[PDF] - Mo dels of the System Ov erview Standard F ormalisms soft PDF Document

SLIDE 1 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (1) Mo dels

f

the System Ov erview Standard F

rmalisms

soft w are engineering notations used to sp ecify the required b eha viour

f

sp ecic in teractiv e systems In teraction Mo dels sp ecial purp

se

mathematical mo dels

f

in teractiv e systems, used to describ e usabilit y prop erties at a generic lev el Status/Ev en t Analysis an example

f

an engineering lev el metho d dra wing

n

b

th

formal mo delling and na

v

e psyc hology

SLIDE 2 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (2) Relationship with dialogue Dialogue mo delling is link ed to seman tics. System seman tics aects the dialogue structure. But the bias is dieren t. Rather than dictate what actions are legal, these formalisms tell what eac h action do es to the system.

SLIDE 3 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (3) Iron y Computers are inheren tly mathematical mac hines. Humans are not. F

rmal

tec hniques are w ell accepted for cognitiv e mo dels

f

the user and the dialogue (what the user should do). F

rmal

tec hniques are not y et w ell accepted for dictating what the system should do for the user!

SLIDE 4 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (4) General computational formalisms Standard soft w are engineering formalisms can b e used to sp ecify an in teractiv e system. Referred to as formal metho ds Mo del based describ e system states and

p

erations

Z,

VDM Algebraic describ e eects

f

sequences

f

actions

OBJ,

Larc h, A CT-ONE Extended logics describ e when things happ en and who is resp

nsible
temp
ral

and deon tic logics

SLIDE 5 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (5) The uses

f

SE formal notations F

r

comm unication

common

language

remo

v e am biguit y (p

ssibly)
succinct

and precise F

r

analysis

in

ternal consistency

external

consistency { with ev en tual program { with resp ect to requiremen ts (safet y , securit y , HCI)

sp

ecic v ersus generic

SLIDE 6 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (6) Mo del based formalisms Mathematical coun terparts to common programming constructs Programming Mathematics t yp es sets basic t yp es basic sets constructed t yp es constructed sets records unordered tuples lists sequences functions functions pro cedures relations

SLIDE 7 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (7) The mo del based metho d Example: a graphic dra wing pac k age

Line Rectangle Ellipse Unselect

P

in

ts are

rdered

pairs. P

int

== N

N

Shap es can b e

f

v arying t yp es. S hape ty pe == l inejel l ipsejr ectang l e

SLIDE 8 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (8) More t yp e denitions A graphic

b

ject is dened b y its shap e t yp e, width, heigh t, and cen tre p

sition.

S hape ty pe : S hape ty pe w idth; heig ht : N centr e : P t A collect io n

f

graphic

b

jects can b e iden tied b y a `lo

kup

dictionary' [I d] S hape dict == I d 7 ! S hape

SLIDE 9 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (9) Dening the state The system state con tains a dictionary

f

created

b

jects and a set

f

selected

b

jects. S tate shapes : S hape dict sel ection : P I d sel ection

dom

shapes Initiall y , there are no shap es in the dictionary . I nit S tate S tate shapes = fg

SLIDE 10 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (10) Dening

p

erations State c hange is represen ted as t w

copies
f

the state b efore | S tate after | S tate S tate S tate S tate The U nsel ect

p

eration deselects an y selected

b

jects. U nsel ect S tate sel ection = fg shapes = shapes

SLIDE 11 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (11) In terface issues F raming problem \ev erything else sta ys the same" can b e complicated with state in v arian ts In ternal consistency do

p

erations dene an y legal transition? External consistency m ust b e form ulated as theorems to pro v e Clear for renemen t, not so for requiremen ts Separation

f

system functionalit y and presen tation is not explici t

SLIDE 12 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (12) Algebraic notations Mo del based notations emphasise constructing an explici t represen tations

f

the system state. Algebraic notations pro vide

nly

implicit information ab

ut

the system state. Mo del based

p

erations are dened in terms

f

their eect

n

system comp

nen

ts. Algebraic

p

erations are dened in terms

f

their relationship with the

ther
p

erations.

SLIDE 13 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (13) Return to graphics example t yp es State, Pt

p

erations init : ! S tate mak e el l ipse : P t

S

tate ! S tate mov e : P t

S

tate ! S tate unsel ect : S tate ! S tate del ete : S tate ! S tate axioms for all st 2 S tate; p 2 P t

1.

del ete(mak e el l ipse(st)) = unsel ect(st) 2. unsel ect(unsel ect( st)) = unsel ect(st) 3. mov e(p; unsel ect(st)) = unsel ect(st)

SLIDE 14 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (14) Issues for algebraic notations Ease

f

use a dieren t w a y

f

thinking than traditional programming In ternal consistency are there an y axioms whic h con tradict

thers?

External consistency with resp ect to executable system less clear External consistency with resp ect to requiremen ts is made explicit and automation p

ssible

Completeness is ev ery

p

eration completely dened?

SLIDE 15 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (15) Extended logics Mo del based and algebraic notations mak e extended use

f

prop

sitional

and predicate logic. Prop

sitions

expressions made up

f

atomic terms p; q ; r ; : : : comp

sed

with (; ); ^ ; _; : ; ), etc. Predicates prop

sitions

with v ariables, e.g., p(x) and quan tied expressions 8 ; 9. These are not con v enien t for expressing time, resp

nsibili

t y and freedom, notions sometimes needed for HCI requiremen ts.

SLIDE 16 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (16) T emp

ral

logics Time considered as succession

f

ev en ts Basic

p

erators: alw a ys 2 2(G funnier than A) ev en tually 3 3(G understands A) nev er 2: 2: (rains in So. Cal.) Other b

unded
p

erators: p until q w eak er than 2 p bef

r

e q stronger than 3

SLIDE 17 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (17) Explicit time These temp

ral

logics do not explici tl y men tion time, so some requiremen ts cannot b e expressed. Activ e researc h area, but not so m uc h with HCI Gradual degradation more imp

rtan

t than time-critical i t y Myth

f

the innitely fast mac hine

SLIDE 18 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (18) Deon tic logics F

r

expressing resp

nsibilit

y ,

bligation

b et w een separate agen ts (e.g., the h uman, the

rganisation,

the computer) p ermission per

bligation
bl

F

r

example,

w

ns( Jane ; le `fred' ) ) per ( Jane; r eq uest(`prin t fred')) per f

r

ms( Jane; r eq uest(`prin t fred')) )

bl

( lp3 ; pr int(le `fred'))

SLIDE 19 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (19) Issues for extended logics Safet y prop erties stipulating that bad things do not happ en Liv eness prop erties stipulati ng that go

d

things do happ en Executabilit y v ersus expressiv eness easy to sp ecify imp

ssible

situations; dicult to express executable requiremen ts; settle for ev en tual executable Group issues and deon tics

bligations

for single-user systems ha v e p ersonal impact; for group w are, w e m ust consider implications for

ther

users.

SLIDE 20 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (20) In teraction mo dels General computational mo dels w ere not designed with the user in mind. W e need mo dels that sit b et w een the soft w are engineering formalism and

ur

understanding

f

HCI. formal the PIE mo del for expressing general in teractiv e prop erties to supp

rt

usabilit y informal in teractiv e arc hitectures (MV C, P A C, AL V) to motiv ate separation and mo dularisation

f

functionali t y and presen tation semi-formal status-ev en t analysis for viewing a slice

f

an in teractiv e system that spans sev eral la y ers

SLIDE 21 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (21) The PIE mo del A blac k-b

x

mo del

result display

P E R D

I result display

SLIDE 22 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (22) More formally [C ; E ; D ; R ] P == seq C I : P ! E displ ay : E ! D r esul t : E ! R Alternativ el y , w e can deriv e a state transition function from the PIE. doit : E

P

! E doit(I (p); q ) = I (p a q ) doit(doit(e; p); q ) = doit(e; p a q )

SLIDE 23 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (23) Expressing prop erties WYSIWYG What do es this really mean, and ho w can w e test pro duct X to see if it satises a claim that it is WYSIWYG? Limited scop e general prop erties whic h supp

rt

WYSIWYG. Observ abilit y what y

u

can tell ab

ut

the curren t state

f

the system from the displa y Predictabilit y what y

u

can tell ab

ut

the future b eha viour

SLIDE 24 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (24) Observ abilit y and predictabili t y Tw

p
ssible

in terpretations

f

WYSIWYG: What y

u

see is what y

u:
wil

l get at the prin ter

have

got in the system Predictabili t y is a sp ecial case

f
bserv

abilit y

SLIDE 25 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (25) F

rmally

P E R D

I result display

Determining result from displa y: 9 f : D ! R

8

e : E

f

(displ ay (e)) = r esul t(e) Determining eect from displa y: 9 f : D ! E

8

e 2 E

f

(displ ay (e)) = e

SLIDE 26 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (26) Relaxing the prop ert y

P E R O D

I result

bserve

9 f : O ! R

8

e 2 E

f

(obser v e(e)) = r esul t(e) 9 f : O ! E

8

e 2 E

f

(obser v e(e)) = e

SLIDE 27 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (27) Reac habilit y and undo Reac habilit y | getting from

ne

state to another. 8 e; e 2 E

9

p 2 P

doit(e;

p) = e T

w

eak Undo | reac habilit y applied b et w een curren t state and last state. 8 c 2 C

doit(e;

c a undo) = e Imp

ssible

except for v ery simple system with at most t w

states!

Better mo dels

f

undo treat it as a sp ecial command to a v

id

this problem

SLIDE 28 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (28) Issues for PIE prop erties Insucien t dene necessary but not sucien t prop erties for usabilit y . Generic can b e applied to an y system Pro

f
bligations

for system dened in SE formalism Scale ho w to pro v e man y prop erties

f

a large system Scop e limiting applicabili t y

f

certain prop erties Insigh t gained from abstraction is reusable

SLIDE 29 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (29) Status/ev en t analysis semi-formal tec hnique \engineering" lev el analysis based

n

formal mo dels uses na

v

e psyc hology clo c ks and calendars as example status { analogue w atc h face ev en t { an alarm

SLIDE 30 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (30) Prop erties

f

ev en ts status c hange ev en t

the

passing

f

a time actual and p erceiv ed ev en ts

usually

some gap p

lling
glance

at w atc h face

status

c hange b ecomes p erceiv ed ev en t gran ularit y

birthda

y { da ys

app
in

tmen t { min utes

SLIDE 31 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (31) Design implications actual/p erceiv ed lag: : : matc hes application timescale? to

slo

w

resp
nse

to ev en t to

late
e.g.,

p

w

er plan t emergency to

fast
in

terrupt more immediate task

e.g.,

sto c k lev el lo w

SLIDE 32 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (32) Na

v

e psyc hology Predict where the user is lo

king

mouse { when p

sitioning

insertion p

in

t { in termitten tly when t yping screen { if y

u're

luc ky Immediate ev en ts audible b ell { when in ro

m

(and hearing) p eripheral vision { mo v emen t

r

large c hange Closure lose atten tion (inc. mouse) concurren t activit y

SLIDE 33 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (33) Example { email in terface mail has arriv ed! timeline at eac h lev el

file system mailtool screen user

mail is received polls see file is changed change icon polls perceived event for user

P erceiv ed ev en t in min utes { not guaran teed alternativ e timescale explicit examination { hours/da ys audible b ell { seconds w an t min utes { guaran teed

SLIDE 34 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (34) Example { screen button widget (i) screen button

ften

missed, : : : but, error not noticed a common widget, a common error: Wh y? Closure mistak e lik ely { concurren t action not noticed { seman tic feedbac k missed Solution widget feedbac k for application ev en t a p er c eive d event for the user N.B., an exp ert slip { testing do esn't help

SLIDE 35 Human{Com puter In teraction, Pren tice Hall A. Dix, J. Finla y , G. Ab

wd

and R. Beale c

1993

Mo dels

f

the System Chapter 9 (35) Screen button widget (ii) a HIT

application dialogue screen user

depress mouse button

ver ‘delete’

highlight ‘delete’ release mouse button do delete remove highlight changes in text closure so NO perceived feedback

r

a MISS

application dialogue screen user

depress mouse button

ver ‘delete’

Mo dels of the System Ov erview Standard F ormalisms soft - - PDF document

Line Rectangle Ellipse Unselect

result display

P E R D

I result display

P E R D

I result display

P E R O D

I result

file system mailtool screen user

mail is received polls see file is changed change icon polls perceived event for user

application dialogue screen user

depress mouse button

highlight ‘delete’ release mouse button do delete remove highlight changes in text closure so NO perceived feedback

application dialogue screen user

depress mouse button

highlight ‘delete’ move off ‘delete’ remove highlight release mouse button no feedback