Local Spatial Data Infrastrutures Based on a Service-Oriented Architecture

1. Local Spatial Data InfrastruturesBased on a Service-Oriented Architecture Clodoveu Davis Leonardo Lacerda Alves PUC Minas

2. Introduction • There is demand for GIS in large computational environments, with a large number of users, spread over numerous locations, and high volumes of data • Factors: • The fast development of networking technology and the Internet • The success of Web-based GIS • The wide applicability of mobile and ubiquitous computing

3. Introduction • Many challenges arise when someone tries to share (geographic) information: • Data format/data structure issues • Data quality issues • Content issues • Semantics

4. Introduction • Early solutions included • Data translation tools • Offline data replication • “Negotiated” semantics through (poor) metadata • This led to intensive data redundancy and numerous updating/synchronization problems

5. Introduction • Data sharing and exchange among several organizations is hard to achieve • It involves at least • A lot of political negotiation • Agreements on standards • Agreements on costs and cost sharing • Agreements on maintenance • ....

6. Introduction • The impact of such data exchange and sharing efforts is very positive • We need technology that can make such efforts pay off • Easier data access • Semantics • Less maintenance and synchronization issues • Semantics • Data usage directly from the source, without translation or replication • More applications • Free access to base maps • Public services • Value-added services

7. Our proposal • Extension of the concept of Spatial Data Infrastructures (SDI) to local data and applications • Definition of the architecture and creation of a local, intra-organizational, service-oriented local SDI • Most SDIs are national or regional in scope • Service-oriented architectures help with interoperability issues

8. A Brief History of GeoData Sharing • Spatial data transfer standards • Spatial data clearinghouses • Spatial data infrastructures • Geoportals

9. Early GIS • Dataset creation was really expensive and complex • Redundant efforts were commonplace • Potential data providers were late adopters of GIS technology • Incipient market of data conversion services

10. Spatial Data Transfer Standards • Only helpful with syntactic problems • Data sharing requires • An offline semantics agreement, plus • Select-export-copy-translate-import, plus • Some luck • There are a number of proposed standards, but none has achieved widespread acceptance

11. Spatial Data Clearinghouses • SD Clearinghouses have been described as sites through which a number of services related to spatial data can be accessed • The emphasis on services is recent: this concept has been initially implemented as a means to obtain “off-the-shelf” data

12. Spatial Data Clearinghouses • A 2004 study on national SDCs showed that there is a growing dissatisfaction from users as to the functional capabilities available • The study indicates that the focus should be user- or application-oriented, instead of data-oriented

13. Spatial Data Infrastructures • Evolution from the clearinghouse perspective • “Infrastructure” implies that there should be some sort of coordination for policy formulation and implementation • US definition (NSDI): • “technologies, policies, and people necessary to promote sharing of geospatial data through all levels of government, the private and non-profit sectors, and the academic community”

14. Spatial Data Infrastructures • SDI objectives are similar to the ones pertaining to other kinds of infrastructure • Fostering economic development by means of a range of publicly-available, multiple-use goods or services • This does not mean that the services are free of charge: do not confuse “publicly- available” with “government-supported”

15. Spatial Data Infrastructures • SDIs should • Provide standardized access to data • Have multiple participants, in the role of information services providers and/or users • Have a broad thematic scope • Facilitate data sharing

16. Spatial Data Infrastructures • Widespread adoption of Internet-based tools allowed the creation of Web portals to multiple geographic information sources • Geoportals

17. Geoportals • Geoportal • Web site that presents an entry point to geographic content on the Web • Includes tools for • The discovery of information sources and content • Online access to Web-based applications

18. SDI vs. Geoportal • SDI • Confluence of several different geodata providers • Access through specific Web services • Requires a repository or catalog of available services • Metadata • Usage by applications • Geoportal • An SDI encapsulated by a human-computer interface • Should be considered a component of an SDI

20. Service-based Distributed Systems Architecture • Service-oriented architecture (SOA) • Involves service providers, service aggegators and service users • Users may be human or software clients • Available services are listed in directories by providers • Aggregators design compositions of rules based on primary services

21. SOA

22. OWS vs WS • There is currently some confusion between OGC Web services (OWS) and W3C Web services (WS) • OWS do not necessarily use the usual W3C standards, including SOAP and WSDL • Instead of UDDI, OWS propose the use of catalog services • OWS have a particular interface for binding • OWS use GML, and not plain XML

23. OWS vs WS • The OGC has specified a range of basic Web services • Web Feature Service • Web Coverage Service • Web Map Service • Web Gazetteer Service • Web Registry Service/OpenGIS Catalog Service • Web Coordinate Transformation Service • Web Terrain Service

24. Local SDI • Most SDI development efforts focus on the usual activities of national mapping agencies, as the usual providers of basic mapping data • The richness of local GIS applications indicates that SDIs focused on local data can also be important

25. Local SDI • Motivation • The large number of actors involved in spatial activities in urban areas • The wide variety and high detail level of information classes involved in urban applications • The potential for fostering new applications on commercially interesting subjects • Location-based services • Personal routing • Convenience shopping • ...

26. Local SDI • A local SDI and an urban GIS are very different • SDI: widely-available, general-use data, accessible through services • GIS: organizational tool, built and operated around a definite set of goals

27. Local SDI • An urban GIS (maintained by the local government) may or may not become an SDI, depending on how freely the government distributes its information, and on how well other urban data providers are integrated in the maintenance effort

29. Services of Interest for a Local SDI • Basemap • Personal location • Geocoding/address recognition and location • Routing service • Public transportation system • Public services • Private services • Emergency services

30. Services of Interest for a Local SDI • The proposed services are rather different from the ones specified for NSDIs • It should be possible to chain these services to generate more complete or integrated services • Value-added commercial services based on public data • Public information of commercial interest available through a fee • Coordination of efforts during emergencies and natural disasters

31. Conclusions and Future Work • This is a work in progress • We are currently working to create a prototyping environment for Web services and SOA • An undergoing project on geocoding is being reconfigured to work under Web services

32. Conclusions and Future Work • Our goals include • Develop a better understanding of the SDI approach • Including performance, privacy, and security concerns • Develop a method for the design and implementation of WS/OWS for the SDI • Design means to publish metadata on services • Study the architectural possibilities for chained Web services • Mobile GIS applications

33. clodoveu@pucminas.br leonardo@lacerda.eti.br