ITU Workshop on “Developments regarding
telecommunication network architectures
and services”
(Kampala, Uganda, 2 April 2012)
Socio-Economic Aware Design of
Future Network Technology
(Y.FNsocioeconomic)
Martin Waldburger,
University of Zurich,
Editor of Y.FNsocioeconomic in Q21/13,
waldburger@ifi.uzh.ch
Kampala, Uganda, 2 April 2012
Outline
•  SESERV Goal
•  Socio-economic Awareness
•  Design Goals and Objectives in Y.3001
•  Scope of Y.FNsocioeconomic
•  Proposed Structure
•  Tussle Evolution: Bandwidth Sharing
•  Tussle Analysis Meta-method
•  Methods Overview
•  Application Case
•  Summary and Conclusions
•  Q&A
2
SESERV Goal
•  Based on the fact that ICT contributes to social
inclusion, economic development, sustainability:
•  SESERV shall bridge the gap between
•  Especially socio-economic (SE) priorities and
•  Research objectives of European ICT projects in FP7
•  SESERV does offer a service that provides
•  An open access to SE and networking experts FISE Community
•  Investigation methodologies for relationships
Tussle
between Future Internet technology, society, and
Analysis
the economy
•  Knowledge and findings through research reports,
Event
workshops, and various research support on SE Organization
upon request
3
Socio-economic Awareness (1)
End-users
ISPs
Regulators
ASPs
…
Stakeholders with
varying socio-economic
interests
Out-of-network socioeconomic interactions
Technology choices
(including investments,
configurations)
Socio-economic
layer
Technology
layer
Technology outputs
(connectivity, QoS,
mobility,
security, etc.)
Network
applications
Network protocols
Links
Routers
Switches
Middle
boxes
Servers
Firewalls
Technology
components
3G towers
Socio-economic layer is governed by laws of socio-economics,
while technology layer by laws of physics
4
Socio-economic Awareness (2)
Stakeholders strategies/
policies with respect to
a specific technology
(functionality)
Dimension resources
Configure technology
Longer
Adaptation timescale
ISP
Adopt technology
Feedback
Internet SocioEconomic layer
Use technology
Shorter
tussle outcome
At each stage conflicts of interest (incentives) may arise at the
socio-economic layer.
The combination of actor strategies lead to a tussle outcome,
characterized by stakeholders benefits.
5
Socio-economic Awareness (3)
•  Traditional engineering goals (technology design)
•
•
•
•
Effectiveness
Efficiency
Modularity
Security
•  Technology will reach multiple stakeholders
•  Different interpretations of goals
Stakeholders •  Different incentives
engage in •  Different choices
tussles
•  Understanding socio-economic aspects
•  Complete view on ecosystem
•  Assess technology adoption and long-term success
6
Design Goals and Objectives in Y.3001
•  „Future Networks: Objectives and Design Goals“
... to reduce barriers to entry for the various actors involved •  Objectives
in the network ecosystem. •
•
•
•
Service awareness
... to reduce life cycle costs in order for them to be deployable and sustainable. Data awareness
... allow appropriate compe88on and an appropriate return Environmental awareness
for all actors Social and economic awareness
•  Design goals
•
•
•
•
•
•
•
•
•
•
•
•
Service diversity
Functional flexibility
Virtualization of resources
Data access
Energy consumption
Service universalization
Economic incentives
Network management
Mobility
Optimization
Identification
Reliability
FNs are recommended to be designed to provide a sustainable compe88on environment for solving tussles among the range of par8cipants in the ICT/
telecommunictaion ecosystem Ra8onale: Many technologies have failed to be deployed, flourish, or be sustainable because of inadequate or inappropriate decisions of the architect, ... Sufficient aHen8on therefore needs to be paid to economic and social aspects such as economic incen8ves in designing and implemen8ng the requirements, architecture, and protocol of FNs in order to provide a sustainable compe88on environment to the various par8cipants 7
Scope of Y.FNsocioeconomic
•  Y.3001 lists...
•  Candidate technologies
•  But no methods to achieve goals and objectives
•  „Socio-Economic Aware Design of Future
Network Technology“
This Recommenda8on lists methods to achieve socio-­‐economic design goals and objec8ves for Future Networks (FNs). When a candidate FN technology is provided, the methods listed provide a structured approach •  to an8cipate at technology design 8me the socio-­‐economic impact of the technology taking into account the relevant set of stakeholders, tussles emerging among them, and the range of available choices, •  to an8cipate either a stable and incen8ves-­‐compa8ble or an unstable outcome resul8ng from deploying the technology, •  to iden8fy poten8al spillover (unwanted) effects from the technology’s primary func8onality to another func8onality, •  and to help design technology for Future Networks that is in-­‐line with the respec8ve socio-­‐economic design goals and objec8ves. 8
Proposed Structure
•
•
•
•
•
•
•
•
Summary
Scope
References
Definitions
Abbreviations and acronyms
Conventions
Introduction
Socio-economic Aware Deployment of Future Network
Technology
•  Design for Tussle
•  Technology Deployment Cycle
•  Tussle Evolution
•
•
•
•
•
Tussle Analysis
Stakeholder Identification Methods
Tussle Identification Methods
Tussle Impact and Tussle Evolution Methods
Appendix: Methods Overview
Tussle concept
Meta-method
Methods to
implement steps
of tussle analysis
9
Tussle Evolution: Bandwidth Sharing
interac9ve users ? ? Func%onality: bandwidth sharing Regulator announces fines due to VoIP providers’ complaints ISPs Regulator p2p users get dispropor9onate bandwidth share ISPs use DPI technology to degrade quality of rival VoIP services ASPs Func%onality: VoIP service delivery Func9onality Stable outcome Evolves Ini9al state Unstable outcome Affects Tussle outcome ISP (neutral en9ty) p2p users configure applica9ons to open p2p applica9ons mul9ple TCP configured to connec9ons for the perform traffic same session ISPs thro2le obfusca9on bandwidth of p2p applica9ons by using What if ISPs deploy DPI technology. p2p conges9on exposure applica9ons technologies & mo9vate mul9ple conges9on pricing schemes? TCP connec9ons for the same session fair bandwidth sharing Tussle outcome Stakeholders’ strategies/policies Stakeholders’ strategies/policies peer-­‐to-­‐peer (p2p) users no discrimina9on interac9ve users get dispropor9onate bandwidth share 10
Tussle Analysis Meta-method
Func9onality I Func9onality II Step 1: Iden%fy all primary stakeholder roles and their characteris%cs for the func%onality under inves%ga%on new itera9on spillover Step 2: Iden%fy tussles among iden%fied stakeholders tussle tussle tussle tussle Step 3: For each tussle assess the impact to each stakeholder and poten%al spillovers 11
Methods Overview
Step 1: Stakeholder Iden%fica%on Step 2: Tussle Iden%fica%on Step 3: Tussle Impact and Tussle Evolu%on Interviews Highly relevant Relevant Relevant Personal observa%on Highly relevant Highly relevant Less relevant Role-­‐playing simula%on Relevant Highly relevant Highly relevant MACTOR method Prerequisite Highly relevant Relevant SWOT analysis Prerequisite Relevant Relevant Game theory Prerequisite Prerequisite Highly relevant Risk management Prerequisite Highly relevant Highly relevant System dynamics Prerequisite Highly relevant Prerequisite 12
Application Case (1)
•  Focus Group
•  During SESERV workshop in early 2012
•  Case of technology developed by SAIL project
•  Role-play simulation
•  Introduction into focus group format (moderator)
•  Technology presentation (project representative)
•  Stakeholder role assignment (10-15 participants)
•  Moderated tussle identification
•  Moderated tussle evolution
•  Transcription and analysis
13
Application Case (2)
!"#$%&'()*&+"#+'#,-+./")0/"%/'.$1%)
H"2$+<>K0JG+
>F;9DL=>F@+
0/"%$12')3425&'/",5%)6+"$7+5&$'2'%)
<==4,5+./")>2?24/=2'%)
?(/()$.+<>K0JG+
>F;9DL=>F@+ 3#F2)!*A)
R$/&#3&"#+
R"23(#+E+<>F;9DL=>F@+
R"23(#+M+<>F;9DL=>F@+
H'+"%,&)!*A)
*>@)A$B4,%C2'%)
0/"&2"&)>,%&',B$./")82&9/':%)
D2%2+'5C)A'/E25&%)
>,'25&/'()*2'?,52)A'/?,#2'%)
!"##$%&'()*+
,-"'(.$-/+
,-".4%$-/+
9:#$-/+
<==4,5+./")*2'?,52)A'/?,#2'%)
Outcome for assessed Future Network technology: An important interface was missing in the technology design! I'/:2'%)J6+':2&)A4+52)A'/?,#2'%K)
A'/72%%,/"+4%)
0/11$",5+./")A'/?,#2'%)
<1+&2$'%)
!"#)$#)+
9:#$-/+
0#1"-23&"#+
,-"'(.$-/+
L+1,"F)A'/?,#2'%)
M,"+"5,+4)*2'?,52)A'/?,#2'%)
!"&2'"2&)D2&+,42'%)
D2F$4+&/'%)
<#1,",%&'+./")<$&C/',.2%)
736(%("4/+<F;0J9LM@+
0#1-3/)-4%)4-$+
,-"'(.$-/+
,"6(%*+738$-/+
F-4/):"-)O*+<366+
A-"P$%)/@+
D2%2+'5C2'%)
*25$',&()<F2"5,2%)
D2%,#2".+4)
I$%,"2%%)
;$/"4-%$+,-"'(.$-+<;=>=;?90;@+
5A6"3.$-+<,B,CD=EFG+>799FHC0F@+
K%&'$+
;$63*+<!B,9I=;G+5J99,@+
;$Q4$/)"-+<366+A-"P$%)/@+
,3//('$+
Stakeholders’ strategies/policies CP Edge-­‐ISP Edge-­‐ISP delivers CP’s content over his localized CDN 5/$-/+
0/"%$12'%)
D/+12'%)
3"#G$%2'%)
82&9/':)0/1=/"2"&)
A'/?,#2'%)
82&9/':)3O5C+"F2)A/,"&%)
P+%&)6,42)A'/?,#2'%)
*2"%/')N=2'+&/'%)
>+':)M,B2')A'/?,#2')
;2"$2)/9"2'%))
L+&29+()A'/?,#2'%)
04/$#)N=2'+&/'%)
736(%("4/+<,5;>50FG+
;=>57=D=FG+F;0J9LMG+!HKDL=G+
5D0?=;>=JN@+
;$Q4$/)"-+<;=>=;?90;G+>K0J@+
>$-'$-+<366+A-"P$%)/@+
F-4/):"-)O*+<366+A-"P$%)/@+
CP iden9fies the degraded QoS and decides to announce IOs to other Name Resolu9on Servers, only Edge-­‐ISP is reluctant to update frequently the local caches to avoid increase of interconnec9on costs ? ? ? ? What if the Edge-­‐ISP and CP nego9ate about update frequency? Func9onality: AAA (Security) Edge-­‐ISP inter-­‐
connec9on costs increase and loss of revenues out of the content delivery CP sees degraded QoS and also may lose revenues (due to loss of customers) 14
Tussle outcome 82&9/':)34212"&);2"#/'%)
>"4-%$+
Summary and Conclusions
•  Engineers need to...
•  Be aware of socio-economic aspects of technology
•  Consider socio-economics in technology design
•  For the goal of...
•  Long-term success by incentive compatibility
•  Assessment of adoption potential
•  Sustainable competition environment
•  Recommendation Y.FNsocioeconomic
•  Methods to achieve socio-economic goals, objectives
•  Tussle analysis (meta-method)
•  Several methods to implement tussle analysis
15
Thank you for your attention!
Questions?
16