=Paper=
{{Paper
|id=None
|storemode=property
|title=Towards e-Government Interoperability Framework
|pdfUrl=https://ceur-ws.org/Vol-706/paper11.pdf
|volume=Vol-706
|dblpUrl=https://dblp.org/rec/conf/dateso/FeuerlichtC11
}}
==Towards e-Government Interoperability Framework==
https://ceur-ws.org/Vol-706/paper11.pdf
Towards e-Government
Towards Interoperability
e-Government Framework
Interoperability
Framework
Jiri Feuerlicht1,2, David Cunek1
1
University of Economics, Prague
nám.1,2
Jiri Feuerlicht W. Churchilla
, and David4 Cunek1
130 67 Praha 3, Czech Republic
1 2
UniversityofofEconomics,
University Technology,Prague
Sydney
Broadway,
nám. W. Churchilla 4 NSW,
130 672007,
PrahaAustralia
3, Czech Republic
2
University of Technology, Sydney Broadway, NSW, 2007, Australia
jiri.feuerlicht@vse.cz, xcund01
jiri.feuerlicht@vse.cz, @vse.cz
xcund01@vse.cz
Abstract. Over the past decade most countries with advanced economies have
made major investments in various e-Government initiatives attempting to
replace traditional paper-based communications between government
departments, citizens and private organizations with electronic
communications. Notwithstanding such costly initiatives the success of e-
Government efforts has been limited. In this paper we discuss e-Government
initiatives in the Czech Republic and argue that successful introduction of e-
Government must be done in the context of a service-centric interoperability
framework that facilitates the development of e-Government services with well-
defined and stable interfaces.
Keywords: e-Government, interoperability, data integration, e-Government
services, service repositories, SOA
1 Introduction
Over the past decade many countries, including the Czech Republic made major
investments in various e-Government initiatives attempting to replace traditional
paper-based communications between government departments, citizens and private
organizations with modern computer-based systems and electronic communications.
While the beginnings of e-Government in the Czech Republic can be traced to the end
of the last century, it was only recently that the legislative changes that facilitate e-
Government have been introduced and became the subject of public debate. State
Information and Communications Policy - e-Czech 2006 [1] that incorporates the
concepts the e-Europe 2005 Framework was approved in 2004, and work on a number
of initiatives including the Information System of Databoxes (ISDS) [2] and the
Government Portal [3] have started shortly afterward.
Currently available public sector services in the Czech Republic can be categorized
into three types based on the type of interaction and on the style of interface (Table 1).
The type of service consumer (i.e. end user) typically defines the style of the
interface; web-based interfaces are widely used for interactions between PA (Public
Authority) and citizens, while interactions between PA and business entities typically
require a programming interface (i.e. API – Application Programming Interface).
V. Snášel, J. Pokorný, K. Richta (Eds.): Dateso 2011, pp. 39–48, ISBN 978-80-248-2391-1.
40 Jiri Feuerlicht, David Cunek
Table 1. Categorization of public sector services.
Interaction Service Type Service Consumer
PA -> End User Online information on web Citizens, Businesses
Codebooks Businesses
End User -> PA Downloadable forms on web Citizens, Businesses
Submissions over internet Businesses
PA <-> End User Signed email communication Citizens, Businesses
Data exchange via Databoxes Citizens, Businesses
Marketplace for public contracts Businesses
Interoperable services Businesses
The first type of services involves one way interactions between PA and service
consumers. This category of services includes information published on the websites
of various government departments such as codebooks for standardized data transfer
(e.g. units of measure for the Customs or Data Standard for the Ministry of Health).
These kinds of services are widely used and are relatively simple to implement. The
second type of services involves citizens or businesses sending data to PA, typically
using electronic forms that are downloaded from a website, completed by the end
users and uploaded. A more sophisticated version of these types of services are
services available via the Government Portal, which offers (in addition to a standard
web-based interface) a set of Web Services that can be integrated into client’s
information system (see section 2.2 for additional discussion). Finally, the third type
of services facilitates two-way interactions between PA and service consumers (i.e.
citizens or businesses). These interactions can take different forms, for example using
secure email with digital signatures or the ISDS [4]. Another form of interaction
involves specialized electronic marketplaces for public sector e-Procurement. The
most challenging type of two-way interactions involves interoperable services that
enable automation of business processes between PAs and business entities. Such
services should allow sharing of information among different government
departments avoiding duplication and reducing the amount of unnecessary paperwork.
Similar classification of e-Government services is used by the European
Interoperability Framework [5] that classifies types of services according to their level
of sophistication into four stages: stage 1) publishing information online, stage 2)
downloadable forms returned via email or mail, stage 3) online forms submitted
electronically, and stage 4) full automation and integration of services. Alternative
classifications of e-Government services can be found in the literature, for example in
[6].
Implementing interoperable services between government departments and
business entities involves overcoming a number of challenges, in particular
standardization of data structure and semantics across the public administration
domain, and specification of well-designed stable interfaces for various types of e-
Government services. Notwithstanding many costly initiatives the success of such e-
Government efforts has been limited [7]. There is some evidence that the focus of
such initiatives has been mainly on providing technical-level interoperability (i.e.
specifying data formats, data exchange and security protocols, etc.) rather than
providing an overall interoperability framework for e-Government services.
Towards e-Government Interoperability Framework 41
Furthermore, there is a widespread tendency to use existing (paper-based)
documents as the basis for developing electronic documents (i.e. XML data
structures) and map the data elements from paper forms directly to corresponding
XML message data structures. This is evident in data structures used for various
Government Portal services, e.g. Customs Administration, Czech Social Security
Administration, etc. This leads to a suboptimal solution with complex message
structures that require frequent modifications as new requirements are incorporated
into existing applications. This document-centric approach attempts to achieve
semantic interoperability directly by transformation and mapping of data elements
between disparate systems operated by different government agencies (see discussion
in section 3).
In this paper we argue that interoperable e-Government services can be best
achieved in the context of a service-centric framework that covers the entire services
life-cycle and produces well-defined and stable interfaces for e-Government services.
To achieve high levels of interoperability, e-Government services must be defined
from a global perspective (i.e. not separately for each government department) and
stored in a globally accessible service repository. We note that while e-Government
services represent a special category of interoperable services, experience with the
design of services in other domains, e.g. travel [8] can provide useful guidelines for
developing a suitable methodological framework.
In the following section (section 2) we discuss e-Government initiatives in the
Czech Republic with specific reference to the European Interoperability Framework
(EIF). The next section (section 3) describes the document-centric and service-centric
interoperability models, and in the final section (section 4) we discuss the limitations
of the traditional approach to e-Government interoperability and advocate a service-
based interoperability framework.
2 e-Government Interoperability Framework
Interoperability of e-Government applications has been the subject of extensive
investigations recently [9-12]. There is a wide agreement that interoperability cannot
achieved purely on the basis of providing connectivity at the technical level and that
effective sharing of information among government departments requires the support
of a methodological framework.
The European Interoperability Framework (EIF) [13] represents an attempt to
provide an architectural solution to facilitate automation of business processes among
public administrations, citizens and businesses within the European Union. European
Interoperability Framework defines four interoperability dimensions: legal,
organizational, semantic, and technical that operate within a political context. Political
context is set by government strategies; two documents define strategy for e-
Government in the Czech Republic. “National Information Strategy” [14] published
in 1999 that proposes the creation of information society and advocates the integration
of information systems in public administration, and “National Information and
Communication Strategy” (e-Czech) [15]. This strategy incorporates the concepts of
e-Europe 2005 [16], and describes the development of modern online services for the
42 Jiri Feuerlicht, David Cunek
public sector (i.e. e-Government, e-Learning, e-Health) and the creation of a dynamic
environment for e-Business. The most recent document setting out strategy for the
development of digital economy “A European Information Society for growth and
employment” (i2010) [17] represents a continuation of efforts to build information
society based on electronic services and the emerging services economy through
incorporation of the European Interoperability Framework into the national
framework.
Ensuring legal interoperability typically involves introducing new legislations, for
example the introduction of Information System of Data Boxes necessitated approval
of Act 300/2008 (Digital agenda), and the Government Portal is supported by Act
365/2000 (Information systems in public administration). Other important legislations
in this area include Act 111/2009 (Basic registers) and Act 227/2000 (Digital
signature).
Another aspect of e-Government interoperability that needs to be addressed is
alignment of organizational structures and processes in different government
departments (Organizational Interoperability). According to the EIF requirements of
service consumers (i.e. requirements of businesses and citizens) need to be taken into
account to minimize duplication when interacting with various government
departments. Consider, for example a situation where an individual is setting up a new
company and needs to provide various documents including criminal record,
information about social security debts, etc. Such information is already available in
various government departments and should be accessed automatically. This in
practice implies business process re-engineering and alignment so that individual
government departments share information, avoiding the need for the clients to
provide information that is already held by different government departments. The
pre-requisite for effective data sharing between government departments is semantic
interoperability, i.e. the standardization of data structures and semantics of data
elements used across government agencies.
2.1 Semantic-Level Interoperability
Semantic interoperability deals with the meaning of data exchanged among the
various agencies in public administration and requires that data heterogeneity present
in disparate information systems operated by individual government departments is
resolved. To address semantic interoperability EU established the SEMIC.EU
(Semantic Interoperability Centre) for member states to exchange information about
the meaning of data elements, which following approval by all participants are
published in a data elements repository.
At a national level, in the Czech Republic, metadata about information exchanged
between various PA departments is maintained using Information System about Data
Elements (ISDP) [18]. Public agencies are obligated to use data elements published in
the ISDP database when communicating electronically between PA departments.
However, ISDP data elements are not required for the implementation of interfaces
with private sector companies. The Czech Ministry of Interior has published a
document “Methodological Guidance for Creation of Data Elements” that describes
data element semantics and includes recommendations about data element life cycle
Towards e-Government Interoperability Framework 43
management. Another related document “Method of development of XSD schemas
in public administration” deals with technical issues that concern the development of
XML schemas and describes basic data types, composition of basic data types into
complex types and conventions for the design of schemas (e.g. namespaces, naming
of elements, etc.). ISDP maintains metadata information about data elements used in
public administration, but it should be noted that it is not a repository of definitions of
service interfaces. Although the use of ISDP data elements is mandated for public
agencies, the definition of interfaces for specific services is the responsibility of
individual service providers, typically government departments. This approach is
consistent with EIF, but it limits interoperability as services implemented by a given
government department will be, in general incompatible with services developed by
other departments, although their functionality may be identical. Another important
consequence of this approach is that there is no centralized repository of e-
Government services, potentially causing duplication and inconsistencies in service
definitions across various agencies of public administration.
2.2 Technical-Level Interoperability
Technical interoperability deals with technical aspects of interconnecting PA systems,
in particular with standardization of data formats and data exchange protocols (e.g.
SOAP, AS2). Other technical issues include ensuring performance, scalability and
security (identification, authorization, integrity, non-repudiation, encryption, time-
stamping, and protection against attacks and viruses). When developing interoperable
services, such issues have to be considered and resolved to provide a technical
solution. Two distinct communication channels are available for e-Government
applications to enable technical-level interoperability among PA agencies, citizens,
and private organizations: the Government Portal and ISDS.
The transactional component of the Government Portal is a gateway that enables
interactions between citizens or businesses (using Government Portal standards and
protocols) with PA institutions’ Department Interface Servers (DIS). Government
Portal implementation platforms is MicroSoft Biztalk [19] that manages data flows
between parties (including schema transformation, message routing, etc.) and can be
used to implement business processes that span government departments. Secure
communication between parties is achieved using transmission of XML messages
over HTTPS protocol. Messages are specified using XSD schema and consist of an
envelope containing routing information for a specific DIS server and data payload.
While the technical aspects of the Government Portal communications are
documented on the portal (Operating Rules for Community Portal for Developers), as
already noted, message specifications used by individual government departments are
not centrally managed.
While the Government Portal provides a standard gateway for the implementation
of e-Government services, ISDS emulates an email solution. Information System of
ISDS provides user interface similar to standard email and enables sending and
receiving of data messages that are stored in Databoxes. All Databoxes are stored in
one central location, so that sending and receiving messages does not involve data
transmission, but is implemented by linking sender and receiver records. In addition
44 Jiri Feuerlicht, David Cunek
to a web-based user interface, ISDS provides an API that enables connectivity via
SOAP and HTTPS. Using this public interface, users can integrate ISDS functionality
into their applications. The API is implemented as a set of Web Services (e.g.
CreateMessage, MessageDowload, etc.), that operate on data messages specified
using XSDs and consists of an envelope containing a signature and a timestamp and
data payload. In line with the ISDP approach, ISDS specification does not define the
structure of the data payloads, making the automation of messages processing
impossible.
3. Information-level Interoperability models
To fully appreciate the extent of the interoperability challenge and to explore
alternative solutions we need to refine the interoperability framework introduced in
the previous section (section 2). Interoperability has been the subject of extensive
investigation in the context of B2B (Business to Business) applications [20], [21] and
has been classified into three different types (levels): technical, information, and
business process interoperability [22]. Technical-level interoperability deals with
disparate communication protocols, language environments, and technology platforms
that are used by partner organizations and directly corresponds to technical-level
interoperability in the EIF framework described in section 2.2. Information-level
interoperability concerns data heterogeneity, and can be further classified into: syntax,
structure, and semantic heterogeneity. Syntax heterogeneity refers to differences in
formats used to represent data (e.g. XML, tagged document formats, etc.). Individual
organizations often use schemas with different structure of business documents to
represent the same information. Such schema differences are referred to as structure
heterogeneity. Semantic heterogeneity concerns the differences in the meaning of
individual data items and business process-level interoperability is concerned with
collaborative activities between the partners.
While technical-level interoperability is today largely resolved by adopting
appropriate standards (i.e. XML, HTTP, SOAP, etc.) as illustrated in section 2,
information-level and business process-level interoperability issues remain a
significant challenge, in particular in environments that involve a large number of
autonomous partner organizations. Discussion of business process-level
interoperability is outside the scope of this paper. Focusing on information-level
interoperability, two basic models have been documented in the literature: document-
centric and service-centric models [23].
3.1 Document-Centric Interoperability Model
Most existing implementations (including the e-Government interoperability
framework described in section 2) adopt the document-centric interoperability model
characterized by shipping business documents between partner systems. The main
advantage of this approach is that the use of documents as basic artifacts of business
communication avoids the need for compatibility of underlying technology platforms,
so that technical-level interoperability can be relatively easily achieved. Business
Towards e-Government Interoperability Framework 45
documents provide a level of abstraction that allows automation of inter-enterprise
business processes based on a mutual understanding of the structure and semantics of
documents. However, document-centric approach suffers from limited scalability
associated with its reliance on document translation, i.e. as the number of partners and
the complexity of their data structures increases, the mappings between individual
schema elements becomes unmanageable.
Another principal limitation of the document-centric approach is its tendency to
use large and complex message structures that typically mirror the original paper-
based forms. The complexity of message structures arises from designing message
payloads to include all the information needed to perform the corresponding business
function without any reference to information received in previous messages, making
the interaction essentially stateless. While such stateless interactions reduce the
number of messages needed to implement a particular business function, it results in
complex and redundant message data structures making changes to message formats
difficult to perform without introducing undesirable side-effects that invalidate
existing applications. The message payloads form the interface between applications
and therefore introduce high levels of data coupling and interdependencies by
externalizing complex document data structures [24]. In effect, the document-centric
approach attempts to solve a data integration problem for a potentially large number
of participants with diverse schemas and business semantics [25], and as is evident
from decades of research in the area of integration of heterogeneous databases, this
problem does not have a satisfactory solution.
3.2 Service-Centric Interoperability Model
The emergence of SOA (Service-Oriented Architecture) and Web Services provides
an opportunity for a new approach to addressing the interoperability challenge. Web
Services and SOAP (or REST services [26]) remove the need to use document
interchange as an interoperability mechanism by providing technical-level
interoperability. Service-centric interoperability model relies on well-defined service
interfaces that typically implement a single business function as a Web Service
operation. This reduces the problem of standardizing document formats and data
semantics to a more manageable task of standardizing service interfaces for a given
application domain. Such domain-specific service interfaces are conceptually similar
to APIs that are used extensively in programming environments. Interoperability of
service-oriented applications relies on stable service interfaces used consistently
across the e-Government application domain. Standardization of services ensures that
service providers (i.e. different government agencies and departments) publish
identical interfaces, avoiding the need to interpret interface semantics. The key
difference between the document-centric and the service-centric interoperability
models is that service interfaces can be designed to minimize interdependencies (i.e.
coupling) between services by encapsulating message data structures and exposing
method (operation) signatures that constitute a stable contract between the service
provider and the service consumer. Data engineering principles can be applied to the
design of service interfaces maximizing service cohesion and resulting in stable and
maintainable services [27], [28].
46 Jiri Feuerlicht, David Cunek
4. Conclusions
It is evident from our discussion in section 2 that the e-Government interoperability
framework adopted in the Czech Republic and based on the European
Interoperability Framework provides effective solution for technical-level
interoperability. However, a number of issues remain to be resolved to achieve
information-level interoperability. While the ISDP system maintains metadata
information about data elements, and the use of these standard elements is mandated
for public agencies, the use of ISDP data elements is not prescribed for private sector
companies. Furthermore, the definition of interfaces for specific services remains the
responsibility of individual service providers, so that while semantic consistency at
the data element level is assured, it is highly likely that individual government
agencies will produce incompatible services with overlapping functionality. The
absence of a centralized repository of e-Government services makes it difficult to
avoid duplication and inconsistencies in service definitions.
With growing acceptance of service-oriented computing and successful application
of services in application domains such travel [29] and healthcare [30], it is likely that
a service-centric interoperability framework would bring similar benefits to e-
Government applications. An important advantage of the service-centric approach is
that service interfaces can be designed to significantly limit the exposure of metadata,
considerably improving the stability and robustness of e-Government applications.
Other advantages of the service-centric approach include improved software
reliability simplified development, and support for evolution of interfaces [31].
A key requirement for achieving semantic interoperability in service-oriented
applications is standardization of service interfaces for a given application domain
(i.e. e-Government). Without such interface standards, equivalent services published
by different providers will not be compatible, placing a burden for resolving the
inconsistencies on service consumers.
Semantic interoperability is a necessary, but not sufficient condition for the
automation of e-Government interactions. Equally, adopting the service-centric
approach does not eliminate the need to agree on information and business process
semantics, but it makes the problem more manageable by limiting the scope of
agreement to service interfaces that typically correspond to simple operations (e.g.
change of residence). Well-designed service interfaces address the equally important
problem of limiting exposure of complex and often redundant data structures typical
of the document-centric approach. Moving away from an interoperability model based
on document interchange and adopting a service-centric approach results in a higher
level of abstraction associated with the use of application programming interfaces. As
experience with programming APIs demonstrates, the benefits of standardized
interfaces include improved software reliability, reusability, extensibility, and
maintainability, and can ultimately lead to significant application development
productivity gains.
Finally, it is evident that in order to address information-level and business process
interoperability issues data and business processes that are currently owned by
different departments have to re-engineered and integrated into a consistent set of e-
Government services maintained in a centralized service repository.
Towards e-Government Interoperability Framework 47
Acknowledgments
This research was supported by GACR (Grant Agency, Czech Republic) grant No.
GA406011 P403/11/0574.
References
1. Státní informační a komunikační politika. 2006 [cited 02-02-2011]; Available from:
http://knihovnam.nkp.cz/docs/SIKP_def.pdf.
2. Informační systém datových schránek (ISDS). 2010 [cited 02-02-2011]; Available
from: http://www.mvcr.cz/clanek/informacni-system-datovych-schranek-isds.aspx.
3. Portal verejne spravy. 2011 [cited 02-02-2011]; Available from:
http://portal.gov.cz/wps/portal/_s.155/6966/place.
4. Information System of Databoxes (ISDS). [cited 02-02-2011]; Available from:
http://www.datoveschranky.info/.
5. Chen, D. and G. Doumeingts, European initiatives to develop interoperability of
enterprise applications--basic concepts, framework and roadmap. Annual Reviews in
Control, 2003. 27(2): p. 153-162.
6. Curtin, G.G., Issues and Challenges: Global EGovernment/E-Participation, Models,
Measurement and Methodology. E-Participation and EGovernment: Understanding
the Present and Creating the Future, 2006.
7. Otesanek Egon. 2010 [cited]; Available from: http://ekonom.ihned.cz/c1-44167410-
otesanek-egon.
8. OTA. OTA Specifications. 2010 [cited 6 May 2010]; Available from:
http://www.opentravel.org/Specifications/Default.aspx.
9. Otjacques, B., P. Hitzelberger, and F. Feltz, Interoperability of e-government
information systems: Issues of identification and data sharing. Journal of
Management Information Systems, 2007. 23(4): p. 29-51.
10. Gottschalk, P. and H. Solli-Saether, Stages of e-government interoperability.
Electronic Government, an International Journal, 2008. 5(3): p. 310-320.
11. Gottschalk, P., Maturity levels for interoperability in digital government.
Government Information Quarterly, 2009. 26(1): p. 75-81.
12. Scholl, H. Interoperability in e-Government: More than just smart middleware. 2005:
IEEE. 0769522688.
13. European Interoperability Framework (EIF) Version 2. 2004 [cited 02-02-2011];
Available from: http://ec.europa.eu/isa/strategy/doc/annex_ii_eif_en.pdf.
14. National Information Strategy. 1999 [cited 02-02-2011]; Available from:
http://www.epractice.eu/files/media/media_425.pdf.
15. State Information and Communications Policy (e-Czech 2006). 2006 [cited 04-02-
2011]; Available from: http://www.epractice.eu/node/281047.
16. eEurope 2005. 2005 [cited 07-02-2011]; Available from:
http://europa.eu/legislation_summaries/information_society/l24226_en.htm.
17. i2010 - A European Information Society for growth and employment. 2010 [cited
06-02-2011]; Available from:
http://ec.europa.eu/information_society/eeurope/i2010/index_en.htm.
18. INFORMAČNÍ SYSTÉM O DATOVÝCH PRVCÍCH 2010 [cited]; Available from:
https://www.sluzby-isvs.cz/ISDP/DefaultSSL.aspx.
19. Microsoft. Microsoft BizTalk Server: Home. 2010 [cited 02-02-2011]; Available
from: http://www.microsoft.com/biztalk/default.mspx.
48 Jiri Feuerlicht, David Cunek
20. Ouksel, A., Sheth, A. , Semantic Interoperability in Global Information Systems: A
brief introduction to the research area and the special section. SIGMOD Record
Special Interest Group on Management of Data, 1999. 28(1): p. 5-12.
21. Medjahed, B., et al., Business-to-business interactions: issues and enabling
technologies. The VLDB Journal, 2003. 12(1): p. 59-85.
22. Bussler, C., B2B Protocol Standards and their Role in Semantic B2B Integration
Engines. Bulletin of the Technical Committee on Data Engineering, 2001. 24(1): p.
3-11.
23. Feuerlicht, G., Meesathit, S.,. Service-Centric Interoperability Model for Inter-
Enterprise Applications. in 2nd InterOP Workshop, in conjunction with the 9th IEEE
International EDOC Conference. 2005. Enschede, The Netherlands. 952-10-2698-7.
24. Feuerlicht, G., Wijayaweera, A. . Determinants of Service Resuability. in 6th
International Conference on Software Methodologies, Tools and Techniques, SoMet
06. 2007. Rome, Italy: IOS Press. 0922-6389.
25. Nagarajan, M., et al., Semantic interoperability of web services-challenges and
experiences. 2006.
26. Fielding, R.T. Architectural Styles and the Design of Network-based Software
Architectures. 2000 [cited]; Available from:
http://www.ics.uci.edu/~fielding/pubs/dissertation/top.htm.
27. Feuerlicht, G., Design of service interfaces for e-business applications using data
normalization techniques. Information Systems and E-Business Management, 2005.
3(4): p. 363-376.
28. Feuerlicht, G., Meesathit, S. Design framework for interoperable service interfaces.
in 2nd International Conference on Service Oriented Computing. 2004. New York,
NY, USA: ACM Press. 1-58113-871-7.
29. Papazoglou, M., B. Krämer, and J. Yang. Leveraging Web-services and peer-to-peer
networks. 2010: Springer.
30. Ko, L.F., et al. HL7 middleware framework for healthcare information system. 2006:
IEEE. 0780397045.
31. Bieber, G. and J. Carpenter. Introduction to Service-Oriented Programming (Rev
2.1). 2001 (updated 9-Apr-01) [cited 27 June]; Available from:
http://www.openwings.org/download/specs/ServiceOrientedIntroduction.pdf.