From Pace Layering to Resilience Theory

1
From Pace Layering to Resilience Theory

• The complex implications of taggingfor
  information architecture
• D. Grant Campbell Karl V. FastFaculty of
  Information Media StudiesUniversity of Western
  OntarioInformation Architecture Summit
• March 25-27, 2006. Vancouver, B.C.

2
William Dampier, 1652-1715

• Naturalist
• Explorer
• Travel Writer
• Sea Captain
• Hydrographer
• Circumnavigator
• and
• Pirate

3
Two views of tagging
Tagging is Progress A natural and logical
evolutionary step forward in our search for
better and more sophisticated ways of generating
useful and usable information structures. A
radical break with previous categorization
strategies, rather than an extension of them
Shirky
Tagging is Regress A step backwards, because tag
structures are not as rich or sophisticated as
traditional structures, such as CVs and thesauri
and faceted classification schemes A reinvention
of the wheel, because librarians have long used
free-form indexing methods and the benefits and
problems are well known.
4
Tagging contradicts central message of IA
5
pace layering(stewart brand)
6
complex systems
7
decomposed into multiple layers
8
layers change
9
at different rates
10
fast layers
learnproposeabsorb shocks
11
slow layers
rememberintegrateconstrain
12
fast layers
get all the attention
13
slow layers
have all the power
14
Pace Layering
Fast Layers learn propose absorb shocks get all
the attention
Slow Layers remember integrate constrain have
all the power
at least, thats the theory
15
Pace layering of buildings
Brand, S. (1994). How Buildings Learn.
16
Pace layering of civilization

• Fashion
• Commerce
• Infrastructure
• Governance
• Culture
• Nature

Brand, S. (1999). Clock of the Long Now.
17
The Iceberg of IA (Morville)
18
Elements of user experience
Garrett, J. J. (2003). The elements of user
experience.
19
Pace layering for architectural components
Morville, P. (2005). Ambient Findability.
20
IA frameworks suggest pace layering
Influence of pace layering may or may not be
explicitly acknowledged, but there is at least an
implicit influence.
21
Pace layering for architectural components
Morville, P. (2005). Ambient Findability.
22
Pace layering in information architecture

• Over time, the lessons learned at the top are
  passed down, embedded in the more enduring layers
  of social and semantic infrastructure
• Ambient Findability, p. 141

23
in other words
24
information architectscan continue doingwhat
they have always done
25
architectural work that isslower, deeper, richer
26
meanwhile.
27
tagging canspin madly on the surface
28
anything worth keeping will
29
slowlynaturallyinexorably seepdownto
thelower layers
30
while at the same time,anything not worth
keeping will be thrown away
31
thats what pace layering says will happen
Fast Layers learn propose absorb shocks get all
the attention
Slow Layers remember integrate constrain have
all the power
32
what have we just done?
33
we had two views of tagging
Tagging is Progress A natural and logical
evolutionary step forward in our search for
better and more sophisticated ways of generating
useful and usable information structures. A
radical break with previous categorization
strategies, rather than an extension of them
Shirky
Tagging is Regress A step backwards, because tag
structures are not as rich or sophisticated as
traditional structures, such as CVs and thesauri
and faceted classification schemes A reinvention
of the wheel, because librarians have long used
free-form indexing methods and the benefits and
problems are well known.
34
tagging isprogress
Choose oneor the other,but not both.
tagging isregress
35
tagging isprogress
pacelayering
integratedholisticworldview
tagging isregress
Replaced two opposing and fractious
conceptualizations of tagging in information
systems with a single, integrated, and holistic
conceptualization.
36
added tagging to ourconceptual framework
37
without upsetting our comfortable worldview
38
in this new frameworkIA still hasenormous value
39
CVs and hierarchy and facetsare stillricher and
deeperand more desirableforms of structure
40
that is, better than tags
41
tags works independentlyof other structures
42
yet deeper structures reap benefitsover time
43
be patient
44
are we missing something?
45
Pace layering of buildings
Brand, S. (1994). How Buildings Learn.
Because of the different rates of change of its
components, a building is always tearing itself
apart. (Brand, pg. 13)
46
frictiontensioninteraction between layers
47
is this important?
48
tagging based on disregard for

• Skilled cataloguers and indexers
• Group consensus about desired goal states

49
Disregard is based on several assumptions

• If enough people do what they want, useful
  patterns and structures will emerge
• The patterns and structures get more useful and
  interesting as more people do what they like
• They will also be different and more useful than
  those produced by experts or guided by group
  consensus
• Systems that exploit this can scale in ways that
  traditional systems, like library catalogues,
  cannot
• Systems with fewer rules and constraints are more
  likely to gain widespread adoption and more
  likely to generate useful and usable patterns and
  structures

50
if you let people tag as they wish
the information structureswill only and always
increase insharpness,utility,and useful
complexity
51
laissez-faire librarianship
52
the tagging discourseimplicitly relies
ontheories ofcomplexity and emergence
53
complexity theoryis the study ofcomplex,
self-organizing,adaptive systems.
Waldrop, M. (1992). Complexity.
54
emergent systemssolve problemsby drawing on
massesof stupid elements,rather than asingle,
intelligentexecutive branch.
Johnson, S. (2001). Emergence.
55
is this approach hostile to traditional forms of
structureand traditional ways of generating
those structures?
56
in other wordsis it hostile to the practice of
information architecture?
57
Question One
Between layers in a complex systemsis
therefrictiontensioninteraction?
58
Question Two
Are there positive benefitsto thisfrictiontensi
oninteraction?
59
complexity theorysaysyes and maybe
60
YES!there are hostile relations between layers
in a complex systems
61
MAYBE!these hostile relationsare
beneficial,but only under certain circumstances
62
if conditions are good,system will be
healthyand changes in fast layerswill have
positive effects
63
but if conditions are poor,system will
deteriorateand changes in fast layersmay have
negative effects
64
?
tagging isprogress
pacelayering
integratedholisticworldview
tagging isregress
65
what about ecology?
ecology and environmental studieshave used pace
layering tomodel interactions over time
66
do you meaninformation ecologies?
interaction of playersoverstratification of
layers
67
asmetaphor
ecology
conceptualizing relationships between people and
artifacts in an information system
68
slowlynaturallyinexorably seepdownto
thelower layers
if true, we should treat ecology as a concrete
phenomenon, not metaphor
69
ecology research pace layering resilience
theory
70
resilience theory
explain role of changein complex adaptive systems
Holling et. al. (2002). In quest of a theory of
adaptive change.
71
four assumptionsof resilience theory
72
assumption onechange is neither continuousnor
chaotic
73
assumption twochange isdiscontinuouspatchyno
n-linear
74
assumption threedestabilizing forcesjust as
importantas stabilizing forces
75
assumption fourconstant yieldsare an
indication of false stability
76
Four basic assumptions of resilience theory

• Change is neither continuous nor chaotic
• Change is discontinuous, patchy, and non-linear
• Destabilizing forces as important as stabilizing
  forces
• Constant yields indicate false stability

77
so what does this mean?
78
1.tagging may have little impact for a long
time, then suddenly cause significant changes
79
2.even if tagging is a disruptive force, it
will have some positive effects, increasing
diversity and flexibilityand resilience of the
system
80
3.if we ignore tagging,we do so at our peril
81
4.if we are wrong about 1, 2, and 3 then our
architectures are much less complex,
sophisticated, and resilient than we think
82
so.we are using complexityto mean difficult
83
we are not dealing with trulycomplex systems
84
at least, not yet
85
if thats the case.
86
ecology is only a metaphor
87
pace layering may be aweak toolfor
understanding IA
88
butif we believe our architecturesshould be
complex systemswe need new techniquesBEYOND
TAGGING
89
ecology is not rocket science
it is much harder
90
two basic responses

• Reaffirm our position as guardians of a systems
  architectural stability and utility
• Reposition ourselves as designers of a systems
  ecological resilience

91
user-centered research

• Longitudinal and observational studies of how
  users create, manage, and interact with tagging
  structures
• Prototypes that allow for novel ways of
  exploring, navigating, representing, and
  manipulating tagging structures

92
system-centered research

• Identifying system metrics
• Modeling for prediction and forecasting
• Simulating of environmental changes

93
(No Transcript)
94
tagging is not justa mechanismfor
creatinginformation structures
95
tagging is not justanother toolin the IA toolbox
96
tagging is alsoa tool for reconceptualizing
information architecture
97
tagging isprogress
tagging isregress
98
tagging isprogress
pacelayering
integratedholisticworldview
tagging isregress
99
tagging isprogress
ecology asmetaphor
complexitytheory
pacelayering
?
resiliencetheory
tagging isregress
100
From Pace Layering to Resilience Theory

• The complex implications of taggingfor
  information architecture

101
Contact Information

• D. Grant Campbell Karl V. Fast
• Faculty of Information Media Studies
• University of Western Ontario
• London, Ontario, Canada
• gcampbel_at_uwo.ca
• kfast_at_uwo.ca
• This research was supported by the
• Social Science and Humanities Research Council
  (SSHRC)
• and the
• Faculty of Information Media Studies at the
• University of Western Ontario