=Paper=
{{Paper
|id=Vol-2383/paper3
|storemode=property
|title=Phenomenological Ontology Guided Conceptual Modeling for Model Driven Information Systems
|pdfUrl=https://ceur-ws.org/Vol-2383/paper3.pdf
|volume=Vol-2383
|authors=Tomas Jonsson,Håkan Enquist
|dblpUrl=https://dblp.org/rec/conf/vmbo/JonssonE19
}}
==Phenomenological Ontology Guided Conceptual Modeling for Model Driven Information Systems==
https://ceur-ws.org/Vol-2383/paper3.pdf
Phenomenological Ontology Guided Conceptual
Modeling for Model Driven Information Systems
Tomas Jonsson1 and Håkan Enquist2
1
Genicore AB, Göteborg, Sweden
tomas@genicore.se
2
Department of Applied IT, University of Gothenburg, Sweden
hakan.enquist@gu.se
Abstract. Langefors infological equation states the infological perspec-
tive that data alone is not information, but can give rise to information
in the minds of people, if data is presented within a frame of reference
i.e. knowledge or perception of reality, in these minds. In a model driven
information system, the model directly defines the structure of data pre-
sented to its users and should therefore be founded in users knowledge or
perception of reality. This paper describes a draft for a phenomenological
ontology for modeling executable conceptual models in accordance with
the infological equation.
Keywords: Executable Models, Information Systems, Ontology Driven Con-
ceptual Modeling, Ontology, Phenomenology, Conceptual Modeling
1 Introduction
Enterprise Information Systems (EIS) are IT-based Information Systems func-
tioning as an integral part of the enterprise actors work life i.e. integrated in
the socio-technical system of the organization. EIS should provide information
management support for enterprise actors according to their specific view of and
understanding of that enterprise i.e. with semantic consistency.
The old vision of semantic consistency between minds, conceptual models and
data in executing Information Systems (IS), in accordance with the Infological
Equation (IE) [1] may be within reach even for full scale life cycle support of
EIS.
IE : I = i(D, S, t)
I = conveyed information in specific actors mind
i = information function
D = set of data received by actor
S = actors pre-existing knowledge, perception of reality (structure of concepts)
t = time available to interpret data
II
When implemented appropriately, a semantic consistency will be achieved be-
tween minds of actors, codified concepts, terminology and data with the factual
data management and information in user interfaces of the EIS. Discrepancies
in semantic consistency decrease information quality and thus quality of actors
work, productivity and effectiveness.
A major issue of EIS Life Cycle management in the digitization of work life
is how to effectively achieve and maintain semantic consistency, with accept-
able cost and time constraints. Model driven generation and execution of EIS,
from conceptual models of the enterprise information, has potential to improve
productivity, decrease error rate and costs at orders of magnitude.
The conceptual models play a fundamental role in facilitating semantic con-
sistency since it must be valid both as a description of enterprise actors view
of their work and also constitute an accurate and traceable design blueprint for
the EIS.
Implementing conceptual modeling guidance by application of a phenomenol-
ogy based ontology in an Ontology Driven Conceptual Modeling (ODCM) instru-
ment could be a pragmatic way to capture enterprise actors views in modeling
for EIS design.
A systematic literature review on ODCM [2] does however indicate a lack
of an ontology for executable conceptual models, with the purpose of modeling
perception of reality (IE:S) in such a way that information systems will commu-
nicate relevant data (IE:D) about this reality.
Our research scope is based in seamless life cycle management of EIS, which
are semantically consistent with their users view of the enterprise and its envi-
ronment. This paper focus on phenomenology as an ontological foundation for
ODCM in that context.
2 Pragmatic foundations and knowledge needs
Core Enterprise Architecture Framework (CoreEAF) include method, tools and
language for developing model driven EIS, based on conceptual models of the
enterprise. CoreEAF has been applied for successful and proven EIS life cycle
management, in large scale information systems (1500+ users) [3] over a period
of 25 years. Lately, it has also been applied for successful design and maintenance
of EIS for the social network enterprise Project Lazarus as demonstrated at ER
2018 [4]
Development of Phenomenological Foundational Ontology (PFO) is an at-
tempt to further develop, formalize and disseminate the conceptual modeling
approach in CoreEAF, by merging the pragmatics of CoreEAF modeling with
matching theoretical contributions, e.g. Husserlian phenomenological philosophy
and resent findings in ODCM research.
III
3 Phenomenology, Ontology and EIS Modeling
PFO is a novel ontology for ODCM, different from the most referred-to ontolo-
gies [5] Bunge Wand Weber (BWW) and Unified Foundation Ontology (UFO),
in that modeled entities are phenomena representing mind objects, based on
comprehension of existence in a metaphysical sense, Lifeworld (Lebenswelt) as
in the philosophical school of phenomenology, with philosophers such as Husserl,
Heidegger and Schutz. Lifeworld is that which is perceived as existing in reality,
perception of phenomena and the foundation for shared human experience.
If accepting the Lifeworld definition in phenomenological philosophy as the
foundation for shared experience, and in context of EIS for sharing information
about enterprise experiences, between actors of an enterprise, we must ask the
question: What are these objects of shared experience, how do we capture them
and how do we describe them? In order to answer the question we start with a
brief summary of a part of phenomenological philosophy.
Fig. 1. Unified Theory of the Husserlian System
In Logical Investigations [6] and later works, Husserl lays out the founda-
tion for phenomenological philosophy. Woodruff-Smith [7] summarize the Uni-
fied Theory of Husserlian System as depicted in Fig.1. Our focus of interest is
the right side of the figure, the Formal Ontology and the Material Ontology Re-
gions, indicating the existence of two orthogonal ontologies. Formal Ontology,
the structure of (mental) existence as objects and Material Ontology Regions as
mental (perceptual) regions to which these objects belong.
In this Husserlian System there is no explicit notion of time, however Hei-
degger’s work Being and Time [8] extends phenomenological philosophy with
aspects of time as a fundamental ingredient.
Starting from Husserlian System we added the time aspect, both to formal
ontology and to ontology regions. In the formal ontology a state graph is added
representing state concepts and transitions between these concepts. In the ontol-
IV
ogy regions, the region of Nature is further specialized into static physical world
and temporal physical world.
Fig. 2. CoreEAF ontological system with Model Language and Ontology Regions in
relation to the Husserlian Ontological System. Concepts augmenting the Husserlian
System in blue text and dotted lines.
Some other additions and modifications were made to transform formal on-
tology into an ontology for conceptual modeling languages, such as the notion
of object type and formal expressions to augment definitions of property and re-
lation concepts. The elements object type, property, relation and state represent
concepts which can be labelled and defined in a structure given by the language
ontology.
An overview of CoreEAF ontological system and it’s relation to the Unified
Theory of Husserlian System is depicted in Fig.2, where Phenomenological On-
tology Regions represents that what should be modeled (using the language) and
is the starting point for designing PFO.
Phenomenology of Husserl and others was from the beginning focusing on
perception of reality of individuals (as first person view), not in social con-
texts such as societies and enterprises. The work of Schutz [9] and others, put
phenomenology into the context of social worlds, adding the notion of intersub-
jectivity, a shared worldview of phenomena among groups of individuals. When
applying PFO in the context of modeling EIS, it is specifically this shared world-
view which is of interest, modeling the types of phenomena which are shared and
communicated about, between actors related to an enterprise.
V
4 PFO and Phenomenon Kinds
Basic components of PFO are phenomena, properties, relations and state con-
cepts. In this paper, we focus only on ontology regions and phenomenon kinds.
Ontology for phenomenon abstractions, relations, properties and state concepts
are still subjects for research and empirical evaluation.
Phenomenon kinds are grouped in four regions, Fig.3, representing world-
view awareness domains, building outwards from the most concrete awareness
of existence, the static physical world. We use the term physical object to refer
to that which exist outside the mind, observable through perception.
Static physical world: Phenomena as mind items, representing that which
is considered to exist in a static physical world. Physical here means, that outside
minds, which could be observed. However, could be observed, does not mean that
something has to exist for observation, it can be just thought of or imagined to
exist as observed.
– Things: Predominantly inanimate physical objects which are not considered
to act, have their own will.
– Actors: Predominantly animate physical objects and groups of such objects,
e.g. people, organizations, animals, herds. Also automata such as automated
machines, robots and information systems could be considered actors, if they
seem to act i.e. take actions.
– Localities: Concepts of location e.g. spatial relations, position, area or vol-
ume, coordinates in a coordinate system defining position, area or volume.
Fig. 3. Phenomenon kinds, their world-view awareness domains, combined with a col-
oring scheme used in modeling tool and runtime environment.
Temporal physical world: A physical world which changes over time. All
phenomena have a past, a present and a future, but when we want to understand
and reason about change, change itself is the phenomenon.
– Events: Phenomena of change. Events can be observed only when they hap-
pen, are in progress, but as mind items they can exist as plans before and
memories after they are observed.
VI
Social World: The social world is regulated, driven, by agreements between
actors in relation to the temporal physical world.
– Agreements: Phenomena of relationships between actors, possibly also with
relations to other phenomena included in the agreement. Agreements are
either informal, undocumented, subconscious or formalized and documented,
e.g. legal systems or written contracts.
Intentional World: A world of intentionality which motivates, drives, gives
purpose for individuals and organizations.
– Value: In the perspective of an actor, related to social and physical world,
describing that which is value and possibly its measure. For actors as indi-
viduals, Maslows hierarchy of needs indicates a starting point for possible
values. For profit making enterprises, value is money but also other values
are considered, such as customer value.
Additional Entity Kinds: The following two kinds of entities are not con-
sidered to be pre-conceptional mind items from the point of view of phenomeno-
logical philosophy. They are conceptual abstractions, which however play an
important role as entities in conceptual modeling for information systems.
– Message: As in the widest sense, generally unstructured (non modeled) in-
formation, usually as textual or image data. E.g. email, documents, pictures,
books in their non-physical sense i.e. in the digital universe.
– Generic: Should not be used when modeling according to PFO, except when
a phenomenon type represents an abstraction of two different phenomenon
kinds, e.g. phenomenon sales item can be either something physical (thing)
or service (event).
5 Continued Research
– Extending the description of PFO foundation, related to empirical work
and theoretical support for PFO for ODCM in literature and contemporary
research.
– Enrichment of PFO with regards to ontological detail, conceptual modeling
language implementation and modeling rules to support is use by practition-
ers.
– Development and integration of PFO in ODCM enabled EIS life cycle man-
agement.
– Exploring relationships between PFO and other ODCM approaches based
on different theoretical foundations and their consequences for model driven
information systems.
VII
References
1. Langefors, B.: Infological models and information user views. Information Systems,
5(1), 1732, 1980,
2. Verdonck, M.; Gailly, F.; de Cesare, S.; Poels, G.: Ontology-driven conceptual model-
ing: A systematic literature mapping and review. Applied Ontology, 10(34), 197227,
2018
3. Jonsson T.; Enquist H.: CoreEAF – a Model Driven Approach to Information Sys-
tems. Proceedings of CAISE Forum 2015, p137-144, 2015
4. Jonsson, T.; Enquist, H.: Phenomenological Ontology Guided Conceptual Modeling
for Enterprise Information Systems. In Advances in Conceptual Modelling (Vol.
11158, pp. 3134) 2018. Xian China: Springer.
5. Verdonck, M.; Gailly, F.: Insights on the Use and Application of Ontology and
Conceptual Modeling Languages in Ontology-Driven Conceptual Modeling. In: ER
2016. Lecture Notes in Computer Science, vol 9974. Springer, Cham (2016).
6. Husserl, E.: Logische Untersuchungen. (1900-1901, second edition 1913 and 1921).
7. Woodruff Smith, D.: Pure logic, ontology, and phenomenology. Revue internationale
de philosophie. 2144 (2003).
8. Heidegger, M.: Being and Time, trans. J.Macquarrie and E. Robinson, SCM Press
(1962) (first published 1927)
9. Schutz, A.: The Phenomenology of the Social World, Trans. G. Walsh F. Lehnert,
Northwestern University Press (1967) (first published 1932)