "in range", "closing", "opening", "error". These 12 additional LED lights should greatly increase the user's awareness of the system status and ease frustrations.

Affordance is some property that a system offers. It is functionality. Signifiers are indications that let the user know how they can interact with a system. On tangible products, usually the affordances are relatively obvious, e.g. to push on the side of the door that is not attached to a beam. Because these products need to obey the laws of physics, people have some intuition when interacting with them. On software user interfaces, programmers are not bound by the limitations of the natural world. Items can "float in air" without any apparent support, such as the restaurant application that Norman mentioned. Another example is the maps application where the map is not bound and floats in the screen as the user pinches in or out and pans around. In these cases, signifiers are more important to show the user how they can interact with the application. For example, in Apple's Map, the bottom of the map is curled up to show that there is a menu that the user can access by tapping the curl.


Peter Wysinski - 1/29/2014 16:04:00

A device with a bad design that I use on a daily basis is the hight adjustment knob on my chair. For the chair to go up, one has to push the knob in; for the chair to go down, on has to pull on the knob while sitting on the chair. I did not know that my chair was adjustable until a friend with a similar chair pointed it out. Having played with the knob countless times I’ve always assumed that it’s purpose was the disassembly of the chair, I never pulled on it in fear of breaking it. The knob violates the design principles of Human Centered Design and the Fundamental Principles of Interaction. A knob with grooves on its side had the affordance of the ‘ability to turn’; I as a user did not consider pushing/pulling on it. Upon closer inspection I discovered that the knob had the words ‘push/pull for hight adjustment’ written on it (a signifier) but since the knob was under the chair I wasn’t bothered enough to get on the floor with a flashlight just to discover the purpose of a knob that I’ve already assumed in the mind as something else. Affordance is the relationship between a physical object and an interacting agent. Simply defined it the service/function that something can offer to a specific user. This perceived functionally many be different from one user to another. Signifiers are cues that explicitly state the action something is able to perform. In physical products signifiers tell us what interactions are available, for example a push sign on a door tells us that it can be pushed. However, in software, signifiers tell us where to do something to get a result — we already know that our smartphone screen can be touched, but we do not know where to touch it to have it do what we want. Adding and arrow to indicate where to touch something would not be an affordance as the user knows that a touchscreen can be touched but rather a signifier.


Bryan Sieber - 1/29/2014 16:05:56

Clamshell packaging, the packaging that is heat-sealed to avoid tampering with the product. It is terrible. I've even hurt myself before because of it. I've had to open many packages with this type of packaging, and every time it has been a painful process. The experience of using and enjoying the product is halted by this type of packaging; as such the total experience and enjoyment is brought down. If products normally using heal-sealed clamshell packaging were to instead adopt an alternate design there is the possibility of increasing the overall enjoyment of the product. What first came to mind was the use of recycled materials, perhaps cardboards. This would not change the enjoyment, but instead--since the product is not visible through the cardboard--make the product seem more like a gift. To ensure that the products are not tampered, the use of a "Tear Here" signifier with a perforated pull tab could provide the needed security of a tamper-free product. (Perhaps a picture would help with what I mean by perforated pull tab. Here's an example with a Microsoft Points card.: new_xbox_card_3.jpg) A product that can be easily opened, but still secure, would increase the enjoyment. Also, it could reduce the harm that can be caused by the clamshell packaging. The principles of feedback and mapping are not visible with this product (the packaging). Feedback in this circumstance could be the frustration by the user and the nonexistent tear in the packaging when the user attempts to open the product with his or her bare hands. Mapping is also lost in this type of packaging. The way that it should be opened is left to the user to figure out.

Signifiers ("the where") signal where the affordances ("the what") occur. The affordances in software user interfaces may not always be as easily perceived as those that are physical products. In the example of the app in the reading, half of the functionality was lost. The product is intangible and solely on the screen, making the extra functionality lost unless it is explicitly seen with a signifier. With a tangible, physical product signifiers might not always be visible or might be accidental, but most functionality can be discovered with ease. Signifiers in software user interfaces are extremely important; they need to convey a large message without taking up much space. In a physical product the signifiers may not always be necessary if the affordance is perceived. Affordances in software user interfaces need signifiers, otherwise the functionality of the affordances is lost.


Seyedshahin Ashrafzadeh - 1/29/2014 16:09:45

I have a night lamp stand and its switch is hidden behind the head of the lamp. Also, it is very small and oddly positioned that it makes it hard to turn on the lamp. Therefore, this is a design flaw in the design of this night lamp stand. This design violates the discoverability that Norman talked about. The signifier that the designer used to show the switch is not well designed and it adds complexity and difficulty to find and pickup this signifier. If I were to re-design this night lamp stand, I would make the switch big enough and put it somewhere on the body of the stand so that it is easily visible. Also, I would put a small LED indicator near it so it could be easily visible in the dark. According to Norman, affordance is the relationship between the properties of an object with how an interacting agent can use the object. This affordance is determined from the qualities of the object and capabilities of the interacting agent. However, the component that shows these affordances is called the signifier. For example, a flat plate on a door signifies that the user has to push the door to open it. This flat plate is the signifier. In design signifiers are more important than affordances because they show how to use the design. Affordances for physical products are somewhat different from software. In a software, affordances are like being able to click, touch, drag, swipe and more. But in a physical product, the affordances have more physical dimensions to them, like being able to lift and sit on in case of a chair. In a software, signifiers are like icons and different UI widget that show an information to the user. But in a physical product, these signifiers can be from the material they are made, their shape, their roughness, and many more. So there are many differences between physical product designs and software designs.


Sijia Li - 1/29/2014 16:11:28

(1). The fridge in my apartment has a really funny/bad design. When my friends came to my place and tried to open the fridge, they usually failed to open the fridge at their first try, (Why?), because the fridge has handles on both left side and right side! So, when a person enters the kitchen, he put his hands on the handle which is closer to the entrance of the kitchen and try to open the fridge. But that side (the left side) of handles can not open the fridge! It is the right side of the handles that can be used to open the fridge. And, there is no signs on the front part of the fridge!!! I have to say that I have no idea why the fridge company installed handles on both sides of the fridge. Maybe, it could be easier to take the whole fridge door off?! This design violates the "understanding" principle. Users at first do not understand why there are handles at both sides of a fridge; they also do not know which one to use to open the fridge. I would just put handles on only one side of the fridge so that the users will be very clear about how to open the fridge. For a simple device like this fridge, external signifies (signs) are not needed, according to Norman's comments on doors (Page 15).

(2). Affordances determine what actions are possible; Signifiers communicate where the action should take place. An affordance is a relationship; whether the affordance exists depends upon the properties of both the object and the agent. Signifiers signal things, in particular how actions should be done.


Kaleong Kong - 1/29/2014 16:11:42

1. My example is the back door of MUNI which is a public transit in San Francisco. As far as I remembered, there are 2 or 3 different designs of MUNI’s back door. My example is the one that needs you to step down one step of stairs (buses usually have two steps of stairs) in order to open the door. I remembered the only signifier is a bright yellow line on the first step, but I saw a lot of people simply overlook it. I usually expect to push the door open or the driver would open the door for me, but I didn’t expect that the button is actually on the floor. Its signifier is failed. The designer thought the yellow line is enough to make us notice the ground but it isn’t enough to make us realized that that is actually a button. In order to fix this problem, I would add a better signifier on the door. I would write a big label on the door saying that “Don’t Push, Step on the Yellow Step to Open the Door!” I think this would be obvious enough for most people. 2. According to the reading p14, it states that “affordances” determine what actions are possible, and “signifiers” communicate where the action should take place. I think the “affordance” in physical object is more obvious, since physical object are more natural and everyone can interact with them without using any device as an intermediate. Moreover, physical objects are really depending on how users want to use it and same object may have different “affordance” to a person at difference time. For example, usually people use umbrella to protect them against rain or sunshine, but people can also use it as a weapon. To me, “affordance” is more relating to the users experience and how users want to use it, it can be against the design purpose of this product. The less artificial of a physical object, the less signifier you need to show how user can use the object, since user can use it anyway as they want. However, in a software user interface, you need more signifier to show the user how the software you want them to use. First, it’s completely artificial. Second, it’s only work if you are using it in the right way. Since software is completely artificial, everyone needs to learn about it before they use it. “Affordance” in software UI comes only when users have been exposed a lot to similar kind of technology and built up an expectation of what that kind of software/technology can do.


Sijia Li - 1/29/2014 16:12:02

(1). The fridge in my apartment has a really funny/bad design. When my friends came to my place and tried to open the fridge, they usually failed to open the fridge at their first try, (Why?), because the fridge has handles on both left side and right side! So, when a person enters the kitchen, he put his hands on the handle which is closer to the entrance of the kitchen and try to open the fridge. But that side (the left side) of handles can not open the fridge! It is the right side of the handles that can be used to open the fridge. And, there is no signs on the front part of the fridge!!! I have to say that I have no idea why the fridge company installed handles on both sides of the fridge. Maybe, it could be easier to take the whole fridge door off?! This design violates the "understanding" principle. Users at first do not understand why there are handles at both sides of a fridge; they also do not know which one to use to open the fridge. I would just put handles on only one side of the fridge so that the users will be very clear about how to open the fridge. For a simple device like this fridge, external signifies (signs) are not needed, according to Norman's comments on doors (Page 15).

(2). Affordances determine what actions are possible; Signifiers communicate where the action should take place. An affordance is a relationship; whether the affordance exists depends upon the properties of both the object and the agent. Signifiers signal things, in particular how actions should be done.

Thanks.


Emily Reinhold - 1/29/2014 16:30:18

1) One of the most poorly designed "everyday things" that I encounter every day is the blinds on my windows in my dated apartment. They are composed of tens of thin aluminum 1 in. x 18 in. strips, fastened together by 2 strings - one on each end. To raise the blinds, pull a string; if you want to lock the blinds at a particular height, angle the string to the left. These particular blinds violate Norman's design principle of discoverability, because they offer no signifier as to how one would lock the blinds at a desired height. If you simply pull the string at any angle, the blinds will raise. If you then let go of the string because you seem to have achieved your goal of raising the blinds to an acceptable height, the blinds will fall back down. Without having seen this mechanism before, it would take a user many trials and errors before discovering that the string needs to be angled just right to cause the blinds to stay put.

Further, they do not employ HCD well, in that they do not perform well when something goes wrong. In particular, the windows that these blinds cover have protruding handles at the center. Since the blinds are composed of many strips as I mentioned before, there are spaces between the strips. When you try to raise the blinds, a strip will get stuck on the window's handle, causing the aluminum strip to get permanently bent and preventing the user from raising their blinds as they hoped. What an awful experience! This shows poor industrial design, as the choice of material allows the blinds to get permanently damaged for a simple mistake.

Finally, these blinds do not perform their function as well as they could. I use them primarily to block out the sun, but since they have space between all the slits, sunlight slips through the cracks.

To redesign these blinds, I would make them out of opaque fabric for two reasons - to block sunlight, and to prevent permanent damage if they get bent. I would make the entire set of blinds one piece of fabric so sunlight could not slip through the cracks. I would make the raising and lowering mechanism a taught string on a rod that connects to a spool (common design for more modern blinds). This mechanism is a natural mapping - if you move the string upwards, the blinds move upwards until you stop (or the blinds are fully raised); if you pull the string downwards, the blinds move down until you stop (or the blinds are fully lowered).

2) An affordance defines what actions a user can perform on a given system, while a signifier acts as a clue informing the user where to initiate the action allowed by the affordance. In general, I think affordances for physical products have to be much more obvious in order for someone to want to buy the product. For example, no one is going to buy salt and pepper shakers without knowing that such is their purpose. In that way, physical products either need to have visible affordances, or well-planned signifiers. I think signifiers for physical products are primarily governed by 3-dimensional shape or text. The shape of the object informs the user about what actions can be performed on it, and text is used for extra clarification (ie. push, pull on a door).

There are typically a significantly larger number of features/functionalities in a software UI than in a physical product. As such, I think software affordances need not be as obvious, since a user will likely download software for its main purpose, and discover other added features when they play around with it. In software UI's, signifiers tend to be governed by 2-dimensional shapes or icons. For example, a small icon with three parallel, horizontal lines is typically a "drawer", often used for settings. Shapes like tabs indicate that a user can "pull" down some added functionality.


Sol Park - 1/29/2014 16:36:57

1) I always have a problem with shower faucets controls. When i move or stay in hotel, my question always arises 'where is the thing that changes the water flow from bath faucet to showerhead?'. In Korea, where I was raised, I never had shower faucets that had to be pushed up the small metal pin to move water to the showerhead. So, when I first came to America, it took me forever to find where the pin is. I am still not use to the shower faucets, and I always have trouble figuring out how it works whenever I encounter new shower faucets. I think such variation violates "signifier" in Norman's design principles because I am clueless with new shower faucets. Instead of the metal pin, a metal switch, which moves downward when the water flows downwards and upward otherwise, would be easily recognizable at a glance. 2) An "affordance" is the relationship between any interacting agent such as a physical object and a person, an animal and a person, a machine and a robot, etc. It determines what actions are possible. A "signifier" communicates where the action show take place. The author uses somewhat differently. He defines it as any mark or sound, any perceivable indicator that communicates appropriate behavior to a person. It can be deliberate and intentional. Signifiers provide valuable clues of social activities. An "affordance" is more difficult to understand since it is about relationships, not properties. So designers come up with idea, "signifier". Both are important factors in design.


Juan Pablo Hurtado - 1/29/2014 16:39:34

1) Beverages carton boxes with screw caps. When you are pouring your beverage from the box the liquid won't come out as a steady stream, instead will come out like alternating air and liquid making you sometimes spill all over the place. This is violating the signifiers, because if you instead of pouring the beverage through the "short side" as almost everyone does because is more intuitive, you pour it over the "long side", air will enter to the box making the stream of milk more steady.

So for solving this I would add a sign so the user can know how to pour it correctly or change the design and add a hole where air can enter making the pouring more steady.

2) Affordance is what you can do with something and the signifier is how the designer communicates what can be done and how to do it.

The main difference from a physical product and a software user interface is that in the physical product you can use the form of it as a signifier and also for communicating the affordance of it and in a software you have to explicitly show the affordance with ver clear signifiers on the same interface (As the example of the reading with the designer and the mentor).


Stephanie Ku - 1/29/2014 16:40:56

1) I travel quite frequently, and have had the opportunity to stay in several hotels in different countries. One object that is not really standardized, but is present in all these hotel rooms, is the lamp. Sometimes they are just simple desk lamps, sometimes they are fancy reading lamps by the bed, and sometimes you find them as tall standing lamps. However, turning on this lamp always took longer than it should.

Firstly, attempting to locate the switch would take extensive seconds (or even minutes). The switch can be found in multiple locations and forms. Sometimes it is a basic square switch [O -- ] at the base of the lamp, and sometimes they are tiny knobs hidden beneath large lamp shades close to the bulb. This violates the discoverability principle. If we cannot find the switch, we cannot figure out what actions are possible (maybe the lamp has dimming qualities) and how to perform these actions.

Additionally, the design of the switch itself violates the understanding principle. With the little knob under the lampshade, there are no signifiers telling the user how to turn on the lamp. Since it’s a knob, it could potentially have dimming qualities too, we won’t know until we try because there’s nothing that tells us that it dims. Moreover, what do we do with the knob? By instinct, we would think to turn it – now the question is do we turn it clockwise or counterclockwise? However, I have also encountered knobs that you don’t turn, but actually pull out or push in.

If I were to redesign this lamp, I would first locate the switch in a perceivable area of the lamp (i.e. not hidden under the lampshade). It would also be approximately eye-level. For example, I would not place the switch at the base of a standing lamp because the user would have to crouch down to the floor to turn it on. Furthermore, since there are so many different switch designs out there, I believe that it would be best to have some sort of signifier. Although it may detract from its aesthetic appeal, even an ‘On/Off’ text with arrows directing the way to turn the knob would prove helpful to the user. If we decide not to go with the knob, a simple push button (with the word ‘PUSH’) could also suffice. Lastly, as we are in a hotel room of the 21st century, it may be ideal to place the switch away from the device itself and with all the other switches on a wall (for ceiling lights, air conditioning, etc.) such that the user only has to go to one place to find all the appliance switches.

2) An affordance is a relationship between the properties of an object and a person (or any interacting agent), whereas a signifier is a communication device that tells the user where an action should take place. A simplified definition would be are that affordances determine what actions are possible, whilst signifiers communicate where the action should take place.

A physical product’s affordances will differ from a software user interfaces’ in a few ways. Firstly, sometimes a physical object would have infinite affordances. Take for example, a really heavy cup. The cup can afford drinking, but can also afford to be used as a paperweight. Yet, for software user interfaces, there is only so many ways an agent can interact with it. For physical objects, they have physical properties that could serve other purposes, whereas software user interfaces cannot serve these physical affordances. Given a user interface (e.g. iPhone), affordances may be limited to use of the fingers with pinching, swiping, and pressing. Therefore, the user interface’s affordances are, more often than not, limited to those that it was intended for. Additionally, user interfaces usually need to employ signifiers to communicate un-perceivable affordances.

In terms of signifiers, a software user interface does not have physical properties to signify the user, unlike physical objects. Everything that the software user interface wants to communicate has to be perceivable to the user. Both physical objects and software user interfaces can employ the use of signs, icons, labels, drawings, and colors to signify the user of what to do. For example, showing user arrows lets the user know that there is a next page. However, physical objects do not necessarily need to use these as they can communicate with a physical property, such as its structure. On the other hand, user interfaces is much more limited on signifiers.


Everardo Barriga - 1/29/2014 16:43:29

One example of a physical device that I have had problem with in the past is a pasta drainer. Recently while cooking pasta I tried to use a drainer I had that was bowl shaped and had two small handles on the side. I found it extremely difficult to drain the pasta in my sink without the actual bowl touching my sink. Although I am pretty strong, I am not strong enough to carry the weight of the pasta with one hand and pour the pasta with the other. I think this violates the consistency design principle Norman talks about in his article because the interface of the pasta drainer doesn’t seem to follow any rules and there is no clear way to drain the pasta. It doesn’t provide a consistent solution and instead can be used in any way the user wishes. I think one solution could be to make a pasta drainer that actually stretches across your sink so that you can drain the pasta and not have to actually carry the drainer and also the drainer doesn’t touch the bottom of the sink. Draining pasta doesn’t have to be a two-person operation anymore!

Affordances are the relationships between physical objects and an interacting agent. They basically tell the user what actions are possible. For example a table can only hold so much weight and what affordance tells you is that an elephant for example cannot “afford” to sit on it. A signifier tells you where the action should take place and they should signal to you to direct you in how the interaction takes place.


Munim Ali - 1/29/2014 16:44:16

1) The poor design that I'm going to talk about is pretty ubiquitous - electrical switches. An array of them are always placed next to each other without any indicator of what each switch is supposed to do. For example, the kitchen in my apartment has the light switch and sink garbage disposal switch right next to each other and I always end up turning the wrong one on. Another example, is when you have a large room with lots of lighting you never know which switch turns on which lamp - people inevitably end up turning all of them on (waste of energy).

I would say that this "everyday" object violates Norman' principle of mapping.

One way to fix this would be to engrave or emboss some kind of symbol on to these switches so even if it is dark you know what the switch is going to do just by the sensation of touch.

2) According to the reading affordance determines what actions are possible and signifiers communicate where the action should take place. I think affordances and signifiers do not differ much for physical products and software user interfaces. A possible difference is that a natural intuition of the physical world (the physics of objects) allows humans to deduce the "affordances" of a physical object when compared to a software user interface although this is soon becoming rare.


Justin MacMillin - 1/29/2014 16:50:27

1) I am a frequent runner, and use a GPS watch to keep track of how far I run, how long it takes me, and how fast I went for individual intervals within the run itself. The particular watch I use is called the Garmin Forerunner 405. This watch has two buttons on the right side of it, and a circular touch area around the screen called the "bezel." On the bezel, you can touch and hold on 4 different parts of it (top, bottom, left, right) to access different menus. First off, the user can use the bezel to select different items on the screen. However, you can only select something, and not go back. For example, my first thought when using this device would be to touch the right side of the bezel to go further into a menu (especially because the screens move in that [left] direction when you select an item), and touch the left side of the screen to go back. This is not the case, all parts of the bezel are used to select. So if I push the left side of it, it will select the item highlighted on the screen; which does not make logical sense to me. This is not an affordance issue, but however a signifier issue. Not only does the transition of screens hint to me to touch different parts of the bezel, but also training with other touch devices on the market. In my opinion, Garmin should have conformed a bit more and made it possible to both select an item and go back a screen.

2) The author says that affordances "determine what actions are possible" and signifiers "communicate where the action should take place." Affordances are a relationship between the user and the object, regardless of what kind of object it is. An affordance is a possible action that a user can perform with the object. For example, a table affords sitting at and supporting things. As a result, people can use tables for a wide range of purposes - for eating on (dinner table), for working at (desk), simply placing things on for storage, etc. A signifier is something about an object that hints to the user how to use the object. For example with a keyboard key, a signifier might be an arrow on it that faces a specific direction. If the key didn't have the picture of the arrow on it, the user would not know what the key is used for. They would only know that it is a key and they can push it. The arrow is a signifier to use that specific key to move the cursor in that direction. The fact that the key is on a keyboard affords pushing it, the picture on it is a signifier as to why you might want to push it. Affordances and signifiers differ for physical products and software user interfaces in a few ways. First, anti-affordances for software is never something related to the user being physically incapable of performing an action. Such as a person picking up a couch by themselves would not be a affordance of a couch. Affordances are much easier to list for a piece of software such as in the menu bars and task bars. People are less likely to read directions for a physical object to find out its affordances. Signifiers in user interfaces are often signals that appeal to making the user think to use their hands, such as a circle on the page to touch, rotate, move around, etc. In a physical object, things normally have words on them to signify how to use it (such as a door with "Push" on it).


Meghana Seshadri - 1/29/2014 17:19:06

1) An example of a physical device with bad design that I have used before is a tall lamp. It has a tiny rod that sticks out, located right at the very top of the lamp post (around 6 feet in the air). In order to turn on the lamp, you need to place your thumb and forefinger around it and turn it to the right not once, but twice. Furthermore, the material and shape of the rod doesn't allow for any sort of proper grip. Hence, my fingers continuously slip as I try to turn on the lamp. This device violates affordance and feedback. It violates affordance in the turning on the lamp for people who can't reach 6 feet in the air as the rod is so high up. It violates feedback as it takes two turns of the rod in order for the light to turn on instead of immediately turning on when the rod is turned once. I would re-design this lamp to possibly take out the rod, and replace it with a long chain. This way, the lamp can have the affordance of being turned on by people who can't reach the top of the lamp post, and by pulling the chain once, the light will turn on, giving immediate feedback.


2) The difference between an "affordance" and a "signifier" is that affordances reflect the relationship between the user and the object, while signifiers reflect a characteristic of the physical space on the object. The affordance figure out what actions are possible on the object, while the signifiers convey to the user how and where those actions can take place on the object. Affordances and signifiers amongst physical products are similar to those for software user interfaces. It's just that what Norman tries to point out is that signifiers are more relevant term to use towards software user interfaces than affordances. For example, buttons on a user interface tend to look like buttons and, hence, clickable. Generally, good user interface designs will make sure that elements that look like buttons are meant to act like buttons.


Chirag Mahapatra - 1/29/2014 17:19:31

1. One of the examples of poor design which I have come across is an old version of portable radio player. This has an antenna which is attached to a hinge which rests directly on top of the radio. The hinge can turn the antenna horizontally to rest on top of the device. To use the device one has to turn the antenna vertically and telescope it up. To carry the device, it comes with a handle which is fixed to the sides and rests behind the antenna. The handle rotates forward to the top when one has to pick it up. Hence, one will snap off the antenna if one pick it up too quickly. There are few ways to avoid this in real life while unless one is controlling their behavior by picking up the radio slowly and carefully or lowering the antenna back to horizontal position. I think something lacking is the conceptual model. This is because most radios would work fine without the antenna as well (atleast that is what happened with our radio). Hence, the question arose on the need of the antenna. As it happened, they were removed with subsequent versions. Also, the signifier associated with the handle was inappropriate.

The way I would correct the system would be to place the handles on the sides so that the radio can be picked up sideways. This will avoid damage to the antenna if it was required at all.

2. Affordances are the possible interactions between people and the environment. while signifiers signify what possible actions are possible and how they should be done.

For user interfaces the user can control only the perceived affordances. This is because the real computer system already exists. This is not the same for physical products where the designer can control the real and perceived affordances. However, there is little distinction for signifiers.



Christopher Schechter - 1/29/2014 17:20:35

1) I would consider the radio in my car to be designed poorly. I'm sure that to someone who is familiar with using radios, it would be fairly simple to use. Unfortunately, I'm not one of those people, and it took me a fair bit of trial and error to learn. It offers all the same features you'd find on any other radio, but the difficulty lies in the lack of proper signifiers. For example, most of the controls come in the form of easily visible analog buttons. For a while, I was fine with those simple functions: a power button, scan buttons, and a volume knob, in addition to a few others. But one problem I had was that when I would switch stations, the new station would briefly show up on my car's digital touchscreen display (the kind that shows you gps, lets you mess with the air conditioner settings, etc.), then disappear forever so that I would no longer be able to see what station I was listening to. It wasn't until later I discovered you could access an audio menu from within the car's display, which showed the radio station at all times. Immediately this became puzzling to me, since listening to the radio now involved two separate input methods: using the analog buttons for power, scanning, and volume, and using the touchscreen for basically anything else. Additionally, after messing around with it for a bit I figured out that by holding down certain buttons on the audio menu you could "favorite" your radio stations to reach them faster--but there was absolutely no indication of how you were supposed to do this, I just happened to find it by chance. If I were to redesign that radio, I would firstly make all the input on analog buttons. Splitting the input between the touchscreen and analog buttons is simply too confusing, and I feel that the buttons are advantageous over the touchscreen because they can be used just by feeling around without looking (which safer when driving). I would mark off part of the computer's display to always show what radio station is playing, and include a "favorite" button to save stations as well.

2) An "affordance" is a feature of a thing that benefits its user through some sort of interaction. For example, the power button on an electronic device affords the user the ability to turn on, and consequently use, the device. A "signifier" is something that makes an affordance visible and easy to understand for the user; so in the power button example, the circle-with-a-line-in-it power icon printed on the button would be a signifier.


Patrick Lin - 1/29/2014 17:27:42

1) Give an example of a physical device (an "everyday thing" as Norman would call it) with bad design that you have had to use. Do not think about software! Think about household appliances, sports equipment, cars, public transportation, etc.) Which of Norman's design principles did this device violate? How would you re-design it to solve the problem? I recently went on a tour of Japan and had trouble with several devices and appliances. Some of the issues were due to my inability to read Japanese, but a few had no instruction whatsoever in any language. For example, a shower at one of the hotels I stayed in had literally no visible controls except for the faucet and a single, cryptic, crystal knob. I tried the pulling, pushing, and twisting the knob to avail, and it took around ten minutes searching the bathroom with my sister before we realized a tube needed to be connected from the bathroom sink to the shower faucet; the shower itself did not produce any water, and any temperature control had to be managed reaching over to the sink and balancing the hot and cold knobs there. This example clearly violates the affordance of the knob, as Norman states: “Knobs afford turning, pushing, and pulling,” whereas in this case the knob was unmovable until the sink was connected, and even then was only used to toggle on the shower head and not control heat. There was also a lack of any signifier that the sink was necessary at all, as the only hint was that water did not come out of the bath faucet. Redesigning this would obviously be best if there wasn’t any hassle with the sink, but barring any plumbing changes, the process would also be made easier if there were instructions or pictures to guide a confused person through the process. Using a smaller metal tab/button that clearly only pushes in or pulls out instead of a knob that is often expected to turn as well would also have made the temperature controls more apparent.

2) Give a concise definition of the difference between an "affordance" and a "signifier". How do affordances and signifiers differ for physical products and software user interfaces? An affordance between an interacting agent and some object is a way that an agent could use the object. A signifier is a clue put in by designers that hints to an agent what sort of interaction is possible. With physical devices, people can use all their senses to test for affordances, for example pushing and pulling on a door to test which way it goes, or observing what sort of handle it has as a hint. In software, unless the designers anticipate what sort of interactions a user will attempt, the interface may have no response to a user’s trial and error (e.g. not knowing if a button is clickable unless some sort of audio or visual cue is given upon clicking). Signifiers are also more limited with software for the same reason, and generally relies on visual, audio, or haptic feedback, whereas physical objects can be manipulated in 3D space (e.g. a child can try spinning a wheel to determine its purpose) with much fewer limits (e.g. a child could also try biting a wheel to come to the conclusion that it’s not edible).


Sijia Li - 1/29/2014 17:28:52

(1). The fridge in my apartment has a really funny/bad design. When my friends came to my place and tried to open the fridge, they usually failed to open the fridge at their first try, (Why?), because the fridge has handles on both left side and right side! So, when a person enters the kitchen, he put his hands on the handle which is closer to the entrance of the kitchen and try to open the fridge. But that side (the left side) of handles can not open the fridge! It is the right side of the handles that can be used to open the fridge. And, there is no signs on the front part of the fridge!!! I have to say that I have no idea why the fridge company installed handles on both sides of the fridge. Maybe, it could be easier to take the whole fridge door off?! This design violates the "understanding" principle. Users at first do not understand why there are handles at both sides of a fridge; they also do not know which one to use to open the fridge. I would just put handles on only one side of the fridge so that the users will be very clear about how to open the fridge. For a simple device like this fridge, external signifies (signs) are not needed, according to Norman's comments on doors (Page 15).

(2). Affordances determine what actions are possible; Signifiers communicate where the action should take place. An affordance is a relationship; whether the affordance exists depends upon the properties of both the object and the agent. Signifiers signal things, in particular how actions should be done. The signifiers in software, unlike affordances, afford nothing new; they could be arrows and icons that signals permissible operations.

Thanks. Sijia Li


Robin Sylvan - 1/29/2014 17:29:58

A physical device I find myself consistently frustrated with is my car's clock. I drive a '94 civic, and the clock has a very simple design – the time, a 12 hour scale with no AM/PM markers, then two buttons: one for adding minutes, and one for adding hours. As an old car, I've had to do a lot of work on the car, meaning I've had to disconnect the battery many times. Every time I disconnect the battery, the clock completely forgets about the time, and I need to reset it. Setting the time on the clock fails in a multitude of ways. The most annoying part is its lack of account for human error – the two buttons only increment and loop around when they get past a certain time, I regularly pass the time by a minute while I'm holding down the button and driving trying to pay attention to the road. I then have to wait another 20 seconds or so for it to loop back around. Another way it fails is discoverability – the buttons aren't marked, and don't have any illumination the rest of the dashboard does at night. When I'm trying to readjust the timing at night, it can be hard to find the small buttons to use them. The major way I see of fixing the design without an entire overhaul would be to replace the small buttons knobs – this could allow for faster changing of the time as well as allowing the user to turn back in case they went too far. Also if light illumination was added to the knobs, it would be very helpful for adjusting the time at night. On a side note, a small rechargeable battery could be used to help hold the time while the car is disconnected from the battery, and would make for much fewer time resets. Affordance's are things that CAN be done, while signifiers tell the user what SHOULD be done. Signifiers point the user towards using the affordances of an object properly. In everyday life an affordance would be the doorknob and hinges on a door, and a signifier would be the sign saying “pull”. In software interfaces, affordances are all the ways a user can interact with a program, and signifiers can help guide them through usage in a reasonable manner. For example, when a user is filling in a sign-up sheet they are given a large amount of text box affordances, and signifiers can lead them through the sign up process by highlighting proper information.


Qianyun Li - 1/29/2014 17:30:01

1) Milk containers always lack of the signifier of where to tear open. I always tear the wrong side and spill all my milk. It violates "signifier" principle. If I can redesign it, I would write where and which direction I should tear it apart.

2) Affordance suggests the possible activities and relationship between a person and a object. However, a "signifier" is some sort of indicator, some signal in the physical or social world that can be interpreted meaningfully. They are fundamentally different. Physical products requires a lot of coordination with with the whole body while software usually only interact with hands and eyes. Therefore the interactions are more complicated and multi-dimensional with physical products.Physical products look for ease to operate with body and software look for ease to use with hand, eyes and brain.


Opal Kale - 1/29/2014 17:30:20

An example of a physical device with a bad design is the double fridge door in my house. When I open one door to get food out of the fridge and release it so it closes, the other door opens because of the impact of the first door when it shuts. This is a problematic because sometimes it causes food to rot. This violates the mapping design principle because some effects are out of their control (the other door opening when closing the first door). If I were to redesign it, I would have a more magnetized strip on each of the doors so that when closed, they require much more force to open, and when closing them, the force of the magnetism of the other door keeps it shut tightly. The difference between an affordance and signifier is that an affordance determines what actions are possible and a signifier determines where the actions take place. Affordances and signifiers differ for physical products and software user interfaces because affordances relate more to physical use and can get messy with software interfaces and signifiers over different interfaces are more universal.


Cheng Sima - 1/29/2014 17:31:41

1) Last week at an office, I had to use a microwave that I have never used before. Similar to many microwaves today, it had a display of time, and had many labelled buttons beneath it. After I put my food in, I saw that the time display already had numbers displayed, so I want to set my own time as to how many minutes I want the food heated. I pressed the CANCEL button, nothing happened. I directly inputted the number of minutes I want, nothing happened. Seeing my difficulty, a worker at the office told me that I had to first turn off the time display using the TIME button, and then select heating mode and minutes.

This device violated many design principles. It had ambiguous signifiers, as canceling the heating used CANCEL, but canceling the time on the display used TIME. It also violated "Conceptual Models" because it had just one display but two different buttons to cancel the display (only one button works for one unknown situation).

I would redesign the device by taking out the confusing TIME button. One display is canceled by one button "CANCEL" at all times.

2) Affordances are the universe of possible relationships that an agent and the object can have. Signifiers are perceivable indicators that help communicate where and how these affordances can happen. For physical products, signifiers are usually physical labels or buttons that indicate the affordance. For software UI, affordances can always be updated and signifiers usually become hidden after the first tutorial, or they may be found in the "Help" menu.


Sergio Macias - 1/29/2014 19:46:58

1) There is a lint roller I have been using for a few years now. It works great as a lint roller but the design was not given much thought. The lint roller was designed so that once you finish with the current layer, you could just peel it off to get a fresh layer and continue using. The problem arises with the fact that there is no signifier on how or where one should peel off the current layer. For a few months, I would just peel wherever, in whatever motion I thought seemed best at the moment. This would make removing the current layer difficult for sometimes it would rip and would not peel off correctly. After months, I learned that there was a small (read: very tiny) perforation in a diagonal line across the roller, but even then it was hard to find the exact spot one had to begin peeling off since it was not marked at the top. My solution to this problem would be to add a signifier as to where to begin peeling, perhaps a red line at the top of the roller where the perforation begins (which I actually added myself with a black marker). Also, one could add a light red dotted line where the perforation is so that one can see that that is how it must be peeled off. The problem with the lint roller is that it did nothing to communicate how to remove the current layer. Simple signifiers would fix this problem.

2) An affordance is a property of an object, with respect to the capability of the agent using the object. Signifier is a means of communication of which/where an action should take place. For physical objects, things are usually more natural since we have more experience with the affordances and signifiers of actual objects and affordances are usually in plain sight, i.e. a chair, I have never seen before, looks like other chairs I have sat in so I think that it has the affordance of supporting me. With software, because good design designates that things must kept orderly and not too over complicated, one cannot have all functions and affordances out in the open, but instead hidden, in some manner. Therefore it is particularly important for software to have good signifiers, otherwise you are left with a cluttered design in which all the functions are out in the open or in which you have functions which will remain hidden to the majority of your consumers.


Insuk Lee - 1/29/2014 19:47:37

The light switches inside my room have some bad design principles. There are three switches in the room and there are no indications as to which one turns on the main light and which one is the operating switch for which outlet. The mapping of this design is violated. I would try to switch this by replacing the main light switch with something more prominent and distinguishable from the other ones. As for the outlets, I would color the corresponding outlet and the switch the same so there would be no confusion.

An affordance refers to some function or purpose of the object whereas a signifier is some indication or hint as to what the purpose or function of the object is. In physical products these signifiers can be various but to my understanding of signifiers in software ui's, they can be symbols or already well established images that everybody can interpret. In this way, there are more limitations and time required for people to get used to these signifiers. Also in general there are more things that a physical product can provide than software products so the range of affordances of physical products is vast whereas for softwares it is limited to the current technological feats of pc/smartphones.