TA
MO
DIA
20
02
managing the ecology of interaction
Alan Dix
Lancaster Universitywww.hcibook.com/alan/papers/Tamodia2002
TA
MO
DIA
20
02
overview
• rationale
• phenomena of rich interaction
• new paradigms
TA
MO
DIA
20
02
the problem
• task models– formal description
• situatedness– unique contexts
• ethnography– rich ecologies
bringingthem
together?
TA
MO
DIA
20
02
option 1 – reject formalism
everything
formaltasks
TA
MO
DIA
20
02
option 2 – enforce formalism
everything
formaltasks
TA
MO
DIA
20
02
option 3 – expand and accept
everything
formaltasks
TA
MO
DIA
20
02
HCI – dialogue with the computer
TA
MO
DIA
20
02
dialogue with the environment
ubicomp – no computer/artefact divide
wearable/cyborg – no computer/user divide
TA
MO
DIA
20
02
phenomena
TA
MO
DIA
20
02
phenomena
• collaboration
• information
• triggers
• artefacts
• placeholders
• situatedness
• intentional cycle
+ continuity & duration
TA
MO
DIA
20
02
collaboration
• already in several notations– e.g. CTT
• add artefacts too ?
TA
MO
DIA
20
02
information
pre-planned cognitive model
goal action
situated action
environment action
TA
MO
DIA
20
02
control
• open loop control– no feedback– fragile
controlsystem
environmentactions
TA
MO
DIA
20
02
control
• open loop control– no feedback– fragile
• closed loop control– uses feedback– robust
controlsystem
environmentactions
feedback
TA
MO
DIA
20
02
adding information
boilkettle
get outcups
make potof tea pour tea
pour teaPlan 0: 1 then 2 when kettle boils 3 then 4
0.
1. 2. 3. 4.
how manycups?
TA
MO
DIA
20
02
triggers
process – what happens and order
get post frompigeon hole
bring postto desk
open post
TA
MO
DIA
20
02
triggers
process – what happens and order
triggers – when and why
first thing in themorning
holding post at coffee time
get post frompigeon hole
bring postto desk
open post
TA
MO
DIA
20
02
artefacts
• ethnographic studies
• as shared representation
• as focus of activity
• act as triggers, information sources, etc.
TA
MO
DIA
20
02
placeholders
• knowing where you are in a process– like a program counter
• coding:– memory– explicit (e.g. to do list)– in artefacts
TA
MO
DIA
20
02
where are you?
1. controllerchoose newflight level
2. controllertell pilot newflight level
3. pilotconfirm newflight level
4. pilotascend tonew level
5.new flight
level achieved
TA
MO
DIA
20
02
step 1. choose new flight level
1. controllerchoose newflight level
2. controllertell pilot newflight level
3. pilotconfirm newflight level
4. pilotascend tonew level
5.new flight
level achieved
TA
MO
DIA
20
02
step 3. flight level confirmed
1. controllerchoose newflight level
2. controllertell pilot newflight level
3. pilotconfirm newflight level
4. pilotascend tonew level
5.new flight
level achieved
TA
MO
DIA
20
02
step 5. new flight level acheived
1. controllerchoose newflight level
2. controllertell pilot newflight level
3. pilotconfirm newflight level
4. pilotascend tonew level
5.new flight
level achieved
TA
MO
DIA
20
02
continuity & duration
• system models – event centric
• status–event analysis– continuous time (status) and discrete (events)– many generic issues and phenomena
• task models:– in the annotations and descriptions– concurrency – true or interleaved?
TA
MO
DIA
20
02
intentional cycle
artefacts artefacts
informationand placeholders
actions
triggersmemory
TA
MO
DIA
20
02
paradigms
TA
MO
DIA
20
02
paradigms
• socio-organisational Church–Turing hypothesis
• embodied computation
• embodied interaction
• incidental interaction
TA
MO
DIA
20
02
the socio-organisationalChurch-Turing hypothesis
TA
MO
DIA
20
02
TA
MO
DIA
20
02
TA
MO
DIA
20
02
TA
MO
DIA
20
02
TA
MO
DIA
20
02
the Church-Turing thesis
• the THEOREM– Church’s lambda calculus and Turing
machines are ‘equivalent’
• the POSTULATE– all computation is ‘equivalent’
TA
MO
DIA
20
02
organisations
• are politcial, social, economic ...
but are also ...
• information processing entities
so ...
TA
MO
DIA
20
02
the socio-organisational Church-Turing hypothesis
similarities to computers and cognition
• computational power
• computational structure
TA
MO
DIA
20
02
the organisation as a computer
• computer: program and data
• organisation: process and information
plus …
• computer data:LTM, STM, program counter
• organisation: ????files, papers … placeholders
TA
MO
DIA
20
02
parallels
computer cog sci organisation
process program proceduralmemory
processes
data data LTM fil es
placeholder programcounter
STM,activation
location ofartefacts
initiative interrupts,events
stimuli triggers
TA
MO
DIA
20
02
parallels
computer cog sci organisation
process program proceduralmemory
processes
data data LTM fil es
placeholder programcounter
STM,activation
location ofartefacts
initiative interrupts,events
stimuli triggers
interpretation ??? ??? people
TA
MO
DIA
20
02
embodied computation
TA
MO
DIA
20
02
ubiquity of compuation
the world is full of computation cognitive social economic
this helps us understand the world
TA
MO
DIA
20
02
physicality of compuation
computation happens in the world memory networks processors
this helps us understand computation
TA
MO
DIA
20
02
embodied computationsome examples
• computation is incremental• interaction not Magnus Magnusson
• pointers take space• where log space comes from
• space means time• o(N) memory takes 3√N time
• representation not information• importance of interpretation
TA
MO
DIA
20
02
embodied interaction
TA
MO
DIA
20
02
embodied interaction
• Paul Dourish’s term
• focus on:– tangible computing– social computing
• users create meaning
• designers give them the means
• where does task analysis fit in?
TA
MO
DIA
20
02
incidental interaction
TA
MO
DIA
20
02
incidental interaction
• traditional interaction – purposeful – user as controller– system as slave
• incidental interaction– user acts for one purpose– system observes and acts
TA
MO
DIA
20
02
incidental interaction
• traditional interaction – purposeful – user as controller, system as slave
• incidental interaction – system observes and acts, not user’s purpose
• examples:– car lights– auto-flush toilet– intelligent homes
TA
MO
DIA
20
02
? task analysis
• model main purposeful activity– use to design sensors
• model activity to be aided/enhanced– use to design actuators
TA
MO
DIA
20
02
winding up
TA
MO
DIA
20
02
final thoughts …
• ecologically valid task modelling– incorporate rich phenomena– but also understand limits
• role?– normative or normal?– definitive process or descriptive grammar– main task or subsidiary task