=Paper=
{{Paper
|id=Vol-241/paper-6
|storemode=property
|title=Regulation in Information Systems at the Level of Tunement
|pdfUrl=https://ceur-ws.org/Vol-241/paper6.pdf
|volume=Vol-241
|authors=Michel Léonard,Abdelaziz Khadraoui,Jolita Ralyté
|dblpUrl=https://dblp.org/rec/conf/caise/LeonardKR06
}}
==Regulation in Information Systems at the Level of Tunement==
https://ceur-ws.org/Vol-241/paper6.pdf
818 Regulations Modelling and their Validation and Verification
Regulation in Information Systems at the Level of
Tunement
Michel Léonard, Abdelaziz Khadraoui, Jolita Ralyté
CUI, University of Geneva, Rue de Général Dufour, 24, CH-1211 Genève 4, Switzerland
{Michel.Leonard, Abdelaziz.Khadraoui, Jolita.Ralyte}@cui.unige.ch
Abstract. Information Systems (IS) have a major impact on human activities.
They are not only “objects or products” because they directly concern our
access to knowledge, our possibility to assume activities and even our survival
in this economic world. Due to these IS stakes, we propose to position the
regulation process at the level of initiatives in the IS development, more
exactly, in the context of the IS tunement between the worlds of human
activities, systems, conceptual models and ontology.
Keywords. Regulation, information systems, world-object, alignment, e-
Government, conceptual model, ontology
1 Introduction
The concept of regulation is not often used in the domain of Information System (IS)
although it is a well known concept in the general theory of systems. Obviously,
regulation processes for systems programming and implementation are available for
IS because of the fact that an IS must be programmed and implemented. But, an IS
cannot be reduced to a computerized system. We have to invent a more central
position for the IS regulation process. In this paper, we firstly present the IS stakes for
an enterprise or, in more general sense, an institution, and even for the society in the
case of e-Government. Secondly, we introduce four main worlds of an IS and we
propose a position for the IS regulation in order to observe and regulate crucial points
in the IS development and evolution processes. Finally, we illustrate our approach
with a case of e-Government.
2 IS Stakes
The domain of software engineering contains many situations of regulations which
are pertinent to improve the quality of the systems to be developed. These traditional
artefacts have a local impact on human activities. Due to this locality, we know how
to construct rigorous and formal backgrounds to develop methods for their
development and their implementation. We can obtain a patchwork of implemented
systems, each of them having a specific utility. We can master a system, and even
REMO2V'06 819
more, we can have the feeling of possessing it because that we know how to define it,
how to implement it and how to use it. These artefacts are objects (ob-jectus: which is
outside of ourselves, but not too far for our thinking). For these artefacts there are
users. The users are active subjects taking information from the passive objects.
However, through the IS domain we enter a new dimension. We are not any more
outside the IS itself for the reason that it continuously interacts with our own
activities. We are not any more in the situation of active subjects who can master the
system with passive objects but in the situation of exchanges where the active roles
and passive roles may be inverted. The IS itself becomes a mandatory condition for
accessing knowledge, assuming activities and even surviving. As a consequence, the
IS domain asks us to leave the traditional schema relative to objects. Certainly, an IS
has objective quality as objects, which can be observed and analysed. But, in a more
generic way, it acts on the global constraints of our own life and on our own survival
conditions. For us an IS is a world-object [4]1.
Any traditional attitude to design an IS follows a kind of one-way or linear
causality category, even with sophisticated feedback or spiral movement (to
determine users requirements and objectives, then from these results to construct
specifications and to implement them and finally, to analyse users satisfaction). Such
an attitude is not any more sufficient to surmount the complexity of the IS domain
even if it keeps some pertinence locally: a pertinent IS investigation must take into
account new categories of interaction, transaction, organization and teleology2.
3 Position of IS regulation: IS Tunement
A lot of papers concern the alignment of the business policy and the informatics
policy [1]. Some authors claim that it is necessary “to bridge the gap” between these
two domains. This point of view does not take into account the power of initiatives of
these two domains and that their processes of initiatives are independent. Even if a
bridge between these two worlds can be built, it will be immediately destroyed
because of the movements of these two worlds. Therefore, for us, no bridge is
possible and no alignment (in the strict sense) can be achieved. The IS domain is
much more complex.
Our approach is based on four worlds: conceptual model, activity, system and
ontology. The notion of conceptual model was introduced in the last 80’s by several
research teams but we give to it a new role – the role of interoperability between the
two previously mentioned domains. The world of conceptual model has its own
properties and concepts, roughly speaking, the world of information semantics
integrating static and dynamic aspects in the same model and related constraints.
Furthermore, it has an overlap with the activity world, and this overlap concerns
1 This paragraph is strongly inspired by [4].
2
This paragraph is strongly inspired by: "Compared to the analytical procedure of
classical science with resolution into component elements and one-way or linear
causality as basic category, the investigation of organized wholes of many variables
requires new categories of interaction, transaction, organization, teleology..." [6].
820 Regulations Modelling and their Validation and Verification
enterprise models. It has another overlap with the system world, and this overlap
concerns the specifications. The process to go from one world to another must be
seamless: what is decided in one world must be implemented trusty, without
modification in the other world. For instance, all the informational specifications
decided in the conceptual world must be precisely respected in the implementation of
the corresponding system. The conceptual model can be simplified into a concept
model, generally called ontology. For us, ontology contains all the invariants of the IS
domain, in particular knowledge but also some business rules, roles of persons which
are independent of the IS development. For example, in the domain of e-Government,
the laws belong to ontology.
Therefore, an IS is in fact composed of four worlds, which have their own
principles and properties, their own independent power of initiatives. We identify two
crucial situations in the IS development where we will introduce the IS regulation
process:
− Since each IS is always in movement due to the initiatives taken in any of its parts,
in any of these four worlds, a crucial activity in the IS domain is to assure the
tuning between these four worlds. We call this part IS tunement;
− Due to the IS stakes mentioned before, crucial and autocratic initiatives must be
avoided as their impact on the enterprise, institution or society concerns many
people. For that reason, the initiatives should be submitted to a more democratic
process.
4 IS Regulation
The main property of a system behind regulation is named homeostasis. “Homeostasis
is the property of an open system, especially living organism, to regulate its internal
environment to maintain a stable, constant condition, by means of multiple dynamic
equilibrium adjustments, controlled by interrelated regulation mechanisms” [7]. In
this work we investigate another approach to reach homeostatic IS. To assure the IS
homeostasis, we need a regulation process concerning the crucial point of the
initiative process. Who is allowed to take initiatives and how? The centralized power
is now irrelevant due to the IS stakes, due to the complexity of situations which
cannot be overcome only by general considerations. Furthermore, due the IS stakes,
any autocratic initiative process will fail. The initiative process needs the participation
of many persons, as many approaches claim. But, the question is how to manage it
with a regulation process? For this purpose we introduce the notion of regulation
committee and we illustrate it through an e-Government case.
4.1 Informational Space of the Regulation Committee
The regulation committee must have an informational space to work, which should be
the most objective as possible. If we consider the four IS worlds, only the ontology
world have this quality.
REMO2V'06 821
Authorization Request Patient (Drug Addict)
name_substance name
Doctor dosage first_name
doctor_name mode_of_administration date_of_birth
treatment_duration complete_address
name_of_pharmacy identity_paper_nature
Drug Prescription
Request Treatment
Decision Cantonal Doctor
Drug Drug
Distribution Prescription
Drug Distribution Drug Administration
Accepted Request Refused Request
Drug Notification
Administration Cantonal Pharmacist
Examination
Notified Refusal
Pharmacist Appeal
Examined Authorizations
Appeal
appeal_time
Legend
ClassA Class3
Transaction ClassB depends Class2 is a
existentially of specialization of
tri ClassA Class3
ClassB Class2
Fig. 1. IS Ontology extracted from the Geneva law K 4 20.06
For example, in the domain of the prescription of narcotics (drug) intended for the
treatment of the dependent people (the Geneva law K 4 20.063), we model this law
(see Fig.1) in terms of our ontology model (essentially based on the existential
dependency relationships between concepts) [5, 2]. A transaction tri in this model has
one or more pre-transaction classes clPrej, and one or more post-transaction classes
clPostk. If completed, tri interacts with the objects of clPrej and produces new objects
in clPostk. In this model, conditions in entry or in exit of a transaction can be complex
Boolean expressions.
3 The law describes the procedure that each doctor must follow to fill a request of
authorization in order to prescribe a narcotic for the treatment of the dependent people
(drug addict). The doctor must obtain an authorization from the cantonal doctor
before the prescription of any narcotic. The law also describes how the drug has to be
distributed and administered. The pharmacist, on the basis of the authorization
delivered by the cantonal doctor, provides the doctor, or directly the patient with the
prescribed drug. (http://www.geneve.ch/legislation/rsg/f/rsg_k4_20p06.html)
822 Regulations Modelling and their Validation and Verification
In this paper, we limit ontology to the law. This result is objective and is more or
less independent of modellers (some variants can happen for details).
4.2 Regulation of Information Spaces of Actors
In the law, there is no consideration of the information spaces of the major actors.
Because of the fact that organisations ware paper-oriented, the legislative power did
not take care of it when promulgating a low. But now, in the epoch of
computerization, the IS developers must know the access rights to be implemented.
Without any regulation process, these decisions have to be mostly made by the IS
developers. Thanks to the ontology for the actor “Cantonal doctor” illustrated in Fig.
2, it is possible to decide how the main actors will work together and the regulation
process has then the role of a moderator in the IS development process.
Doctor Patient
(Drug Addict)
Authorization Request
Study Time
Decision Cantonal Doctor
2 days
Accepted Request Refused Request
Specialisation Link Existential Link Instanciation Link
Fig. 2. Informational space for a Cantonal Doctor.
4.3 Regulation of Initiatives Concerning the Way of Organization
Any person competent in observing ontology model can take initiative to elaborate
new kinds of organization. In our example (Fig. 2), there is only one cantonal doctor
and only one cantonal pharmacist. Why? This position could have a political
background: the majority of the Assembly thought that only one person must be in
charge of this activity. Moreover, this position could have been taken implicitly: due
to the paper-oriented organization it was normal to think that one cantonal doctor is
sufficient in order to avoid cumbersome administrative procedures. But now, with
computerized IS, this reasoning is not valid any more. The regulation committee must
REMO2V'06 823
analyse this initiative and, maybe, to decide to transmit for modifications of primary
or secondary legislation.
4.4 Regulation Concerning Grounds of Activities
Any person competent in observing ontology model can take initiative to elaborate
new kinds of activity with respect of the grounds. In our example (Fig. 2), the
cantonal doctor decides to accept or not the demands. Why it is a decision and not
only a control? Indeed, with an IS it is now easy to create a datawarehouse for the
role of cantonal doctor and therefore this role can control the activities of pharmacist
and doctor and eventually to disqualify them in case of failures.
The ground of a decision – the decision is written in the law – could have political
background: the majority of the Assembly thought that only one person must be in
charge of this type of decision. Besides, this ground could be implicit: due to the
paper-oriented organization it was normal to think that only the cantonal doctor must
take the decision to avoid cumbersome administrative procedures. The regulation
committee must analyse this initiative and, maybe, to decide to transmit for
modifications of primary or secondary legislation.
4.5 Participation
The information technology creates new kinds of situation, which were impossible to
foresee before, in particular in the case of promulgating laws. But, these situations
exist and cannot be hindered. How to do with them? In our example, one question
appears: who is allowed to observe the information stored in the IS? Any citizen? Is it
possible for any citizen who wants to propose an amendment to the law to have access
to the IS and to perform some data mining processes? At another level, is it possible
for any citizen to have access to the ontology model and to the conceptual model to be
able to propose consistent evolution at any level? All these questions concern e-
Participation in the e-Government field. But they are broader and relevant for any IS.
5 Conclusion
Regulation is a major concept for any system but it is much more relevant for open
systems than for the closed ones. In the traditional IS approach, the domain of
initiatives is much more a closed system than an open one. Due to the progress of
information technologies and also due the IS stakes and their impact, it is impossible
to stay in this position in a democratic country. This is particularly true in the domain
of e-Government in order to avoid autocratic decisions with severe impact on the
Society. Considering the IS initiatives as an open system, we proposed in this paper
the notion of the regulation committee. The role of the regulation committee is to
work with an objective model from where several opinions can be established
consistently. Such a model is proposed to be an ontological one. Besides, we
824 Regulations Modelling and their Validation and Verification
described in this paper several major situations of the regulation committee activities
and illustrated them with a case of e-Government.
References
1. Henderson, J., Venkatramen, N. (1992) Strategic Alignment: A Model for Organisational
Transformation, in: Kochan, T., Unseem, M. (Eds.), Transforming Organisations. Oxford
University Press, New York.
2. Khadraoui, A., Turki, S., Aïdonidis, C., Léonard, M. (2004) Institutional Information
Systems Alignment on Laws. Proceedings of the International Conference on Advances in
Intelligent Systems Theory and Applications, AISTA 2004, In cooperation with the IEEE
Computer Society, Luxembourg.
3. Krippendorff, K. Krippendorff’s Dictionary of Cybernetics. Last accessed 2006-04-17.
http://pespmc1.vub.ac.be/ASC/indexASC.html
4. Serres, M. (2001) Hominescence, Ed. Le Pommier.
5. Turki, S., Aïdonidis, C., Khadraoui, A., Léonard, M. (2004) Ontologies for Institutional IS
Engineering. Proceedings of the Open INTEROP Workshop On Enterprise Modelling and
Ontologies for Interoperability (EMOI - INTEROP 2004). Co-located with CAiSE'04
Conference, Riga, (Latvia).
6. von Bertalanffy, L. (1968) General System theory: Foundations, Development, Applications.
George Braziller New York.
7. Wikipedia, (2006) http://en.wikipedia.org/.