13/08/2011

1

Participatory Modeling Methodology: Tightening the GeoWeb Design Loop

Tim NyergesDepartment of Geography, University of Washington

Seattle, Washington USA

Piotr JankowskiDepartment of Geography, San Diego State University

San Diego, California USA

Brent HallSchool of Surveying, University of Otago

Dunedin, New Zealand

ICA-ISPRS WorkshopsSimon Fraser University

August 10, 2011

Outline1. GeoWeb as convergence of three realms

• Spatial Data Infrastructure, Online Participatory GIS, Volunteer Geographic InformationVolunteer Geographic Information

• A Synthesis of Values for Design2. Coupling architectures for the GeoWeb

• Service-oriented architecture3. Participatory Modeling Methodology

• System development, use, and evaluation• Tighten the design loop

4. Conclusions and Directions• Results of tightening the design loop• A fourth realm? CyberGIS high performance computing

13/08/2011

2

1. GeoWeb Realmsof GIScience & Technology

• Spatial Data Infrastructure (circa 1990)• Online Participatory GIS (circa 1999)• Volunteer Geographic Information (circa 2005)

Look at a brief overview of these realms…

Motivation within realms

• SDI - increase data sharing• OPGIS - broaden public governance and

community voice• VGI – enhance distributed information collection

All involve values, interests, and concerns, but perhaps from different perspectives…

13/08/2011

3

US SDI

Spatial Data Infrastructure, e.g. three levels• 12 Federal Agencies – geoplatform.gov• 50 States (National States Geographic

Information Councils)• Regional (e.g. Washington State

Geographic Data Archive)Geographic Data Archive)

Data.gov

13/08/2011

4

http://www.naco.org/meetings/participate/Presentations/Federal%20Geographic%20Data%20Committee%20and%20National%20Geospatial%20Advisory%20Committee.pdf

US Federal SDI – current architecture

xter

nal U

sers Lim

ited Ge

Dow

nload & L

Service Acc

ExIn

fras

truc

ture Shareable

eo Local

cess

Prog

ram

s Localized

Jerry Johnston, US EPA, presentation to NGAC “Status Update: Geospatial Platform”http://www.fgdc.gov/ngac/meetings/march-2011/intergovernmental-subcommittee-update.pptx

13/08/2011

5

US Federal SDI – next architecture

Developers

PlatformManager

Service Management Configuration ControlSecurity / Identity MgmtRelease ManagementCapacity PlanningG bl d S h

Access:Advanced Geo Analytical FunctionalityGeo Map and Feature ServicesFl ibl D t D li

Users

Discovery Use Platform Supplier

Data.gov Geospatial Capability – With Geo Platform

Developers Capacity PlanningPerformanceService Desk

Contract Management VehiclesCloud SuppliersSoftware Suppliers

Service ManagementService Level AgreementsCost Benefit AnalysisProduct & Service Catalog

Data.gov Geospatial Platform Platform ManagementSupport Services

PaaSSoftware Components

Software Libraries

Middleware

Geo Applications Geo-Tools Geo

Services

SaaS

ComputingPower

Data Stores

IaaS

Aware of

Aware of

Server

Data

Catalog as a Service

Geo enabled Searchand Visualization

Flexible Data DeliveryImage ProcessingGeo-processingMetadata Editing / ManagementCoordinate Transforms Mashup / Meshup

From SaaS

Abide by Contracts & Service Agreements

Data Administrators Portfolio &Investment Managers

IT Managers

Supply Data

CoordinateIaaS

Developers

Develop in PaaS

Evaluate &Assess

CIO / GIO

Manage

Server

Data

Data Management Systems Management Portfolio Management

Supply Metadata

Jerry Johnston, US EPA, presentation to NGAC “Status Update: Geospatial Platform”http://www.fgdc.gov/ngac/meetings/march-2011/intergovernmental-subcommittee-update.pptx

US – States Level(National State Geographic Information Councils)

• GeoSpatial One-Stop Nodes• Metadata only

13/08/2011

6

Regional SDIWashington State Geospatial Data Archive

(WAGDA) 1.0

Acquiring Data – Data Access

• Data Access– Multiple Services (currently in development)

Function Direct Connection

Geodata Service

Image Service

Web Feature

Service/Web Coverage Service

Web Mapping Service

Geoportal

Fast data view ● ○ ○Remote data analysis ● ● ● ●Complete and ready metadata ● ● ● ●

● supported - ○ unverified - (blank) not supported - (grey) not preferred

Complete and ready metadata ● ● ● ●Geodatabase versions ● ○ ○ ○Exportable data ● ● ● ● ●Interoperability ● ● ● ●Modifiable access permission ● ● ● ● ● ○Replication/Editing ● ● ● ●

12

13/08/2011

7

Regional SDI WAGDA 2.0

• GeoPortal – wagda.lib.washington.edu/geoportal

WAGDA 2.0 Architecture

14

13/08/2011

8

OPGIS

• Online Participatory GIS• Widespread growth since 1999

OPGIS Niches

13/08/2011

9

Support provided by the National Science Foundation Grant No. EIA 0325916, funded through the Information Technology Research Program, and managed in the Digital Government Program.

OPGIS example

VGI

• Volunteer Geographic Information• Wide array of developments since 2005

13/08/2011

10

VGIMashups

d

SMS

and Real-time feeds

VGI - balloon sensor launch

13/08/2011

11

Balloon trajectory

Volunteer Geographic Informationaccuracy comparison of geocoded addresses

13/08/2011

12

Synthesis across SDI-OPGIS-VGIWhat is being valued?

• SDI – people interested in efficient, effective, and equitable access to dataeffective, and equitable access to data

• OPGIS – people interested in stakeholder public values, goals, and concerns

• VGI – people interested in personal values, goals, concerns about what is important

Is their a convergence of “value” in light of geospatial information technology?

Shared interests?Common values?

All realms…E ti i t• Engage participants

• Enable participants• Structure participation

…as human-computer-human interaction…for data, information, evidence, and

knowledge production on the GeoWeb

13/08/2011

13

Coupling SDI-OPGIS-VGIS enables distributed and participatory GeoDesign

• GeoDesign – geography by design (Steinitz 2011)• For example a regional stormwater runoffFor example, a regional stormwater runoff

monitoring network to better understand and act upon non-point source pollution

• GeoDesign can enable large-scale participatory monitoring designs using GIS

• GeoDesign as sustainability management i SDI OPGIS d VGIrequires SDI, OPGIS, and VGI resources

configured into regional enterprise GIS also called consortium GIS

2. Coupling SDI-OPGIS-VGIS

• What architecture design(s) might help us couple SDI OPGIS VGI resources?SDI-OPGIS-VGI resources?

• Service-oriented architecture (SOA) is an approach proposed by many

• SOA – a collection of protocols and components for integrating service resourcesSOA i f t t t• SOA infrastructure connects…– Consumer services: User interface clients– Producer services: Data and software capabilities

13/08/2011

14

Esri example of integration platform

http://www.esri.com/library/whitepapers/pdfs/geospatial-soa.pdf

Expansion of SOA Infrastructure, by example,

for coupling SDI-OPGIS-VGI

http://www.esri.com/library/whitepapers/pdfs/geospatial-soa.pdf

13/08/2011

15

What methodology Helps Guide Coupling of SDI-OPGIS-VGIS Resources?

• Coupling among three technology realms is a complex choice problema complex choice problem

• Need systematic approach to help guide coupling strategies

• Participatory Modeling Methodology brings diverse perspectives into playdiverse perspectives into play

3. Participatory Modeling Methodology(PMM)

• Approach that links system development (information technology) system use (complex(information technology), system use (complex problems), and system evaluation (social-behavioral science) for improving outcomes from systems.

• One application: participatory interaction modeling explores the science of interaction design

• Participatory interaction designs are intended to• Participatory interaction designs are intended to structure participation, while incorporating motivation and information technology

13/08/2011

16

What, Why, and How of Component Coupling?

• What resources are to be coupled?

• Why are resources to be couple?

• How are resources to be coupled?

PMM Framework for Participatory Interaction Modeling

1Cyber-System D l t

What coupling is needed?

Monitor event logs of online

activity

Improve development

artifacts

Development

Why is coupling needed? How is coupling performing?

Create and calibrate activity

models

3 Cyber-System

Evaluation

2Cyber-System

Use

three roles in human-computer-human interaction:(1) developer, (2) user, (3) analyst

13/08/2011

17

Three domains anchor the GeoWeb design loop

1) Cyber-Systems developmentincorporate participatory technologies

2) Cyber-Systems use by participants address pervasive complex problems

3) Cyber-Systems evaluation involve social behavioral studies to improvesocial-behavioral studies to improve designs

Workflow Manager System (Design) Development

13/08/2011

18

Workflow Manager System Design - Class Diagram

LIT System Use – workflow agenda interface

13/08/2011

19

LIT System Use: Review project packages

Client-server interaction “event”

Server locationUW

Time

System (HCHI)

Evaluation

User’s home locationor home zip code

UW

Space

System log: Client-server interaction “events” captured during decision makingoccur in time and space

13/08/2011

20

Entire Timeline:245 registered users

120,396 client-server “events” One Day:81 users7,272 “events”

Spatio-temporal Timeline: Three Counties around Seattle, WA, October 11 – November 25, 2007

October 18 – October 23, 2007

What it suggests: Evolving structure of an analytic-deliberative network constructed from client-server interaction “events”

HCHI Evaluation

75%

100%

Percent of Quota Subjects Still Actively Participating

Oct

ober

15,

200

7

Oct

ober

22,

200

7

Oct

ober

25,

200

7

Nov

embe

r 1,

200

7

Nov

embe

r 7,

200

7

Nov

embe

r 13

, 200

7 30 minutesSub-Step 1b and 1c

Sub Step 3c and 4a

25%

50%

Tim

e (M

Inut

es)

15 minutes

Sub-Step 5a

Sub-Step 3c and 4a

0% 1a 1b 1c 2a 2b 3a 3b 3c 4a 4b 5a 5b

LIT Sub-Step

Average Time Quota Subjects Spent Online

Evaluation: Participant activity contained peaks and valleys of intensity

13/08/2011

21

Roles for Tightening the Participatory Design Loop

among the Three DomainsArtifact Developer User AnalystArtifact Abstraction Level

DeveloperRole

User Role

Analyst Role

Use Case Narrative X X XClass Diagram X X XActivity Model X X XArchitecture Diagram X X XMock up X X X

X = Produced by X = Consumed by

Mock-up X X XPrototype X X XEvent Log X X X

Design Loop Outcomes

Tightening the GeoWeb design loop improves t d t di b toutcomes…understanding about:

• Participatory GeoWeb technology design• Effectiveness of complex problem solving• Participatory workflow patterns of knowledge

productionproduction

13/08/2011

22

Research Focus PMM on GeoWeb

PMM - linking system development, use and evaluation…on theGeoWeb - coupling of SDI, OPGIS, and VGI…enables many research directions, but two in

particular…1) S li ti i ti1) Scaling participation2) Spatio-temporal modeling for GeoDesign3) CyberGIS to support those 2

1. Scaling participation

a) down-up - scaling up moves the research toward topics about regional placestoward topics about regional places

b) in-out - scaling out moves the research toward very large numbers of people participating

c) low-high - scaling high moves the research toward nuanced analyses and rich deliberative processesprocesses

How can these dimensions help us understand convergence of SDI, OPGIS, and VGI?

13/08/2011

23

2. Spatio-temporal Modeling for GeoDesign

For example, couple Regional SDI, OPGIS, and VGI to enable GeoDesign contributions for stormwater runoff monitoring of water quality

3. CyberGIS to support directions…

CyberGIS – a forth realmf f f• a fundamentally new software framework

comprising a seamless integration of cyberinfrastructure, GIS, and spatial analysis/modeling capabilities using “services approach”, and

• promises widespread scientific breakthroughs p p gand broad societal impacts due to new level of performance for intense computational problems.

13/08/2011

24

4. Conclusions• Cyberinfrastructure-enabled participatory

GeoWeb will continue to grow, connecting g gpeople with similar interests more than ever before

• Need systematic and robust, but at the same time comprehensive and flexible, framework for participatory systems design p p y y gto guide that growth

• Synergistic activity among SDI, OPGIS, and VGI developments can help formulate consortium GIS for regional governance

Conclusions - 2

• Synthesize comparative architectures that l d t i d d i f G W blead to improved designs of GeoWeb solutions; a need for computer-enabled case study synthesis

• Tightening the design loop will require metrics for development, use and evaluationmetrics for development, use and evaluation

• Enabling participatory geospatial thinking, learning and decision making using cyber-enabled tools will require nuanced metrics.

13/08/2011

25

Acknowledgements

• Much of this presentation was prepared while Nyerges and Jankowski were on bb ti l t th U i it f Ot D di N Z l d (S i 2011)

Thank you!Comments / Questions?

sabbatical at the University of Otago, Dunedin, New Zealand (Spring 2011) sponsored by a William Evans Fellowship and hosted by Brent Hall.

• LIT research was supported by National Science Foundation Grant No. EIA 0325916, funded through the Information Technology Research Program, and managed in the Digital Government Program.

• PIM research is funded by the National Science Foundation Grant No. BCS-0921688, Geography and Spatial Sciences Program

• CyberGIS research is funded by National Science Foundation Grant No. OCI-1047916, Office of Cyberinfrastructure, Software Institutes, Cross-Directorate Active Programs, Geography and Spatial Sciences, Method, Measure & Statistics.

• Presentation authors are responsible for any and all errors.