Human-Computer Interaction 3e ­ Dix, Finlay, Abowd, Beale

exercises  -  18. modelling rich interaction


Can you suggest any improvements to the screen button feedback problem discussed in Section 18.2 that would distinguish at the interface between the two cases of hitting or missing the button? Is there any guarantee with your solution that the user will notice the distinction?


One fix for the button feedback problem would be to have the function attached to the button be invoked when the mousekey is pressed down, rather than up. But this solution would not work for functions invoked on pop-up menus. Another possibility is to have some visible or audible feedback from the button associated with the invocation of the function on the mousekey release. This solution does not guarantee that the user notices the added effect unless their attention is focused on the added visual or aural effect. Another possibility, subtly different from the last suggestion, would be to associate the addition verbal or aural cue to the error case, when the mouse accidentally slips off the button between the press and release of the mousekey.



Brian wants to make a dinner date with Alison. He knows she will not be able to read email, as she is away for a few days, and he doesn't have her hotel number. He types and prints a letter, which he puts in her pigeonhole. Alison's secretary always checks the pigeonhole several times a day, and when she finds the letter she reads it and rings Alison and tells her.
Analyse this story using a status-event description.

answer available for tutors only

There are many levels of detail. It would be possible to use S-E analysis to describe even the act of typing, etc. However, the following is a high-level description:

  1. Brian intends to make a dinner date (an event in the future).
  2. His intention is also an actual event.
  3. His posting of the letter in the pigeonhole is an actual event.
  4. The letter changes the status of the pigeonhole
    (noting that a mediating status is a common way to communicate events).
  5. Alison's secretary polls the pigeonhole (a common way to find out about status change).
  6. When Alison's secretary finds the letter there is a perceived event for the secretary (this is delayed from the event at (3), as is normal with polling).
  7. Alison's secretary rings Alison causing a perceived event for Alison.

N.B. the data flow of information in the content of the letter is largely irrelevant for the S-E description.



Look again at the tea making task analysis in Chapter 15 (figure 15.4). Go through this and look for triggers and placeholders. You will need to make assumptions (e.g. is the kettle the kind that whistles when it boils?) so document these.

answer available for tutors only

We'll list the main task sequence first and then look at the sub-tasks for tasks (1), (3) and (5).

0. make cups of tea

triggers - the whole task may be triggered by thirst (internal event), or perhaps by guests arriving(external event).

1. boil water

triggers - the trigger for this is that you have just gone to the kitchen to make the tea.
placeholders - in the kitchen with no boiling kettle and no made tea! However, this would also be the placeholder for other tasks in the kitchen (e.g. fetching biscuits). It is likely that capture errors may occur here for less frequent kitchen tasks where you go to the kitchen and bring the wrong thing.

2. empty pot

triggers - just put kettle on and standing in kitchen
placeholders - pot filled with cold tea (yuck!) - not this was already in the plan ... the plan says do (1) first ... there is no intrinsic sequence, so it would be possible to empty the pot then put the kettle on to boil, but this would be 'inefficient', wasting time - does this matter in a domestic kitchen?

3. make pot

triggers - kettle boils ... but how do you notice? Are you observant? Standing waiting for it to boil? Perhaps it has a whistle when it boils? [[Alan says he can spend a whole morning putting the kettle on to boil, not noticing that it has boiled (missed trigger) and then restarting the process ...]]
placeholders - empty pot, kettle hot (??light off for electric kettle) ... if you forget to put the kettle on before emptying the pot, it would be possible (especially with an electric kettle, or if the kettle had already been warm) to pour cold water in the tea pot!

4. wait four or five minutes

triggers - this is not a 'regular' task, so more to remind us, in the HTA, that there is a delay. However, because it is there it makes the trigger for (5) more important as you are likely to do something else (chat, wash dishes, etc.) while waiting.
placeholders - the placeholder for this is "pot ready and cups unfilled" ... which will be the same as (5) below ... how do you know that (4) is over? Do you check the time on the kitchen clock (which is not in the HTA)? Do you set a little clockwork timer (again not in the HTA)? Do we just rely on a good internal body clock, or perhaps occasionally pour a tiny drop into a test cup to see if the tea has brewed?

5. pour tea

triggers - four or five minutes have passed - a temporal trigger, but like all such triggers: how do you know? If you are very forgetful you may set an explicit timer, or simply rely on remembering after the right sort of time. Clearly potentially unreliable.
placeholders - the full pot and empty cups tell you that this is where you are ... but not that a full cold pot is the trigger for (2). If you use a tea cosy (little woollen hat over the tea pot), you may not be able to tell easily whether it is hot or cold. ... Could a very absent minded tea maker pour yesterday's cold tea?

Subtasks of - 1. boil water

1.1 fill kettle

triggers - this is the first sub-task so the trigger is effectively the (1) trigger.
placeholders - largely mental through routine, with backup "kettle empty and no gas"

1.2 put kettle on hob

triggers - kettle is probably in your hand after filling it
placeholders - same as trigger

1.3 turn on and light gas

triggers - just put kettle down (immediate trigger) ... with backup (below) of "kettle on hob and not lit" - however, this is a fairly poor trigger as it is not very salient unless one explicitly thinks "where was I?"
placeholders - partly mental "just put kettle down" ... but if you were interrupted then the secondary placeholder would be "kettle on hob and gas not lit". However, unless you remember to pick up the kettle to check it is full, this is identical to the backup placeholder for (1.1). Then if you erroneously try to refill the kettle this is not too bad ... but if you accidentally skip task (1.1). and do (1.3) instead, a potentially empty kettle is being heated ... wait for that smell of burning metal :-(

1.4 wait for kettle to boil

triggers - like (4) this is a place where the HTA is emphasising a delay
placeholders - the gas is on and the kettle not boiling ... how obvious is the lit gas? Can you mistake this for 1.1. or 1.3 placeholder and stand waiting for a kettle that is not lit? This is perhaps easier with an electric kettle where the signs are even less obvious.

1.5 turn off gas

triggers - kettle is boiling! ... as with (3) how do you notice? If you fail to notice with a gas kettle then the steam will do an effective job of stripping your kitchen wallpaper and eventually the kettle will boil dry. Secondary trigger (see below) is a gas burning with the kettle removed.
placeholders - boiling kettle on hob ... but the boiling kettle is also a trigger for (3) ... what happens if you take the kettle off the hob to pour water into the tea pot (a good idea to make sure it is really boiling when it hits the tea leaves) and then leave the gas burning on the stove. The kettle will no longer be actively boiling; there will simply be a gas burning with nothing on it. You may notice the gas burning (secondary trigger), but with an electric ceramic hob there is no visible sign with consequent danger of touching or putting something down on the hot surface.

Subtasks of - 3. make pot

3.1 warm pot

triggers - same as (3) kettle has boiled. [[note: a really expert tea maker may do this warming using hot water from the tap or with partially boiled water to avoid the kettle cooling or overboiling before (3.3)]]
placeholders - empty cool pot

3.2 put tealeaves in pot

triggers - immediately after (3.1)
placeholders - empty warm pot with no tea in it

3.3 pour boiling water

triggers - immediately after (3.2)
placeholders - empty warm pot with tea in it ... but notice that the placeholders for (3.1) and (3.2) are not so different ... have you ever added tea twice, or poured in hot water without adding tea?

Subtasks of - 5. pour tea

5.1 put milk in cups

triggers - note that this task occurs first as the initial task and then for each task. There is thus an initial trigger for the overall task (5) and then a sense of "I'm in the middle of making tea with milk jug and tea pot in my hand"
placeholders - empty cup ... and we know which cup because it is the 'first' empty one in whatever sequence we are filling them (clockwise, left to right)

5.2 fill cup with tea

triggers - just done (5.1)
placeholders - cup with milk and no tea

5.3 do sugar

triggers - finished last cup for 5.2
placeholders - all the cups are full ... but this will be true after as well?

5.3.1 ask guest about sugar

triggers - first time round because (5.3) has been triggered, subsequently because one is "in the middle" of doing sugar ... if interrupted could you forget to ask all guests?
placeholders - some guests not asked ... need to remember which ones ... but if they are sitting there not so hard

5.3.2 add sugar to taste

triggers - just got answer "yes please, N spoons"
placeholders - you need to recall which guest ... relies on memory, the tea looks the same with or without sugar! If interrupted between 5.3.1 and 5.3.2 two errors may occur: (i) you may have formulated the intention to put the sugar in and not done it ... when you return to the tea process you may recall the intention and think you have done it! (ii) you may lose count, (iii) you may have added the sugar, but the act of putting sugar in cups is similar for each guest so hard to recall and you may erroneously add sugar twice for the guest.

Finally …

Did you realise making a cup of tea was so complicated! In practice we rely on the fact that we do most of the actions habitually, but most of the potential failure points do occur if we are distracted or interrupted in the middle of the processes. For an analyst looking at a situation these may be points for technological intervention to help prevent failures (e.g. adding a whistle to the kettle if it doesn’t have one). If we are a designer planning a technical intervention we should notice where and why the triggers and placeholders work in order to prevent us creating failure points. For example, if we install a new very silent electric kettle it may be hard to notice whether or not it has been turned on or has boiled.



Rank the following in terms of levels of intention or consciousness
automatic doors into hotel, automatic water taps in wash basin, reversing lights in a car, ultrasonic burglar alarm, auto-numbering lists in a word processor, web page counter, font menu in word processor that shows recent fonts at the top of the list
If you have a group, you could each rank them separately and then discuss your answers.
Why are some more consciously considered than others?
Think of more examples.

answer available for tutors only

A bit open ended ... as interesting for the analysis as the ranking


None have high level of intention
automatic water taps in wash basin
these get close to intentional, you put your hands under and often deliberately move them where you think the sensor is


automatic doors into hotel
intention - medium: you step towards the doors because you want to walk through, but you work on the expectation that they will in fact open, and you will usually adjust your gait to give them time to open.
reversing lights in a car
intention - low: you put car into reverse to drive backwards, not to turn lights on.
expectation? - if you think about it you know that the reverse lights will come on, but if they don't come on, even in the dark, you may not realise at first, just feel reversing is more difficult than usual
auto-numbering lists in a word processor
initially low intention - you type the number to format the list yourself, but if the auto-numbering works well (!!) you may come to rely on it "if I just hit return it will fill in the number for me" - that is a move towards expected interaction.
font menu in word processor that shows recent fonts at the top of the list
intention - virtually zero - you use a font in order to use it, not get it put on the menu. However, if you understand the rule you may start to know whether to look at the menu top or in the longer alphabetical listing if you know you have recently used the font ... this is a level of expected interaction.


ultrasonic burglar alarm
intention - none: the burgler certainly does not intend to set off the alarm. Of course, an expert burgler may learn to move to avoid the sensors!!
web page counter
intention - none: you visit the page to see it, not to increment the web counter (unless you are the page author trying to make it look visited!). Of course, the presence of the counter is visible as a trace of previous visitors - and perhaps a measure of quality: lots of visitors = popular (but does the presence of the counter suggest desperation?)


EXERCISE 18.5 [extra - not in book]


answer available for tutors only



EXERCISE 18.6 [extra - not in book]


answer available for tutors only


Individual exercises

ex.18.1 (ans), ex.18.2 (tut), ex.18.3 (tut), ex.18.4 (tut), ex.18.5 (tut), ex.18.6 (tut)

home | about | chapters | resources | exercises | online | editions | interactive | community | search | plus +++
exercises: 1. human | 2. computer | 3. interaction | 4. paradigms | 5. design basics | 6. software process | 7. design rules | 8. implementation | 9. evaluation | 10. universal design | 11. user support | 12. cognitive models | 13. socio-organizational | 14. comm and collab | 15. task models | 16. dialogue | 17. system models | 18. rich interaction | 19. groupware | 20. ubicomp, VR, vis | 21. hypertext and WWW