From the ID Foreward. By Dr. James Foley

Transcription

1 From the ID Foreward By Dr. James Foley

2 Design is a Process It is interdisciplinary Know your user Consider alternatives Prototype early and often Test(Fail) early and often

3 Advised approach Know who your user is User-centric design, YES Techno-centric design, NO Don t jump to the solution

4 From the DOET By Dr. Don Norman

5 Norman on DOET The three outstanding themes: It s not your fault There are Design principles Conceptual models Feedback Constraints Affordances Gulf of execution Gulf of evaluation

6 Terms to be seen in DOET Design model (designers s conceptual model of the world) System model (the model of the world presented by the device) User model (user s conceptual model of the world)

7 Terms to be seen in DOET Knowledge in the world vs Knowledge in the head

8 DOET Chapter 1 Models, Modes, Execution and Evaluation ITEC4130 Dr. Jim Rowan

9 Start with some impressive design failures! Apollo 13 mission representation failure The failure was shown on the interface It was buried in pages of data ~45 minutes before it was found A320 v=_em0hdchvly

10 Stuff in the world DVDs Cell phones Automobiles Airplanes Student IDs with RFID chips Turn arrow at PepBoys on JimmyCarterBvd

11 Poorly designed objects Difficult to use Don t provide clues as to their use EVEN WORSE: They provide false clues They have hidden modes of operation They are generally unpleasant to use, unfriendly

12 Well-designed objects Provide clues as to their use Easy to interpret Easy to understand

13 How does this happen? For objects in the natural world: form follows function For objects in the man-made world: form can be completely disassociated from function Made worse by: The transition from the mechanical to the electronic Mechanical objects are visually inspectable Electronic objects are NOT visually inspectable

14 The glass door trap

15 Design principle: Visibility Parts should be visible If they aren t, how do you know what to do? The principle of visibility the visibility of affordances (possible actions) the visibility of the result (what got done?) Parts must convey the correct message Mappings: What the object tells you about how to use it (through the parts it shows you) Do they match your expectations of those parts?

16 Hidden mappings: (The Leitz slide projector control) Only one button but two operations Not designed for discoverability Does it violate Norman s Visibility? Does it violate Norman s Mapping?

17 The phone example in book The hold function lacked results visibility: Was the call I just put on hold actually put on hold or was it disconnected? Older phones flash a light beneath the hold button (at least it connects the flashing light to the button that activated it) Instructions made it worse: Written in the language of the designer not the language of the user

18 About the telephone: Did you know Dial tone was an intentional addition Made the invisible (do I have a connection?) visible It tells you that you can start dialing

19 Norman s Principles Visibility of: Available actions through appropriate clues Results through feedback of one s actions

20 Affordance and Causality The perceived and actual properties of a thing determines how it is used. Affordances are strong clues. Correlation does not imply causation (Though as humans we REALLY want it to!) Temporal correlation (two things happening at the same time) Implies causality Can give rise to superstition

21 Affordance and Causality For push buttons (especially on remotes) Sometimes they click Sometimes a light flashes The lack (or lateness of) of expected result Gives rise to repeated action Like an idiot, you keep pushing the button hoping to eventually get something to happen

22 Projected keyboards Why? You don t have to carry them with you!

23 Projected keyboards Why not? (next slide )

24 Projected keyboards Why not? fingers block the display no audible click (can be added with software) no physical movement no physical feedback on finger positioning room lighting can interfere

25 Projected keyboards possible partial solution? software generated click results in auditory visibility The typed word appears on the screen results visibility What about solving the finger position problem? (not so easy)

26 Problem: fingers hide keyboard? show the selection larger Tiny keyboards results affordance issues

27 Users bring expectations; where do they come from? Some are cultural black AND white are the colors for funerals Some are explicitly learned (standards) cold water is on the right hot water is on the left Some may be basic to humans The attraction of the campfire The beauty of the arch as a basic form

28 Clues provided by the device can support the user in his quest to use that device Constraints what does the device keep you from doing? Mappings are the controls reasonably placed? do they move in appropriate directions? Affordances what does the device tell you?

29 Constraints example Consider scissors Holes in the handles (afford finger insertion) Hole size constrains what can be inserted fingers, maybe; legs, probably not different size and shape hole for thumb

30 Natural Mappings Relationship between controls and actions are apparent to the user Types of natural mapping Spatial analogy: Up on the control is up Cultural or biological (rising level is more) Additive (loudness, brightness) Substitutive (increasing pitch is more)

31 Models A designer has in his head a model of what is to be made The designer communicates the function of a device through its shape, appearance and controls What the user has in his head that directs his use of a device

32 Models, Mental and otherwise Three different models Design model: The designer s conceptual model. System image: The model implied by and presented by the device s structure User s model: The mental model you hold that guides your actions

33 Models, Mental and otherwise Consider the refrigerator the visible controls (the system image) imply two separate systems in actuality it is only one system Errors arise when the user (using his mental model implied by a flawed system model) tries to operate the system

34 But: Who s Model is it really? The designer has a model when creating a device The user has a model when trying to use a device The system image is how the designer communicates with the user; It s the device s physical and operational form. The problem: You (as the designer) are most probably NOT the user so you d better know the user!

35 Problems: Hidden Modes A mode is a change in context that affects the interaction with a device. A button means one thing in one mode and something different in another mode Many features with few controls give rise to hidden modes featuritis Modes must be made visible

36 Hidden Mode Problems Airbus 320 was a mode problem pilot was in flyby mode the airplane was in landing mode The fact that the plane was in landing mode was not readily visible Flying into the Everglades: A tale of human attention/distraction Going off autopilot was silent The mode change was not announced

37 The Good Visibility Natural mapping from control to controlled Feedback Number of controls = number of functions or make certain the modes are visible

38 The Bad Hidden modes Arbitrary mappings The stovetop example False conceptual model The refrigerator example Many functions, few controls cell phones, tv remotes

39 Feedback: Good! Tactile feedback (keyboards ) Auditory feedback (keyboards, phones ) Visual feedback (pencil, changing lights ) Haptic feedback (vibration )

40 Why are there so many bad designs? Multiple stakeholders in a design venture Designer Manufacturer Retail outlet Purchaser Maintenance

41 Technological Paradox Technology provides more functionality; possibly reducing the number of devices needed More functionality increases complexity Making it more difficult to learn Making it more difficult to use So Life s simpler but more complicated! Creeping featurism: Attractive but deadly!

42 Human Error? You use a device and continually make the same mistake Is it your fault? Is it the device s fault? Is it the designer s fault? Yes to all three but knowing what mistakes humans are subject to make allows the designer to avoid them.

43 Folk Theory and Home Heating Thermostats Humans want causality! Naïve theories abound. The timer theory: Half way is half the time. The valve theory: Half way is half open. Truth: It s on full blast or it s off.

44 Norman s Structure of Action To get something done you need: A goal to achieve in that world A world in which to act Then take action: Make a change in the world Check the world to see if the goal has been met

45 Goals Goals are Underspecified: They don t tell precisely what is needed to be done to accomplish them Get dressed Go to work Drive home

46 The Action Cycle: Two types of Actions Execution Make a change in the world Evaluation Check the world to see if the goal has been met

47 Stages of Execution (teasing it apart) Form the intent to act on the goal (mental) Form the action sequence required to achieve that goal (mental) Execute that action sequence (physical)

48 Stages of Evaluation (teasing it apart) Perception of the world as it is after execution Interpret that perception according to our expectations Evaluate the interpretation against our intentions and our goals

49 Goals (read this counter clock wise) Intentions Evaluate interpretations Action sequence Interpret perceptions Execute Perceive the state of the world World

50 Norman s Seven Stages of Action Form the goal Form the intention Form the action sequence Execute the action sequence Perceive the state of the world Interpret the perception Evaluate the interpretation against expectations

51 Everyday Tasks a class of tasks we all do all the time in the course of each day Humans are Opportunistic in nature They take advantage of circumstances Implies that action is triggered by opportunity Use the environment to trigger activity Not necessarily planned in advance you are reminded as you walk around that the dishes need washing you need to shop for groceries

52 Gulf of Execution The difference between the intentions on the part of the human and the allowable actions on the part of the system VCR bridges the gulf of execution found in the reel-to-reel videotape machines there is no way to foul up the threading the machine.

53 Gulf of Evaluation A measure of the amount of effort on the part of the user needed to interpret the state of the device and determine if (or how well) expectations have been met. VCR: Can t tell if a tape is in the machine. Press play and it just sits there (in some models)

54 Norman s Principles of Good Design Visibility: Can the state of the machine and the possible available actions be easily determined by inspection? Good conceptual model: Is the system image a good one for the intended users? Good Mappings: Are there ties between actions and results? Controls and effects? Feedback: Does the user get feedback from the results of actions taken.

55

56

57