Building a Domain Model for the Governance System - Semantic Scholar

Download PDF
7 downloads 120012 Views 471KB Size Report
Comment
Building a Domain Model for the Governance System policies become even less effective insofar as the vertical breach couples with a horizontal one, as is ...
Building a Domain Model for the Governance System E-GOVERNMENT OR E-GOVERNANCE? BUILDING A DOMAIN MODEL FOR THE GOVERNANCE SYSTEM Vassilios Peristeras, Theodore Tsekos, Konstantinos Tarabanis United Nations Thessaloniki Centre, University of Macedonia [email protected], [email protected], [email protected] ABSTRACT In this paper, we propose, (a) a high-level representation for the overall governance system and (b) two models for describing the overall policymaking system’s function backed up with feedback mechanisms for controlling system’s Critical Success Factors. In recent literature and practice the e-government term has been mainly used for describing systems aiming at electronic service provision by public administration agencies. e-Governance, as introduced here, is a much wider concept as it incorporates information flows between society and political system, political and administrative system, and internal political system and civil society flows. Providing high-level models and definitions for the overall governance system is perceived to be the first step towards standardization and the creation of commonly accepted domain vocabularies and ontologies. The later constitute the sine qua non infrastructure for achieving interoperability at the content level, thus becoming the enabling factors for the development of a governance semantic web infrastructure. Keywords: e-Government, Domain Model, ontology, public administration, public policy INTRODUCTION - MOTIVATION Policy-making dichotomy A dichotomy concerning policymaking occurs usually and globally, due to the fact that design and implementation are two quasi- independent and loosely linked sub-processes. Policy analysis and design constitutes a top-down process involving political -or highly politicized- personnel. It is outcome oriented and operated on the basis of general criteria such as mission and vision concepts, organizational and environmental values and strategies, political priorities etc. On the other hand, policy implementation is mainly a bottom-up process involving P.A. professionals: middle management and low-level personnel. Therefore implementation activities are guided by intra-organizational priorities and day-to-day management requirements and restrictions. They are short-term output oriented with only vague references to the “big picture” and loose links to the organization’s long-term objectives and strategic priorities and goals. The unsuitable connection and mismatching of two complementary steps of a process supposed to be linear, leads to the disintegration of the overall rational policymaking framework and produces incoherent, ineffective and inefficient applied policies. Public 1

Building a Domain Model for the Governance System policies become even less effective insofar as the vertical breach couples with a horizontal one, as is usually the case. Policy outcomes are mostly the conjunction of distinctive processes in different policy fields. The inability of full communication and cooperation between all these distinctive policy constituencies and networks often results in poor policy outcomes. Loose and ineffective horizontal communications is a very common and widespread defect of contemporary policymaking and administrative systems (Tsekos, 2003). The problematic situation described above creates an urgent need for a bi-dimensional integrative interface bridging both the vertical and the horizontal policy gaps and linking all policy field actors to a cooperating “policy community”. This kind of interface can be built through extensive use of modern informational and communication technologies applied at each and every stage of the overall policy making process. One facet of this process must be associated with the transcription of current policy making procedures in ICT applications, in order to standardize, simplify and accelerate vertical coordination and horizontal networking and, thus, facilitate policy integration. Generic process and data structures, a high level model for strategic planning to provide common definitions, vocabulary and conceptual framework for policy making within broader policy field-based sub- categories of public agencies and a unifying “enterprise architecture” on which all public administration processing and information systems should be based in order to become interoperable are some critical prerequisites for integrated public policies Domain description and the Semantic Web Currently, the public standards, definitions administrative systems more problematic and

administration (PA) domain lacks commonly agreed content and vocabularies, not only at the global level among the worldwide, but even inside each country. To make things even although some initials attempts have been made (Tarabanis, Peristeras and Fragidis, 2001), (Tarabanis and Peristeras, 2000), (Inter-Agency Benchmarking & Best Practices Council, 1996), PA theory lacks a commonly accepted upper model description or a domain analysis (Arango and Prieto-Diaz, 1989) (Neighbors, 1980). The latter description could normally constitute the premise and serve at the starting point for the construction of an upper level content ontology for PA. In the computer science domain, ontologies aim at capturing domain knowledge in a generic way and provide a commonly agreed understanding of a domain, which may be reused and shared across applications and groups (Chandrasekaran, Johnson and Benjamins, 1999). Ontologies provide a common vocabulary of an area and define with different levels of formality- the meaning of the terms and the relations between them. Building a PA domain specific content ontology could provide vocabularies about the concepts within the PA domain and their relationships, about the activities that take place in PA, and about the theories and elementary principles governing this domain (Mizoguchi and M., 1996; Van Heijst, Van der Sperk and Kruizinga, 1996; The OntoWeb Consortium, 2001). In addition to the above, a content ontology could enable content-

2

Building a Domain Model for the Governance System based access, interoperability amongst existing Information Systems in governmental agencies, and the realization of semantically-enhanced communication across the Web. The latter constitutes the state-of-the-art today in electronic service provision regardless of domain (e.g. e-commerce, e-banking, e-learning, e-health). Tim Berners-Lee coined the term Semantic Web (Berners-Lee T., 1998) for describing these new possibilities and the new level of sophistication of a Web enhanced with metadata and knowledge structures. Among other features, this “upgraded” Web version will enable and facilitate features as knowledge representation, knowledge engineering, database design, information systems and database integration, natural language understanding, information retrieval and extraction, object-oriented software development, knowledge management and organization, agent-based systems development and semantic portals. The key role of ontologies for content-based data interchange in these areas is testified by the interest shown by many international standardization bodies and initiatives, including IDEF (Knowledge Based Systems Inc., 1994), ISO (ISO/IEC JTC1 SC32), ANSI (ANSI Ad Hoc Group on Ontology Standards, ), the W3C (Connolly D. et al., March 2001), IEEE (IEEE P1600.1, ). From the technological point of view, XML schemas have been proposed as agreed schemas for structurally describing PA services (e.g. e-GIF from UK (Office of eEnvoy UK, 2002) , GovML (Tambouris E., 2001)). Nevertheless, XML-based approaches soon reach their limits as they cannot effectively represent semantics. Using RDF schemas on top of XML could improve the situation but practice has shown that for elaborating truly semantic features, supporting ontologies are needed. Interestingly, two DARPA proposed languages, namely DAML+OIL (Connolly D. et al., March 2001) and DAML-S (The DAML Services Coalition, 2002) amongst others, built on top of RDFS and propose a generic semantic web service framework which can be applicable to the PA domain. Exploiting all the above-mentioned possibilities in the PA domain will be a great challenge for e-government systems designers. The Semantic Web is considered to be the infrastructure upon which all intelligent e-government applications will be built on in the near future. However in order for every domain, to fully exploit the new framework, formal and commonly accepted descriptions (models) are needed. Unless common definitions and vocabularies for describing the public administration domain are built, the steps towards semantic web enabled e-government will lag behind. Therefore, there is a pressing need for the timely construction of a public administration content ontology. In the complex, multi-layered, interconnected but highly fragmented and barely integrated field of public administration, semantics acquire a specific meaning and could be considered as a sine qua non for achieving interoperability and integration. This paper presents part of the work conducted so far towards the development of a highlevel representation for the domain of public administration as a policy making mechanism. PROPOSED MODELS

3

Building a Domain Model for the Governance System

The proposed models constitute two high-level representations: • A high-level domain model of the overall governance system focusing on actors and primary relations amongst these actors and called Governance System ActorsRelationships model (GSAR). • Two system-view models describing in more details the “service and regulatory provision” function, as the transformation of the input (society needs) to output (services and regulations) by the governance system. Namely, Administration to Society Macro-View model (mV-A2S), and the Administration to Society SystemView model (sV-A2S). The Governance System Actors-Relationships model The purpose of this model (fig.1) is to outline the domain of the Governance System in broad terms, presenting the main actors and relationships that exist amongst them. At the GSAR there are two main actors inside the Governance System (Chevallier, 1986):

• •

The Political System Public Administration (or Administrative System)

Figure 1. Governance System Actors-Relationships (GSAR) model •

Additionally there is a third actor lying out of the politico-administrative sphere: Society divided into two sub-categories (businesses and citizens)

The Governance System is defined as the relational model of these three sub-systems: political system and public administration interacting with society.

4

Building a Domain Model for the Governance System Although the separation between the political and administrative systems has been fiercely criticized by PA scholars for the “technical” separation it introduces, its analytical power should not be underestimated. Additionally, the GSAR model depicts the primary relationships among the various actors. Some of these are considered out of scope with regard to our work (eg businessto-citizens). A very interesting relationship exists between citizens and the political system. In democratic regimes citizens enjoy the privilege of electing their representatives and formulating in this way the political system. The relationship and information flow between the citizen and the political system in this area has been usually addressed by the IT industry using the term “e-democracy”. In addition to this, citizens can directly address their needs to the political system. More details on the later process will be discussed below. Looking at the relationship between the administrative system and society (A2S) we could mention that this area has been loosely specified as that of “public service provision”. It is important to mention that the majority of the so-called e-government initiatives address this area. This has been done at an initial stage through the development of Administration to Citizens (A2C) and Administration to Business (A2B) front office applications. Though e-government system designers soon realized the limitations of such approaches that left the back-office of the administrative production intact. Recently, developing Administration to Administration (A2A) (or back-office) systems and applications, has became a prerequisite for realizing electronic services for citizens and businesses. The common use of the term e-government just for “public service provision” may be misleading, as the other two relationships (politicaladministrative and political system-society) are left out of the e-government scope. For clarifying the term we have two alternatives: either to generalize the e-government notion to include all the existing relationships (political-administrative, society-political system and service & regulation provision), or to leave the term with its usual connotation (just electronic service provision) and coin a new word for expressing a super-class to which e-government, e-democracy and politico-administrative IT support are sub-classes. In the figure, the following relationships can also be found: • Citizen to Citizen (C2C): These relations constitutes the basis of what has been called “Civil Society”(Bridges, 1994) and although they belong to the governance system will not be addressed here. • Business to Business (B2B) and Business to Citizens (B2C): As already mentioned, these relationships are out-of-scope in our work. • Political System to Administrative System (P2A): At this point, some very interesting relationships exist and a critical flow of information takes place. At this paper we will address later in the presentation of the mV-A2S and sV-A2S models, only the part of these flows that are connected with the service & regulation provision.

5

Building a Domain Model for the Governance System •

Political to Political (P2P). In these category, we could include various “interior” political system relationships (eg between the parliament and the government, the president and the prime minister, amongst the various political parties, etc). Although these relationships are crucial for the function of the governance system, they have attracted only marginally the interest of IT community. These relationships interests us a lot but they are left out for the purpose of this paper.

The Administration-to-Society Macro-View (A2S-mV) and the Administration-toSociety System-View (A2S-sV) models The two representations that follow below model the part of the GSAR that describes the services & regulations provision. Administration to Society Macro-View model (mV-A2S). The first schema, called Administration to Society Macro-View (mV-A2S) model.(Fig. 2), is quite simple as it presents a macroscopic, high-level and generic representation: The governance system (political and administrative system plus interaction interface with the society, including collective representative bodies, consultation institutions and processes as well as formal and informal communication channels) receives as input, society needs, and after processing them produces services and regulations to Fig. 2 address these needs. This black-box view is useful for better understanding a system’s purpose and telos. The Administration-to-Society System-View model (sV-A2S) The second model called Administration-to-Society System -View (sV-A2S) model goes into more details and decomposes the mV-A2S model into various phases (fig. 3-4). The generic description upon which sV-A2S is based, secures the applicability of the latter to all different public policy fields (eg urban planning, security, public health, education, development, etc) As already mentioned, the sV-A2S model represents in more details the “Service & Regulatory Provision” part of the GSAR model. There are three main sub-systems participating in the sV-A2S model: Society (from where everything starts and ends), the administrative and the political sub-systems. Actually, society is the ultimate “client” that must be served. For this reason, society has delegated power to the political sub-system, acknowledging to the latter functions of a primary “server” towards the society. In democratic regimes, through the election process, society decides whether a specific server (political party) has successfully offered its services or if there is a need to test

6

Building a Domain Model for the Governance System another type of server claiming to offer either something different or the same but more efficiently. In between the primary client and the primary server though lies the administrative system. How does this system gain its legitimacy? Who assigned executive powers to it? Its role, presumed by its position, is that of a broker. Its specific functions as a broker are analyzed separately in the two different parts of the model it participates. During the service & regulation provision, part of the political-administrative relations existing in the GSAR model also have to be activated. More specifically the following information flows occur between the two systems: • From political to administrative, decisions made by the first and imposed through a legitimated dominance afterwards to the latter. This dominance in democracy gains its legitimacy through the electorate and is practically implemented with the physical presence and the executive power of the ministers inside the administrative structure they head (see also Fig.1). • From administrative to political, at the first stage (upwards movement of information) the information gathered by the administrative system describing the society needs and at a later stage (downwards movement of information), expertise and administrative knowledge regarding how to realize political decisions. The sV-A2S model (Fig. 3) is presented as a circle starting from its bottom, where the social need for collective action triggers and activates the whole system. In fact, it is this need that legitimates the necessity of building the governance system and entrusts it with the monopoly of exercising physical violence (M. Weber). Alternatively the sV-A2S model is presented in Fig.4 as a UML Interaction Diagram. Before starting the detailed description of the sV-A2C, it is interesting to map the upper part of the model as presented in fig.3, (political system zone) with the “formulate public policy” major public administration process (Tarabanis and Peristeras, 2000). Accordingly, the middle part (administrative system zone) where the service provision occurs corresponds with the “provide service” major process. Let’s take a closer look at what happens during each phase: Everything starts, as already mentioned, with a societal need for collective action. Generally, the society has two ways for communicating a need to the political system: the formal, bureaucratic channel through the administration, and a number of alternative channels which bypass bureaucracy in order to reach the political system. These channels may vary from a formal political party or NGO to an informal drinking of coffee of the political appointee downtown together with “ordinary citizens”.

7

Building a Domain Model for the Governance System

Administration-to-Society System-View NORMATIVE CYCLE

(sV-A2S) model POLICY FORMULATION & APPROVAL

POLICY DESIGN

POLICY IMPLEMNT

INDIVIDUAL CYCLE

Figure 3 Returning to the formal administrative channel which is of interest in this paper, the first role of the administrative system becomes apparent: it has to develop methods for collecting societal needs. Unlike the flexibility inherited by the non-administrative channels, administration has to set up a system capable of gathering information from the general public. For having this system operating effectively, the capacity of not only collecting the declaratively expressed societal needs but also “sensing” needs, thus acting in a proactive way, must be included in this system. After this first step, “administrative processing” follows. This processing is a first attempt by the administrative system to organize the unstructured information that comes in, from the “collection” phase. Categorizing, translating to administrative language and summarizing are some aspects of this processing. However at this stage there is always a controversial point. The administrative system has no right to choose or evaluate the upward coming needs. Bureaucracy has to be completely neutral, a blind instrument that acts only with logic and professional expertise. Evaluation means judgment and judgment requires a set of values different from the set of values upon which bureaucracy has built its legitimacy (neutrality and professionalism). Though a political question arises: Is it possible for the administrative system to process the addressed or forwarded demands neutrally? Or is there always an indirect intervention of the latter to the flow of information from the primary client to the primary server, acting as a filter based on values and criteria non-explicitly expressed? Of course, these questions will not be addressed in this paper.

8

Building a Domain Model for the Governance System Keeping in mind the controversial legitimacy that characterizes the administrative processing step, we reach the “check-in” point as can be seen in the figure. This is the point where the administrative flow meets the alternative root mentioned previously. From the organizational point of view it is here that we usually find the offices of the political personnel. They are doing the hard job of trying to balance and transfer to the key person all the information they feel critical. The overflow at this point is very usual. The office is bombarded by controversial pressures and additional processing becomes indispensable before the information reaches its target in the political system. Processing by the political layer though is something completely different than the previous type of processing. At this point neutrality is not accepted. Since we have entered the realm of the political sub-system, political choice based on a set of political criteria has to be realized. Society has delegated to them the power to exercise this privilege. So the office excludes some of the needs as inappropriate (e.g. as not compatible with the supported political agenda) and presents a final list of issues to be addressed by the political system.

Figure 4 At this step we have reached the phase of prioritizing. This phase is perhaps the central function of the political system in all regimes. In democracy more specifically, different political approaches are evaluated by the electorate depending on how political personnel prioritize the society needs that address them, after the afore-described phases. For a citizen being a member of a specific party means acceptance of a one proposed prioritization and rejection of another. The positioning of “Prioritizing” at the head of the model emphasizes the prominent importance that this operation has over the whole system. Being a political leader and having a clear political position practically means having explicitly presented your set of values and criteria upon which you (as political leader) will judge all emerging issues and accordingly place them, hierarchically ordered, in your political agenda. The main output of this step should be an hierarchically ranked list of political To-Dos, or in other words a political plan.

9

Building a Domain Model for the Governance System With the political decision occurring at the prioritizing phase, we start moving downwards in the model. The upward movement was the flow of the society needs towards the decision-makers (or primary servers), the downward movement is the flow of the political decisions (or orders) to the ultimate client. The prioritizing phase is followed by the “assigning” phase. If the former clearly expresses political ideology, the second deals with political organization and practicality. At this step the political subsystem delegates its legitimacy to the administrative actor to realize the political system’s ideas and priorities. Though, as can be seen in the Fig.3&4, there exists an alternative channel for realizing political ideas: the private sector can act as an alternative provider of public services to society. Through its legislative power, the parliament assigns duties to organizational entities (either public or private) to realize the agreed - or imposed by the majority - political agenda. The ideas and visions become concrete political plans, with actors, budgets, accountability and management. Amongst others, the specific administrative level (eg central, regional or local administration) for the realization phase is decided. The “check-out” point that follows is where the decision leaves the political sub-system and through the ministers (acting as the main actors) returns to the administrative subsystem (or the private sector). What PA people usually receive from this stage is a law, that has to be enforced by them. Enforcement of course can mean numerous different things: impose and check for compliance, provide a new service to the society, build a new organization, etc. Now the administrative system, with the help of the political appointee (minister) has to organize and execute practically what was asked. What is the role of the minister now? Is the minister acting still as a political personnel during this realization phase? Basically no! The political decisions have been made in the previous steps (eg when the minister/government/parliament prioritized the needs of the society to be covered by the ministry). But now the minister plays the role of the trustee agent appointed by the political system to manage the realization operation that will be executed by the administrative system. The minister becomes a manager, responsible for the practical realization of a political agenda. And as the political system wants a politically dedicated manager, the minister is preferred to a technocrat. Together with the experts of realization (public administration professionals) the minister tries to realize the political decisions and to produce concrete results. So the “realization” phase constitutes the main area for administrative action. The framework and rules have been decided and implementation starts. All activities related to the administrative production of services are linked to the “realization” phase. Building or operating a hospital, safeguarding security in the cities, providing services to entrepreneurs, preventing or reimbursing victims of natural disasters, the production of all services these clearly belong to this “realization” stage. What follows next is the distribution. We have intentionally separated the production phase (realization) from the distribution phase, proposing a technical but useful separation between what has been called front- and back-office. This separation although very common in other industries (eg food, banking & insurance, etc) only recently has

10

Building a Domain Model for the Governance System started to attract interest in public administration, where production and distribution was supposed to be an integrated part of a unique function. Since the end of nineties, there are several initiatives from administrations worldwide to exploit the apparent advantages of dealing separately with production and distribution in public administration. The interesting “shopping mall” concept for public services or the idea of establishing citizens’ centers and kiosks to integrate at the front-end complex administrative processes and provide services from a single point according to the one-stop-shop administrative paradigm, are examples of this trend. Of course here once again the area pertains to the administrative actor. Obviously, with the “distribution” phase the whole governance system hands its output to the society. The circle started from this same point and ends again in society. Society asked for collective action to solve emerged inconsistencies, and finally society gets an output conceptualized by the political system and realized by the administrative system. This output is supposed to constitute the “solution” to the initial societal need. But is it really so? What type of controls and feedback mechanisms are needed to secure compliance between what was asked and what was received? The above described schema, placed on the outer cycle of figure 7, represents the normative stage of the overall process. During this stage output, outcome and process standards are set, generating a policy making system. The internal cycle represents the treatment of individual applications and demands throughout the administrative machinery within the normative framework produced and installed through the external, normative, policy making cycle. In addition to the system’s input-output descriptions, we propose four feedback loops in order to control several systems characteristics that could be perceived as being critical success factors (CSFs) for the overall system’s operation. We envisioned four types of primary controls to be applied upon the system. Each control aims at providing feedback and checks the internal system’s (and sub-systems) capacity in different stages. A short description follows: 1st Control – Political Awareness: Checks the divergence between what society needs and what the political system thinks society needs (effectiveness of the communication between political system and society). 2nd Control – Administrative Accountability: Checks the divergence between what the political system wants to provide to society and what society finally gets out of the administrative system (quality of communication between political-administrative system, administrative capacity, etc) 3rd Control – Political Accountability: Checks the divergence between the received demand from the society by the political system and the final political plan which is communicated by the former to the administrative system for addressing this demand (political liability, capacity for policy making) 4th Control – Governance Responsiveness: Checks the divergence between what society has asked for and what society receives by the governance system (system capacity, entropy, effectiveness, efficiency, etc).

11

Building a Domain Model for the Governance System

Current situation (Fig 5), involves policy making practices that generate a multiplicity of problems of diverse nature in each and every step of the above mentioned process

Policy making deficiencies FORMULATION PROCESS

Stove-piped input & data pools Disparate data Lack of co-variation detection Lack of contextuality Non integrated decision making

Unstructured processes Non standardized input. Lack of feedback Opacity. Bureaucratic arbitrariness Micro-clientelism

Non integrated top management Lack of complementariness Conflictual assignment Fragmented back office processes Non interoperable dpts Lack of data sharing Red-tape

Non adaptive output Inadequate outlet Bottlenecks

Complexity

INPUT PROCESS

Insufficient Monitoring & Evaluation

Political discretion Macro-clientelism

OUTPUT PROCESS

Fig. 5 Policy pitfalls could result first of all from inadequate input mechanisms such as insufficient social dialogue and consultation, restricted representation and other similar deficiencies creating an authoritative and clientelistic political environment. At the administrative layer unstructured transaction processes and a lack of monitoring, feedback and transparency engender red-tape and bureaucratic arbitrariness. Political decision making can become biased if disparate and stove-piped input and data jeopardize the understanding of connectivity and interactions between policy fields and impede effective policy prioritization and consequent rational resource allocation. Non integrated policy formulation couples with uncoordinated and even conflictual policy implementation. Discorded assignments, insufficient horizontal communication and coordination, overlapping and lack of complementariness lead to quantitatively and qualitatively poor policy results. Finally, absence of interoperability between public agencies at the day-to-day and the street level management causes more red-tape, bottlenecks and unsatisfactory policy deliverables. In order to remedy the policy making deficiencies and inadequacies described above an Integrated Policy Making model is required (Fig 6). This system must aggregate by means of enhanced social dialogue, concerted and multi-criteria decision making and

12

Building a Domain Model for the Governance System standardized and simplified processing, all facilitated by ICT, the fragmented steps into a coherent, homogenous and well articulated system

e- Remedies DECISION SUPPORT SYSTEMS- DSS

FORMULATION PROCESS

EXECUTIVE SUPPORT SYSTEMS-ESS

Integrated data bases Common ontologies Compatible info architecture Interoperability MANAGEMENT INFORMATION SYSTEMS- MIS Complexity

Fully interactive portals Multiple outlets

INPUT PROCESS Citizens Panels e- Voting NGO’s participative framework

INTEGRATED POLICY SYSTEMS

OUTPUT PROCESS

Fig. 6 ICT solutions can substantively facilitate the necessary policy rationalization and integration process at different levels. e-Democracy applications, not just e-Voting but chiefly e-Referendum solutions, coupled with on-line monitoring can improve public consultation and transparency at the social needs input and the policy agenda setting level. Transaction Processing Systems at the user’s input level could speed-up intraadministrative and back-office processes. Users Relationship Management Systems, based on CRM applications, could create a user friendly environment for citizens’ and businesses’ transactions with governmental agencies, by multiplying governmental outlets, reducing users’ movements, standardizing and simplifying processes etc. Management Information Systems at the agency level can improve operational and implementation planning. Decision Support Systems at the ministerial level can integrate decision making with monitoring and feed-back improving public action within a specific policy field. Executive Support Systems at the government and parliamentarian level can multiply and diversify decision making criteria and data and, therefore, rationalize policy prioritization and resource allocation. Last, but not least, data and application sharing facilities based on common activities architecture and unified administrative terminologies will permit interoperability between public agencies reducing time and cost requirements and improving quality of service delivery.

13

Building a Domain Model for the Governance System All these, already existing, technologies mostly developed in view of the private sector business requirements, need to be adapted to the public action specificities. Above all, they have to be linked effectively to each other. They must become well integrated components of a coherent policy making mega-system capable to deal with the extreme policy making complexity and create a smooth and traceable policy path from social needs assessment to street-level service delivery. CONCLUSION In this paper, we propose, (a) a high-level representation for the overall governance system and (b) two models for describing the policymaking system’s function backed up with feedback mechanisms for controlling system’s Critical Success Factors. In recent literature and practice the e-government term has been mainly used for describing systems aiming at electronic service provision by public administration agencies. eGovernance, as introduced here, is a much wider concept as it incorporates information flows between society and political system, political and administrative system, and internal political system and civil society flows. Providing high-level models and definitions for the overall governance system is perceived to be the first step towards standardization and the creation of commonly accepted domain vocabularies and ontologies. The later constitute the sine qua non infrastructure for achieving interoperability at the content level, thus becoming the enabling factors for the development of a governance semantic web. REFERENCES ANSI Ad Hoc Group on Ontology Standards available at http://www-ksl.stanford.edu/onto-std/index.html, as accessed in 2 Sep, 2000 Arango, G. and R. Prieto-Diaz (1989). Domain analysis: Concepts and research directions, in Domain Analysis: Acquisition of Reusable Information for Software Construction, IEEE Computer Society Press. Berners-Lee T. (1998). Semantic web road map. Internal note, World Wide Web Consortium. Bridges, T. (1994). The Culture of Citizenship: Inventing Postmodern Civic Culture. NY, State Univ of New York Press. Chandrasekaran, B., T. R. Johnson and V. R. Benjamins (1999). “Ontologies: what are they? why do we need them?” IEEE Intelligent Systems and Their Applications.Vol. 14(1: Special Issue on Ontologies): 20-26. Chevallier, J. (1986). Science Administrative, Presses Universitaires de France. Connolly D. et al. (March 2001) DAML+OIL (Reference Description), available at http://www.w3.org/TR/daml+oil-reference, as accessed in 10 Oct., 2002 IEEE P1600.1 Standard Upper Ontology (SUO) Working Group, available at http://suo.ieee.org/, as accessed in 28 Nov, 2002 Inter-Agency Benchmarking & Best Practices Council (1996) Government Process Classification Scheme, available at http://www.va.gov/fedsbest/index.htm, as accessed in Knowledge Based Systems Inc. (1994) IDEF5 Method Report, Information Integration for Concurrent Engineering, available at http://www.idef.com/Downloads/pdf/Idef5.pdf, as accessed in 2 July, 2002 Mizoguchi, R. a. and I. M. (1996). “Ontology Engineering - Towards the Basic Theory and Technology for Content-Oriented Research.” J. Jpn. Soc. for Artificial Intelligence.Vol. 12(4): 559-569. Neighbors, J. M. (1980). Software construction using components. Irvine, Department of Information and Computer Sciences, University of California.

14

Building a Domain Model for the Governance System Office of e-Envoy UK (2002) e-Services Development Framework Primer v1.0b, available at http://www.govtalk.gov.uk/documents/eSDFprimerV1b.pdf, as accessed in 13 Noe, 2002 Tambouris E. (2001). An Integrated platform for Realising Online One-Stop Government: The eGov Projet. DEXA International Workshop "On the Way to Electronic Government".Vol. pp. p. 359-363:IEEE Computer Society Press. Los Alamitos, CA. Tarabanis, K. and V. Peristeras (2000). “Towards an Enterprise Architecture for Public Administration : A Top Down Approach.” European Journal of Information Systems.Vol. 9(Dec. 2000): 252-260. Tarabanis, K., V. Peristeras and G. Fragidis (2001). Building an Enterprise Architecture for Public Administration: A High Level Data Model for Strategic Planning. 9th European Conference on Information Systems.Vol. pp.:June 2001, Bled, Slovenia. The DAML Services Coalition (2002) DAML-S: Semantic Markup For Web Services, available at www.semanticweb.org/SWWS/program/full/paper57.pdf, as accessed in 10 Noe, 2002 The OntoWeb Consortium (2001). Technical Roadmap v1.0: p.11. Tsekos, T., Ed. (2003). Towards Integrated Policy Making: Remedying the Public Action Dichotomy Through Information and Communication Technologies and Learning. State Modernization and Decentralization, Implications for Education and Training in Public Administration: Selected Central European and Global Perspectives, NISPAcee. Van Heijst, G., R. a. Van der Sperk and E. Kruizinga (1996). Organizing Corporate Memories. Proc. 10th Banff Workshop on knowledge Acquisition for Knowledge-Bases Systems (KAW 196).Vol. pp.:SRDG Publications. Calgary, Canada.

15