6. 13/03/14 pag. 6
7.1 A PROBLEM
Many of us have found ourselves with a
report that has to be completed by a
deadline, with the result (Figure 7.1) that
the dining room table, extended to its 12-
guest state, is covered by piles of paper as
well as reports, books, clippings and
slides; perhaps with more arranged on
the floor and on a couple of chairs.
There may even be piles on top of piles.
Such a presentation of vital information
makes a lot of sense: everything relevant
is to hand (hopefully!) and, moreover,
its very visibility acts as a reminder (Bolt,
1984, page 2) of what might be relevant
at any particular juncture, possibly
triggering a situated action (Suchman,
1987). In this environment I can
concentrate on creative tasks rather than
organisation.
Despite the availability of high-resolution
displays and powerful workstations I
still write most of my reports in this way.
Why? Because the display area provided
by the typical workstation is far too small
to support, visibly, all the sources that are
relevant to my composition.
7.2 THE PRESENTATION
PROBLEM
I am not alone in the sense of having too
much data to fit onto a small screen. A
very large and expensive screen, for
example, would be needed to display the
London Underground map in
sufficient detail(Figure 1.1), and it would
be difficult or impossible to present, on a
normal display, the complete
organisation chart of IBM or ICI.
Moreover, the recent emergence of small
and mobile information and
communication devices such as PDAs
and wearable displays has additionally
identified a pressing need for a solution
to the ‘ too much data, too little display
area’ problem: the presentation
problem. How can it be solved, mindful
of the need to support the activity of
visualising the underlying data?
7.2.1 Scrolling
An obvious solution is to scroll the data
into and out of the visible area. In other
words, to provide a means whereby a
long document can be moved past a
window until it reaches the required
‘page’ (Figure 7.2). This mechanism is
widely used, but carries with it many
penalties. One relates to the "Where am
I?" problem: I’m working on Chapter 2,
(it may be section 2.3, I don’t know)
and I want to remind myself of a figure
that is in chapter 5, it may be in section 5.3
– or was it 5.6? All I can do is operate the
scrolling mechanism and look out for
the figure I need, albeit assisted by
various cues such as the page number
indicated in the scrolling mechanism.
With a scrolling mechanism, most of a
document is hidden from view. I have
the same problem when using a
microfilm reader, with the additional
complication that if I move the tray to the
left, the image moves to the right. A
similar difficulty applies to my use of the
famous London ‘AtoZ’ street directory.
I’m driving along a road that goes off
the edge of the page, so I desperately
need whatever page contains the
continuation of that road (and quickly!).
Even if I get it, I will typically have
trouble locating the same road on the
new page. These and other similar
problems can be ameliorated by the
provision of context. Much of this
chapter, in fact, is concerned with
deciding how to provide context.
Scrolling
15. 13/03/14 pag. 15
Metaphor
illustra+ng
the
principle
of
the
Bifocal
Display
(a) An information space containing documents, emails, etc.
(b) The same space wrapped around two uprights.
(c) Appearance of the information space when
viewed from an appropriate direction
direction
of view
Distortion
18. 13/03/14 pag. 18
Bifocal display features
1. Distortion: available display area is allocated to two different regions
– Focus
(undistorted)
– Context
(distorted)
2. Information moves smoothly and continuously from context to focus
3. Display affords for representation
– opportunity
to
use
two
dimensions
– for
instance,
+me
assigned
to
horizontal
axis
– type
of
item
to
Y-‐axis
4. Main purpose
– Focus:
provide
detail
– Context:
awareness
and
iden6fica6on
5. Manual control
19. 13/03/14 pag. 19
What is the Bifocal Display Doing?
Transforming the information
space to the display space
7.19
Informa+on
space
Display
Space
Normal
display
Informa+on
space
Display
Space
Bifocal
display
context
focus
Slide
source:
Ken
Brodlie
20. 13/03/14 pag. 20
A
sequence
of
amino
acids
within
a
protein
Source:
Courtesy
of
Tom
Oldfield
Applications of distortion technique
28. 13/03/14 pag. 28
11Sun
12 Mon
13 Tue
14 Wed
15 Thur
16 Fri
17Sat
Fly LA
Kathy to airport Model Maker
Check slides, notes.
Family barbeque
Fly LHR Kathy to collect
Chapter 2/ see Dave March
JulyJuneMayAprilMar Aug Sept Oct
Flight to SFO
Tutorial set-up
Tutorial
United flight Heathrow
Pointer
Color OHs
Jane+John
Call Kathy
Combined X- and Y-distortion provides a
convenient calendar interface
29. 13/03/14 pag. 29
Visual
designer’s
sketch
of
the
applica+on
of
the
flip-‐zoom
technique
to
the
presenta+on
of
photographs
on
a
Nokia
mobile
phone
Source:
Courtesy
Ron
Bird
30. 13/03/14 pag. 30
Source:
Courtesy
David
Baar,
IDELIX
SoFware
Inc.
Distorted map on a PDA, showing the
continuity of transportation links
32. 13/03/14 pag. 32
Equal X- and Y-distortion centred around a manually
chosen location in the Macintosh OSX ‘dock’
33. 13/03/14 pag. 33
The Perspective Wall applies a 3D effect to the
bifocal display
34. 13/03/14 pag. 34
Advantages Perspective Wall
• User can adjust ratio of detail to context
• Smooth animation helps user perceive object constancy
• Relationship between detail and context is consistent: objects
bend around the corner
Slide
source:
Ken
Brodlie
35. 13/03/14 pag. 35
Perspective Wall
Perspective gives smoother transition from focus to context
Informa+on
space
Display
Space
Perspective
Wall
context
focus
Slide
source:
Ken
Brodlie
36. 13/03/14 pag. 36
overview
Space
limita+ons
• Scrolling
• Overview
+
detail
• Distor+on
• Suppression
• Zoom
and
pan
Time
limita+ons
• Rapid
serial
visual
presenta+on
• Eye-‐gaze
37. 13/03/14 pag. 37
Suppression
• Applies a distance function
and relevance function
• Less relevant other items are
dropped from the display
• Classic example: New
Yorker’s idea of the world
38. 13/03/14 pag. 38
Suppression
• Originally proposed by Furnas (1986), but many variations of
applications.
• Basic idea: more relevant information presented in great
detail; the less relevant information presented as an
abstraction.
• Relevance is computed on basis of the importance of
information elements and their distance to the focus.
39. 13/03/14 pag. 39
Degree of interest (DOI) function:
DOI(a|.=b)
=
API(a)
–
D(A,b)
• DOI(a|.=b):
DOI
of
a,
given
the
current
focus
is
b.
• API(a):
sta+c
global
a
priori
importance
measure.
• D(a,b):
distance
between
a
and
b.
44. 13/03/14 pag. 44
Each
node
in
the
organiza+on
tree
has
been
assigned
an
a
priori
importance
(API)
10
9 9
8
7 7
7
8 8
6
8 8
6
9
API
45. 13/03/14 pag. 45
Degree of Interest (DoI)
DoI = API – D
Expressed as a function of two quantities:
• A priori importance (API)
• Distance (D) between an item and the item currently in focus
46. 13/03/14 pag. 46
Segng
a
lower
limit
of
6
for
DoI
iden+fies
the
nodes
within
the
shaded
region
8
6 6
8
6 6
6
4 4
4
6 6
4
8
Focus
Context
Nodal values of degree of interest (=API – D)
48. 13/03/14 pag. 48
The
engineering
drawing
simplified
in
the
context
of
a
suspected
fault
Applications of DoI concept
49. 13/03/14 pag. 49
Illustra+ng
the
concept
of
a
magic
lens.
(a)
shows
a
conven+onal
map
of
an
area,
(b)
shows
the
loca+on
of
services
(gas,
water
and
electricity
pipes)
in
the
same
area,
and
(c)
a
(movable)
magic
lens
shows
services
in
an
area
of
interest,
in
context
Application in magic lens technique
51. 13/03/14 pag. 51
A
molecular
surface
of
the
protein
transferase
coloured
by
electrosta+c
poten+al
bound
to
DNA
shown
as
a
schema+c.
(ID
=
10mh).
The
magic
lens
window
allows
a
view
of
the
atomic
structure
bonding
to
be
shown,
with
the
bound
ligand
structure
highlighted
as
cylinders,
thereby
providing
a
view
inside
the
protein
Source:
By
kind
permission
of
Tom
Oldfield
and
Michael
Hartshorn
Magic lens
52. 13/03/14 pag. 52
A 3D Flexible and Tangible Magic Lens in
Augmented Reality
www.youtube.com/watch?v=PKegByAZ0kM
53. 13/03/14 pag. 53
A
combina+on
of
rubber-‐sheet
distor+on
and
suppression
lead
to
a
map
appropriate
to
a
journey
from
one
city
to
another
Combined distortion and suppression
55. 13/03/14 pag. 55
Historical note
• Distortion and suppression appeared in early 1980s
• Need to maintain a balanced view of focus + context
identified earlier – for example by Farrand (1973)
“an effective transformation must somehow maintain global
awareness while providing detail”
“… there is a need for presenting a display with 1. sufficient
detail for interaction, while 2. maintaining global vision of the
entire scene.”
56. 13/03/14 pag. 56
Fisheye view
• Farrand also coined the term “fisheye”
• Nowadays appears to refer to both distortion and suppression
57. 13/03/14 pag. 57
Fisheye Menus
• Here is the same idea applied to menus
– Ben
Bederson,
University
of
Maryland
• See also:
– hDp://www.cs.umd.edu/hcil/fisheyemenu/fisheyemenu-‐demo.shtml
ENV
2006
58. 13/03/14 pag. 58
Fisheye View, Polyfocal Display
Can
distort
boundaries
because
applied
radially
rather
than
x
y
1D
Fisheye
2D
Polyfocal
Slide
source:
Hornung
and
Zagreus
65. 13/03/14 pag. 65
Zooming
is
the
increasing
magnifica+on
of
a
decreasing
frac+on
of
an
image
(or
vice
versa)
Zoom and pan
66. 13/03/14 pag. 66
Zooming
• Conventional zooming-in
– No
change
in
data
or
representa+on
–
only
filtering
– Loss
of
context
• ≠distortion whose purpose is to permit focusing rather than filtering
• Supports two cognitive tasks (Cairns and Craft 2005)
– Zooming-‐in:
extraneous
informa+on
is
removed
from
visual
field
–
more
manageable
view
– Zooming-‐out:
reveals
hidden
informa+on
67. 13/03/14 pag. 67
A space-scale diagram relevant
to combined zooming and
panning
Furnas
and
Bederson
(1995)
72. 13/03/14 pag. 72
overview
Space
limita+ons
• Scrolling
• Overview
+
detail
• Distor+on
• Suppression
• Zoom
and
pan
Time
limita+ons
• Rapid
serial
visual
presenta+on
• Eye-‐gaze
73. 13/03/14 pag. 73
A
collec+on
of
images
is
presented,
one
at
a
+me,
at
a
rapid
rate
(e.g.,
ten
per
second)
time
Rapid serial visual presentation
74. 13/03/14 pag. 74
Tile mode: concurrent presentation of images
opposed to ‘slide show mode’
75. 13/03/14 pag. 75
‘Floa+ng
RSVP’
in
which
images
appear
to
approach
the
viewer
from
a
distance.
Sensi+ve
arrows
allow
the
speed
and
direc+on
of
‘movement’
to
be
controlled
by
a
user
Source:
Courtesy
Kent
WiNenburg
Floating RSVP
76. 13/03/14 pag. 76
The
contents
of
an
online
bookstore
are
presented
in
‘collage
mode’
RSVP,
simula+ng
the
placing
of
book
covers
on
a
table
in
sequence.
The
set
of
arrows
just
under
the
presenta+on
allows
control
of
the
speed
and
direc+on
of
presenta+on
Source:
Courtesy
Kent
WiNenburg
Collage mode
RSVP
77. 13/03/14 pag. 77
An
interface
facilita+ng
the
browsing
of
posters
adver+sing
videos.
Cursor
movement
along
the
stacks
causes
posters
to
briefly
‘pop
out’
sideways,
and
the
whole
bifocal
structure
can
be
scrolled
to
bring
a
video
of
interest
to
the
central
region,
where
a
mouse
click
will
cause
a
clip
from
a
video
to
be
played
(Lam
and
Pence
1997)
RSVP + bifocal principle
80. 13/03/14 pag. 80
An
experiment
to
test
a
subject’s
ability
to
recognise
the
presence
or
absence
of
a
previously
viewed
target
image
within
a
collec+on
presented
sequen+ally
at
a
rate
of
around
ten
per
sec.
Prior instruction
to subject
Subjectsʼ performance
“Here is a target
image. Tell me if
this image
appears in the
sequence of N
images youʼre
about to see”
Recognition
about 80% to
90% successful
time
about 100 ms
unrelated images
Presentation of images
Briefly glimpsed images
81. 13/03/14 pag. 81
Representa+on
of
limits
on
display
area
and
total
presenta+on
+me
by
a
‘resource
box’
Display area
Presentation
time
Space and time resources
82. 13/03/14 pag. 82
Source:
Courtesy
of
Katy
Cooper
Three ‘static’ image presentation modes (A, B, C) and
three ‘moving’ image presentation modes (D, E, F)
90. 13/03/14 pag. 90
The
accuracy
with
which
the
presence
or
absence
of
a
target
image
was
reported
for
the
six
presenta+on
modes,
averaged
over
all
tasks
and
presenta+on
+mes.
1.0
0.9
0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
Slide-show Mixed Tile Diagonal Ring Stream
Recognition
accuracy
Presentation modes
91. 13/03/14 pag. 91
The
(sta+c)
slide-‐show,
mixed
and
+le
image
presenta+on
modes
account
for
three-‐quarters
of
the
preferred
modes
(Cooper
et
al.
2006)
92. 13/03/14 pag. 92
Almost
all
the
least
preferred
image
presenta+on
modes
were
moving
modes
and
the
stream
mode
accounted
for
over
half
93. 13/03/14 pag. 93
A
simple
representa+on
of
eye-‐gaze
behaviour.
The
rapid
saccades
are
shown
green,
the
fixa+ons
(F)
of
varying
dura+on
by
circles
of
propor+onate
size
F
F
F
F
F
F F
F
Eye-gaze
100. 13/03/14 pag. 100
The
acquisi+on
of
an
expanding
target.
(a)
The
dormant
appearance
of
the
image
collec+on,
and
(b)
its
appearance
when
the
cursor
rests
over
image
6
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
10 11 12 13 14 15 16 17 18 19 209861 2 3 4
(a)
(b)
Manual control: ‘expanding target’
presentation mode
101. 13/03/14 pag. 101
An
experiment
in
which
a
subject
first
views
the
rapid
(e.g.,
10
per
second)
presenta+on
of
a
collec+on
of
images,
is
then
shown
a
single
image
and
asked
to
say
whether
that
image
was
part
of
the
collec+on.
Iden+fica+on
success
is
highly
dependent
upon
the
+me
elapsing
between
the
end
of
the
presenta+on
and
the
ques+oning
of
the
subject
Prior instruction
to subject
Presentation of image collection Subject’s performance
about 100ms
unrelated images
time
None
The subject was shown an image and then
asked, ‘Was this image present in the
sequence you have just seen?’
Recognition success was 10% to 20%
unless the question was aksed within about
4 seconds of the end of the presentation
Models of human visual performance
102. 13/03/14 pag. 102
An
experiment
in
which
a
collec+on
of
images
is
presented
to
a
subject.
Each
image
is
presented
briefly
(e.g.,
for
100ms)
and
followed
by
a
‘visual
mask’
las+ng
about
300ms.
Subjects
were
able
to
say,
with
a
considerable
degree
of
success,
whether
an
image
shown
arerwards
had
been
part
of
the
presenta+on
Prior
instruc+on
to
subject
Presenta+on
of
image
collec+on
Subject’s
performance
about
300ms
unrelated
images
+me
None
The
subject
was
shown
an
image
and
then
asked,
‘Was
this
image
present
in
the
sequence
you
have
just
seen?’
Up
to
92%
recogni+on
success
etc
.
.
.
.
Visual
mask
.
Visual
mask
Visual
mask
about
100ms
Models of human visual performance
103. 13/03/14 pag. 103
A
third
palleDe
for
the
interac+on
designer,
addressing
issues
of
presenta+on
Presentation
concepts and
techniques
Scrolling
Overview+detail
Distortion
Suppression
Zoom
Pan
RSVP
Eye gaze
Recap
105. 13/03/14 pag. 105
Shneiderman’s “7 Tasks”
• Overview task
– overview of entire collection
• Zoom task
– zoom in on items of interest
• Filter task
– filter out uninteresting items
• Details-on-demand task
– select an item or group to get details
• Relate
task
– relate
items
or
groups
within
the
collec+on
• History
task
– keep
a
history
of
ac+ons
to
support
undo,
replay,
and
progressive
refinement
• Extract
task
– allow
extrac+on
of
sub-‐collec+ons
and
of
the
query
parameters
108. 13/03/14 pag. 108
References
• Furnas, G. W., & Bederson, B. B. (1995, May). Space-scale
diagrams: Understanding multiscale interfaces. In
Proceedings of the SIGCHI conference on Human factors in
computing systems (pp. 234-241). ACM Press/Addison-
Wesley Publishing Co..
• Shneiderman, B. (1996, September). The eyes have it: A task
by data type taxonomy for information visualizations. In
Visual Languages, 1996. Proceedings., IEEE Symposium on
(pp. 336-343). IEEE.
• Some relevant notes:
http://jcsites.juniata.edu/faculty/rhodes/ida/
presentation.html
110. 13/03/14 pag. 110
Research presentations
• Schedule on PointCarré
• Select a second paper in the same slot for questions: e.g.
session 1: http://doodle.com/rmpc9g8u3p2qzsy4
• Links to doodle polls for all six sessions will be included in the
schedule.
112. 13/03/14 pag. 112
Team project milestones
1. Form teams
2. Project proposal
3. Intermediate presentation
4. Final presentation
5. Short report
due
27
Feb.
due
13
March
due
3
April
22
May
due
29
May
113. 13/03/14 pag. 113
Project proposal
1 page description of your intended project:
– mo+va+on
– which
datasets
you
will
use
– current
status.
If
available,
first
designs.
– problems/ques+ons
due 13 March
If you want earlier feedback, send us your proposal earlier ;-)
114. 13/03/14 pag. 114
Data collection
• https://docs.google.com/forms/d/
1gHwVWHZLzWdSz1F37jA1Gungrl56bT215M6FYW3YqGY/
viewform
Or
• bit.ly/N6JTyD
Anonymous! Choose your own ID.
• Please report your data ;-)