Taming HAL
Chapter Notes and Activities

Taming HAL by Asaf Degani

• Vocabulary
• Chapters 1 and 2
• Chapters 3 and 4
• Chapters 5, 6, and 7
• Chapters 8, 9, and 10

The Wonderful World of Machines

Taming Hal: Designing Interfaces Beyond 2001 uses familiar (and unfamilar) terms in specific ways. Below are some terms about machine "states" and user approaches:

• Event: an action performed on a machine (i.e., press, turn, insert, etc.).
• State: a particular behavior of a machine (i.e. on, off, idle, armed, etc.).
• Transition: the movement from or change of states.
• User model: similar to "mental model," this is what the user sees and knows about as the system (p. 29).
• Machine model: description of ALL the configurations and transitions underlying a system (p. 29).
• Non-determinism: the quality of a system that behaves in a way that can't be determined (p. 35).
  • We saw determinism in a different context in Johnson, didn't we?
• Side effect: events that can bring about multiple outcomes (p. 43).
• Automatic transitions: non-user-initiated transitions (p. 59).
• Concurrency (pp. 70-72): "several things are happening at the same time in the same space."
• Hierarchy (pp. 74-76): a super state that embeds states (or other functions/features) that are not equal. Think of modes or states as containers.
• Synchronization (p. 77): when an event triggers another event at the same time.
• Population Stereotypes (p. 83): culturally held expectations regarding anything; in our case, objects have certain affordances that should act certain ways.
  • For instance, what do you do when you see a door handle...a door knob...a button?
  • What other concept do population stereotypes sound like?
• Universal Stereotypes (p. 93): well-established principles of user-interface design or any design.
  • For instance, if you click on a right-pointing arrow, what do you expect? If you click a down arrow, what do you expect?
  • Well, nothing is truly universal; there are always exceptions...a paradox.
    
• Initialization (p. 124): (initial mode or reference values) configuration on start up.
• Mode (pp. 122-125): the displays of an interface--video, audio, flight map, etc.
• Reference Values (pp. 124-125): the selections a user can make within a mode--levels, choices, channels, etc.
• History Settings (p. 125): the previously selected reference values on from the previous setting.
• Default Settings (pp. 125-126): the action that occurs (the machine reverts to) when a user fails to interact with an interface.
  • Notice the difference in the definition of "default" that Degani gives and our understanding of "default settings" for software such as Microsoft Office.
  • In Word 2003, for instance, the default font size and typeface was Times New Roman 12pt.
    

Chapters 1 & 2

Most of Chapter 1 is about getting us to think about how to approach the book and, therefore, ways of thinking about machines. Although Degani is focusing on machines for thinking through design, we're going to focus on machines and documentation; after all, our overall goal is to consider the user's experience--not just static documents. Remember, no one wants to read what you write...

I would like you to think back to an earlier time when you first learned how to ride an elevator. It might be difficult to recall, but I want you to try and remember how you learned that you could walk into a closet-sized room with Star Trek doors, push a button, and expect to go up...or down. If this doesn't get your mind going, list or describe elevator etiquette or procedures you adhere to and expect of others. Logon to Moodle and respond to the "Elevator Prompt."

Below (the strike through text) was the original plan, but I want to give you a chance to have more participation credit, so let's attempt to do this activity as a larger class. Review all the questions for the next 10 minutes and be ready to say something about them. If the conversation doesn't seem fruitful, I'll just keep the original assignment and have you create a webpage to respond to a prompt below.

This activity is related to your reading for today. Create a Web page that you will link to from your homepage for these responses and write them in paragraph form.

1. Explain abstraction and its relation to interfaces and diagrams of interfaces, systems, phenomena, etc. Why are such "items" important for users, specifically, and technical communication in general?

2. Describe a machine's "state" in your own words. Why is it important to think about machines having "states." In other words, why might knowing a machine's state help a writer of user documents?

3. Look at the following abstractions of the Washington, DC metro (the world's greatest mass transportation system):
   
   All the images are on the Abstractions page
   
   Question A: What systems do the above abstractions replicate? Why are they considered "user documents."
   
   Question B: If you haven't already, read the note for Ch. 2 (pp. 286-287). How are the above abstractions better than the "actual" image (below)?
   
   Image 4 on the Abstractions page
   
   Question C: Refer back to the first three abstractions. What is the average distance between metro stops? Is the average distance relative to distance from the center? Which areas of the DC metropolitan region have greater average distances? Why might distances not be to scale?

Chapters 3 & 4

Chapter 3 and 4 are about machine (or is it human) error. Chapter 3's subjects are quite benign--alram clocks and vending machines--but Chapter 4's subject is grave--plane crash. Let's consider these chapters as a larger class before going into the discussion points below.

I hope the following are discussion points. I want you to have HAL inspire your next user doc (#3). Read over the questions and contemplate them for a bit. Then, we'll discuss them as a class.

If our discussion isn't fruitful, these will be individual assignments you post online.

1. Come up with a list of five non-deterministic features on a technology or several technologies. Give the technology and explain the feature or features that are non-deterministic. If possible, explain how to correct that.

2. Explain a Golem-like technology (I wonder if there's any connection to Gollum...) that you've encountered or heard about (not one from our reading). Describe the technology and explain how it behaves in a non-deterministic fashion? This is related to the question above, but, as you can see, it asks for a single technology and not "features on a technology." This can be a fictional technology.

3. Explain what it means to accept that machines go awry? What does Degani say about this acceptance, and who else agrees with him?

4. Death by technical communication classification. Briefly describe the technical classification that sealed the fate of Korean Airlines flight 007. What are two other technical classifications that could make an item or person go from safe to danger based on user perception? (tough one)

5. The ends of Chapters 3 and 4 both ask, who's responsible for non-deterministic technologies. Considering either a specific technology or many, come up with a reason (almost like a heuristic) for placing blame when it comes to non-deterministic technologies.

Chapters 5, 6, & 7

Taming Hal: Designing Interfaces Beyond 2001 asks us to think about ways to understand machine behaviors.

We're going to try to have a discussion, but I'd like some constructive small-group discussions first to get us thinking about machines and their interactions with humans. In groups of no more than 3, watch this clip from 2001: A Space Odyssey (I'll show it on the big screen), and think about any connections you might find to the chapters we read for today (Degani Ch. 5, 6, and 7). Specifically, consider the following questions:

• What are some population stereotypes that "influence" Dave's approach to disabling HAL?

• How might knowing population stereotypes be beneficial in creating effective user documents?

• Because future "projections" of technologies are atomized or amplified versions of a present's technologies, how might you update the procedure for dismantling HAL? In other words, what are some common population stereotypes that would make Dave's job different in the actual year 2001 (well, a decade-plus later, in fact)?

• Fun question: Why do we often see evil machines in science fiction when most of the time I hear people LOVE technology?

Then, we'll either go chapter by chapter and discuss the main points that help us understand interaction from a user's perspective or jump around across the chapters. Below are some questions to get us thinking:

• Users create simplistic explanations to deal with things beyond their understanding (p.83)
  • How might we use this knowledge?
  • Should we take the time to explain what's going on to a user? What's our focus...how in depth should we be?
  • On the other end of the spectrum, what shortcuts do you experts use when dealing with software?
• What do you do when something goes wrong?
• Cordless Phone Population stereotype
• Light Switches
• Alarm Clock Radios
  • Turning the buzzer down to 0--totally silent
  • Having two alarm times set
  • Radio frequency changes based on its position in the room...and your position
  • What's your threshold for bad or weak signals on the radio or television (satellite, cable, antenna)?

Possible Individual Activity

If the conversation doesn't have lots of contributions from many class members, we'll do the following questions individually. As in, you'll answer the questions on a web page you link to your home page:

Chapter 5:

1. Describe what Degani means by concurrency. Besides the instruments/situations he brings up, briefly discuss the concurrency of another instrument you're familiar with.

2. Comment on hierarchy in systems similar to your description above. Why do designers have hierarchal structures in their instruments?

3. Why do we encounter synchronization in our modern machines? Explain why such a phenomenon makes user testing and effective user documents so important.
   
   Chapter 6:

4. What are population stereotypes with regard to machines? What are some common ones that seem to be apparent across a few instruments?

5. What is the difference between 'population stereotypes' and 'universal stereotypes'? Where might a design run into trouble if based on particular population stereotypes? Be specific about the instruments.
   
   Chapter 7:

6. Let's be practical for once. Many design choices are for cost benefits. Is that such a problem? Why or why not? What might be the end result of pushing economics to the side when taking users' needs into account?

7. Degani mentions on pages 97-98 that "to consider any interface from the user's perspective, we must begin by asking what are the user's tasks." From such a statement, how is Degani viewing human-computer interaction? Who might disagree with Degani's discussion of "task"?

8. Why does Degani claim that combining two technologies into one is different from integrating two technologies into one?

Chapters 8, 9, &10

So you got out of having a quiz for chapters 8, 9, and 10. Well, you should still be able to define or know these fun facts from the reading:

Chapter 8 topics:

• Alarms no one can hear
• No back ups--can you expect users will have back up plans.
• GIGO-Garbage in Garbage Out
  • False interpretations of the facts
• Design flaws in machine-machine interaction

Chapter 9 topics:

• Expectation to use Walk-in interfaces
• History settings and multiple users
• Browser history
• Security vs Convenience
• Users should be aware of the structure of devices...but how aware?
• The future will "grant" us more user interfaces
  • Where might you find them?
  • Where are some that you think mark our culture?
    {Think instant gratification, consumerism, (the idea of) freedom, leisure (entertainment), etc.}

Chapter 10 topics:

• Population stereotypes of the Internet
• Browsers
• Forms
• 'http://' vs. 'https://'
• Bankrate.com
• Is it unethical to go against user expectations?
• eCommerce and the web