TOWARDS A SEMANTICALLY AUGMENTED

COLLABORATIVE WORKING ENVIRONMENT

Designing a Service Oriented Architecture Environment for eProfesionals

Antonio Ruiz-Martínez, M. Antonia Martínez-Carreras and Antonio F. Gómez-Skarmeta

Department of Information and Communications Engineering, University of Murcia, Murcia, Spain

Keywords: eProfessional, Collaborative Working Env

ironment (CWE), interoperability, semantic.

Abstract: Both social and economic changes have favored the appearance of eProfessionals as workers whose

business and tasks can only be achieved using modern cooperative technologies. To be able to achieve this

vision there is a need to research and develop new collaborative working environments. This is the main

goal of the ECOSPACE project. As part of the research in this project, in this paper, we introduce an

architecture based on Service-Oriented Architecture (SOA) which has been improved by means of the

inclusion of semantic information. This semantic information allows us to achieve a better level of

interoperability than we can provide only by using the syntactic information provided by SOA technologies.

1 INTRODUCTION

Within few years, significant social, organizational

and economic changes as well as a relentless

technology evolution will lead the way of working

for eProfessionals into a dramatic change. People

will no longer work according to chain production

models but rather more as dynamically and

spontaneously assembled groups of people working

together in a collaboration mode, which means a

seamless work to achieve common goals. In the

academic research community, these trends lead to

the Social Computing, Social Desktop, and Social

Web or Web 2.0 initiatives.

The “Experts Group on Col

laborative Working

Environments” of the European Commission coined

the term eProfessional to address workers in such

environments. We consider an eProfessional as a

professional worker whose business and tasks can

only be achieved using modern cooperation

technologies. These technologies enable an

eProfessional being part of groups and communities

as well as knowledge networks, and being involved

in distributed cooperation processes that have not

been possible before.

According to this assumption, our vision is that

“Th

e Network” (in fact the Internet and the web of

people) will become a global virtualized

collaborative workplace where the contextual social

exchange will be located through people-concepts

connectivity. In a recent study it is shown that

workplace innovations account for 89% of multi-

factor productivity gains (source: Black and Lynch,

San Francisco Federal Reserve, 2004). This global

virtualized collaborative workplace which is one

goal of ECOSPACE will enable knowledge workers

to get access to both, their individual shared

workspace and groups or communities shared

workspaces wherever they are, whenever they need

it independent of organizational boundaries.

To be able to achieve this vision there is a need

o research and develop new collaborative working

environments as it is reflected by the groups of

experts of the European Commission in the idea of

CWE 2020 (Laso-Ballesteros et als 2006) (Laso-

Ballesteros 2005). This research must lead to a better

understanding of the work environment, the

development of an upperware and collaboration

services as the collaboration platform, and new

innovative user-centric collaborations tools that

reduce the complexity of today’s techno-centric

communication applications.

A critical element in an increasingly

terconnected world will be the software that will

implement most functionality and will also ensure

the secure and reliable integration and

interoperability of mobile, fixed, personal and

corporate heterogeneous resources and applications.

Ruiz-Martínez A., Antonia Martínez-Carreras M. and F. Gómez-Skarmeta A. (2007).

TOWARDS A SEMANTICALLY AUGMENTED COLLABORATIVE WORKING ENVIRONMENT - Designing a Service Oriented Architecture Environment

for eProfesionals.

In Proceedings of the Second International Conference on e-Business, pages 31-38

DOI: 10.5220/0002110200310038

 SciTePress

This will require advanced technology for the

development of highly functional and high-quality

software, and a truly semantically enabled

middleware infrastructure for its interoperability.

Well accepted standards and open-source software

are also crucial.

In this context the ability of organizations to

embrace change, uncertainty and opportunities in the

global economy are directly related to the

empowerment of knowledge workers

(eProfessionals) at all levels of the organizations.

This means making collaboration with colleagues

down the corridor or around the world as natural as

working alone. It means making access to

information secure, ubiquitous and unobtrusive. In

that sense it is our intention to contribute to the

European industry by means of defining secure,

robust, common and open infrastructure standards

for collaboration based on a middleware opened for

seamless integration of components and their

convergence into new types of systems and

environments.

This middleware will be aligned with the

methods, services and tools in order to define a

collaborative support for combining these

technologies into the defined architecture. In order

to reach this goal, ECOSPACE major objective is

the integration of existing platforms provided by

partners in the project (Prinz et als 2006). They have

been selected to represent state of the art systems in

collaborative environments. At the same time we

propose new approaches that will enable the

integration of existing technologies across different

usage environments, but also support the new

concepts and services for an eProfessional

workplace according to the paradigm shift advocated

by the project.

The objective of ECOSPACE is to define a

robust upperware that:

• Allows interoperation with open standards and

contributes to them, to facilitate the integration

of a range of technologies and platforms,

including existing ones provided by partners as

well from developers outside the consortia.

• Support the integration, sharing and distribution

of content and concepts develop in the project

based on semantic information in mediation

between the structure and content of knowledge

bases from different living cases.

• Allow the support of mobile, nomadic and fixed

professionals in a seamless approach, as well as

presence and awareness services that enable the

integration of the context of users and their

collaborative activities.

• Support security, trust and privacy, in the access

and use of the services and tools defined over it.

In this project we extend classical middleware

vision with the adoption of a semantic-based

upperware to manage the interoperability, access and

integration of distributed data and resources, and to

provide a platform for value-added services. This

middleware will contribute and collaborate in the

definition of open standards for business and

collaborative service processes.

The middleware that we will develop is able to

structure and focus these needs, adding coordination

and knowledge management. This middleware

allows changes and adjustments to both the

organizational structure and the coordination rules

depending on how the global knowledge of the

collaborative environment evolves along the time.

It is important to stress that the middleware will

be compound by several layers that allow the

independence of the computing paradigm specific

aspects and at the same time provide the abstraction

to access several components distributed in the

network to offer the iterated view of the

collaborative services required by the user. This will

imply the location, discovery, naming, and in

general invocation of the different distributed

components.

As different computing paradigms need to be

integrated in the same middleware (i.e. service

oriented computing, P2P computing, Web 2.0

methods), the middleware should allow to work

either on centralized and totally decentralized

environments. Thus we need to address the aspect of

synchronization and persistence for distributed

workspaces. Further, appropriate protocols for

knowledge dissemination and recovering must be

defined to allow for an appropriate context

information management.

At the ECOSPACE defines and develops the

user-centric interoperability middleware and new

core collaboration services: e.g. persistent

distributed workspaces, group management,

collaboration context and resource location.

The rest of this paper is organized as follows.

Section 2 provides a description of the main feature

that the ECOSPACE CWE should provide. Then, in

section 3, we describe the initial steps that we

propose to achieve the goals commented in the

previous section. Then, in section 4 we mention

some considerations about the involvement of the

end-users in order to improve the results proposed.

Finally, in section 5 we finish the paper with some

conclusions and future work.

ICE-B 2007 - International Conference on e-Business

2 VISION OF THE ECOSPACE

CWE

Future eProfessional collaborative working

environments requires a shift from application

oriented developments towards the design of

collaboration-aware work environments that support

cooperation and interaction in terms of activities

instead of technical functions.

ECOSPACE aims to support this shift by

applying a paradigm shift from simple cooperation

services – currently often stand-alone applications –

to user and activity oriented cooperation

environment. ECOSPACE will develop such an

environment based on the integration of existing

services that the project partners contribute to the

project, and which will be also open to include third-

party services. For this purpose, a reference model

and middleware for collaboration environments will

be developed. On top of this middleware, innovative

collaboration tools will be developed to increase

cooperation awareness for users and tools (Figure 1).

The few basic components for such a

cooperation environment is formed by existing

cooperation services such as email, shared

workspaces and application sharing or task

management services. Among the new services,

presence and awareness services will play an

important role. These services are needed in

distributed cooperation to support users in their

mutual understanding of the status and progress of

work as well as the work rhythms of other

organizations.

On top of these services an open integration layer

enables the horizontal and vertical integration of the

services. It supports the interoperability of similar

services, e.g. two shared workspace systems from

different vendors, of complementary services, e.g.

user management between workspace system and

instant messaging system. An important pre-

requisite for the realization of such a layer is the

development of interoperability standards.

The integration platform integrates existing

services on a technical level. This includes the

exchange of authentication data (single sign login)

as well as metadata of collaboration artifacts. In

addition new core collaboration services are

developed that are needed for the collaboration

tools. On top of the service integration layer an

activity and collaboration integration layer provides

the user-centric integration of the collaboration

service. This enables the organization of resources

according to users’ collaboration context.

The activity and process support level abstracts

from the cooperation application to provide an

activity-oriented collaboration environment. Within

this environment, users can organize their resources

according to their processes, activities, teams and

communities. I.e. the documents and messages

exchanged within a project will no longer be

scattered over the attachments of emails in email

folders, the local disk and a shared file system or a

shared workspace. Based on a semantic integration

of the cooperation activities as well as the services,

users will be able to organize the environment

according to their project, team or community

contexts.

Figure 1: From Cooperation Services to Cooperation

Aware Environments.

3 TOWARDS TO

THE ECOSPACE CWE

In the previous section, we have introduced the

ambitious goals required to the ECOSPACE CWE.

Next, in this section, we provide the different steps

followed in order to go towards this CWE.

In order to define a basic interoperable

infrastructure for the collaboration, one important

initial step is to analyze the architectures and

interfaces defined in the different already existing

platforms to define a cooperative architecture being

able to cope with the different models. ECOSPACE

has a significant number of systems provided by the

partners such as shared workspaces (BSCW, BC,

SAP NetWeaver KMC,…), virtual presence

(Jaytown, OpenScape) and conference toolkit (Arel

Spotlight, Isabel).

With the aim of establishing an interoperable

activity-oriented collaboration environment we need

to move from the standalone applications approach

that presents the previous collaboration tools to the

services approach. Therefore, we need to define the

architecture as Service-Oriented Architecture (SOA)

TOWARDS A SEMANTICALLY AUGMENTED COLLABORATIVE WORKING ENVIRONMENT - Designing a

Service Oriented Architecture Environment for eProfesionals

as we describe in the following section. For this

architecture we are based on the initial results

proposed in (Martínez-Carreras et als 2006), where

the authors presented a generic architecture based on

SOA. However, in this architecture we lack the

incorporation several components, such as the need

ones to discover activities as well as the

incorporation of semantic information in order to

provide a better level of interoperability.

3.1 Service Oriented Architecture

For this new collaborative working environment we

propose the decomposition of the different tasks that

can be carried out by a CWE using a set of web

services (WS). Therefore, in this way we have a

Service-Oriented Architecture. Between the main

advantages of the SOA approach we can point out:

the easy integration of applications with other

systems, the adaptation of the applications to

changing technologies, the reuse of code, the

creation of business processes from the existing

services in a faster way, better scalability, etc.

Then, according to this approach all new services

are developed as web services. These web services

are based on several technologies such as SOAP,

WSDL, UDDI, etc. The access to a web service is

based on the exchange of XML messages according

to the definition of the service which is provided by

means of the WSDL language. These service

definitions are published in a repository in order to

easily find where the services, that satisfy some

requirements to carry out a task, are located. The

repository could follow the UDDI specifications or

other that is considered adequate. UDDI provides

services for the description, discovery and

integration of web services. However, this

specification is not as broadly implemented as other

specification related to web services such as SOAP

or WSDL.

For the exchange of messages to access to the

service we could follow the SOAP protocol or the

REST protocol. SOAP is largely extended and has

become a de facto standard used for communicating

heterogeneous platforms due to it is transport-

independent. Therefore, it is used as the main

protocol to access to web services. Other important

of SOAP is that it can be used several transport

protocols, both synchronous ones, such as HTTP,

HTTPS, SMTP…, and asynchronous ones, such as

JMS. When this protocol is used, we recommend the

definition of web services according to the document

style because it fits better to SOA. More details can

be found in (Erl 2004). On the other hand, REST is

being used more and more in order to create

lightweight access to the web services from the Web

2.0 technologies such as AJAX. Although for Web

2.0 are proposed this lightweight protocol, it is

important point out that SOAP web service could be

also used from Web 2.0, using a Web 2.0 technology

such as AJAX. From our point view, for the

definition of web services in this architecture both

approach can work together. Then, in order to

support them in a unified way we proposed the use

of WSDL 2.0 since it supports the definition of both

kinds of services.

However, as we have mentioned previously, we

already have an important amount of tools that are

not described as web services. The migration of

these tools to the new approach would suppose a lot

of effort and sometimes is not possible. In order to

overcome this situation, we propose the use of

wrappers, called Wrapper Web Services (WWS)

(see Figure 2), with the aim of encapsulating the

functionality already offered by means of these

tools. Next, we provide a detailed description of

these services.

3.1.1 Wrapper Web Services (WWS)

A Wrapper Web service is a web service that

encapsulates a legacy or non-SOA application.

There are several ways to develop this sort of

service. It depends on the API provided by the

legacy or non-SOA application. Most of the tools

provide a HTTP interface that is based on the use of

servlets, php, XML-RPC … Thus, the wrapper web

service is a web service that receives SOAP or

REST requests according to a WSDL specification

and creates a request according to the API provided

by the tool. Once, the web service receives the

response from the tools, it creates the response as a

web service response. The following figure depicts

this interaction.

WS-Client

Client Application

WWS

Non-SOA application

Collaborative Application

1. WS-Request

2. Non-SOA

Request

3. Non-SOA

Response

4. WS-Response

WS-Client

Client Application

WWS

Non-SOA application

Collaborative Application

1. WS-Request

2. Non-SOA

Request

3. Non-SOA

Response

4. WS-Response

Figure 2: Wrapper Web Service.

Both WS and WWS are the base of the

architecture that we propose because they can

provide the desired interoperability between

heterogeneous platforms and applications.

The whole design of our architecture is shown in

Figure 3. The Service layer represents both the

Cooperative Services and the Service Integration

ICE-B 2007 - International Conference on e-Business

layers that we commented previously in Figure 1.

On the one hand, it represents the Cooperative

Services layers because it offers collaborative

services from the different existing applications. On

the other hand, it represents the service integration

layer thanks to the definition of web services that

allow us to achieve the interoperability between

applications and platforms developed with different

technologies. In that layer, we can see that there are

some collaborative services that have been

developed according to this approach. The new web

services are only represented as a box with the term

WS. Additionally, the web services that represents

the adaptation of existing applications to this new

approach are represented under a box named WWS.

For example, so as to depict how we have included

the shared workspace named BSCW as a web

services through a wrapper web service, we have put

a box with WWS on the top of BSCW.

At present, in ECOSPACE project we have

several wrappers services that encapsulate part of the

functionality of groupware such as shared

workspaces, e-mail and Ldap. Among these services

we have forum, document management, calendar, e-

mail, Ldap, etc.

Moreover we have a service of presence and

availability developed without using wrappers for

invoking it.

We also point out that the new web services

developed following this approach could use another

web services or even other wrapper web services in

order to make a task. This interaction is shown with

an arrow in that layer.

3.2 Composition of Services

As we commented in Section 2, in any collaborative

working environments it is required the provision of

collaboration based on activities. With the previous

layer we provide only a set of interoperable services.

With the aim of reflecting the complexity of

activities in any environment, composition and

orchestration of the various collaboration services

have to be considered.

For this purpose we propose the introduction of a

new layer called Composite Collaborative Services

(CoCos) layer. This layer is on top of Service layers

as we can see in the Figure 3. The goals of this layer

can be achieved by means of standards such as WS-

CDL (Web Services Choreography Description

Language) (Kavantzas et als 2005), WS-BPEL (Web

Services Business Process Execution Language)

(Alves et als 2006) or WSMO (Web Services

Modeling Ontology) (Roman et als 2005). These

different languages allow us to model the behaviour

of business processes or activities and the messages

to exchange in order to achieve the goals of the

processes. The exchange of messages can be seen

from the point of view of a particular user or from

the point of view of all participants in the exchange

of messages. These standards could be even

combined in order to perform a particular business

process. Then, these languages allow the definition

of both sequences control between activities as

concurrency and synchronization between them.

In this layer, a CoCos represents the combination

of several services, in some specific way, in order to

carry out a specific task or activity. In order to

discover the different services that a CoCos may

need, we can make use of the service repository that

we commented in the previous layer. The task

description carried out by a CoCos is stored in the

Activity repository to facilitate its search. Hence,

this new CoCos can be used in other services and/or

CoCoS.

With this layer we satisfy the goals mentioned in

Section 2. This functionality was shown in Figure 1

as the activity based collaboration support.

Based on the WWS previously mentioned, in this

project, we have created several CoCos as a proof of

concept. Some of them are: the event notification to

several users by means of e-mail or the creation of a

forum to discuss on a document that has been

uploaded to the workspace. Currently, these CoCos

are described by means of WS-BPEL.

3.3 Applications

From the services and the CoCos we can develop

new collaborative applications. These applications

are represented in a new layer on top of CoCos

layer. These new applications follow the SOA

approach. Then, taking into account the Model-

View-Controller pattern, the new applications will

be centred in the view and the controller. On the

other hand, the access to the different functionality is

provided by the services and the CoCos. This kind

of applications is represented as the squared box in

Figure 3.

However, we have non-SOA application. In this

case, the question will be how we can enhance them

and take advantage of this architecture. The

approach to follow depends on the kind of

application. In proprietary applications we have to

find out if the new functionality could be provided

by means of plug-ins. In this case, we would develop

a plug-in that would access to web services.

Otherwise, it would not be possible. On the other

hand, in open source applications could be possible

TOWARDS A SEMANTICALLY AUGMENTED COLLABORATIVE WORKING ENVIRONMENT - Designing a

Service Oriented Architecture Environment for eProfesionals

to recode some part of the code in order to

incorporate the use of web services.

In this project, as a proof of concept we have

developed an AJAX application that allows the users

to use some of the CoCos previously mentioned to

start the discussion on a document among a set of

users.

Figure 3: SOA-based CWE Architecture.

3.4 Semantic Augmented CWE

Until this moment we have commented a basic

infrastructure based on SOA that be developed in

order to achieve a minimum level of cooperation

between heterogeneous collaborative working

environments. However, in the basic Service

Oriented Architecture (SOA), one weakness in

adopting web services is its lack of semantic

information.

The web services technologies SOAP, WSDL,

UDDI etc, rely exclusively on XML for

interoperations, but the structural XML guarantees

only syntactic interoperability. Expressing message

content in XML allows web services to parse each

other’s message but does not allow semantic

‘understanding’ of the message content. In order to

overcome this limitation, the efforts in the Semantic

Web hold great promise of making the Web a

machine-understandable infrastructure. In the line

within this project in order to be able to integrate the

technologies and platforms and generate a new range

of services available through the middleware, we

adopt the extension of the classical middleware

based on semantic information.

Nowadays there exist different proposals for

including semantic in web services. Then, we

propose to extend the previous architecture (Figure

3), that only supports interoperability at syntactic

level, with the advantages of semantic solutions like

WSDL-S (Web Services Semantics) (Akkiraju et als

2005), OWL-S (OWL-based Web Services

Ontology) (Dean et als. 2003) or new proposals like

WSMO (Web Services Modelling Ontology)

(Roman et als 2005). These technologies allow

formal representation of content to the web services

specification and allow the semantic in the

interactions and capabilities. This improves the

retrieval of accurate information (services or CoCos)

regarding the users’ preferences

Then, as a final result what we generate is a

collaborative SOA (c-SOA) that extends WS-I

(Motahari et als 2006) recommendations and some

of the existing proposals for peer-to-peer service

composition and collaboration such as WS-CDL,

WS-BPEL or WSMO that we commented

previously. We also propose to improve these

proposals with semantic information so as to

improve both their search according to the user’s

preferences and the composition of these composite

services into high-level composite services. This

semantic information is included by means of

ontologies as mentioned below.

Therefore, we add semantic information to the

services layers obtaining a semantic web service

(SWS), as it is represented in the Figure 4, on top of

WS element. On the other hand, we extend the

information provided in the definition of CoCos with

semantic. This is also shown in Figure 4, on top of

CoCos, we have a new box called CoCos with

semantic.

Both in web services and composite services the

semantic information is provided by means of

ontologies. These ontologies are described using

proposals such as RDF(-S) or OWL, Finally, in

order to support this semantic c-SOA, in this project

it is been working in the analysis and the definition

of the ontologies needed for the description of the

services mentioned. Subsequently, these ontologies

will be used to improve semantically the description

of services and CoCos previously described.

Figure 4: Semantically augmented SOA-based-CWE.

ICE-B 2007 - International Conference on e-Business

This semantic information included in this

architecture comes to support the semantic

integration that was commented as one of the goals

of future CWE mentioned in Section 2.

From this semantic information we are able to

compose and create new applications in a more

intelligent way and achieving a better result in the

end-user satisfaction.

4 USER INVOLVEMENT

The experimentation and evaluation methodology

across the different lab settings will allow

comparison and mutual learning. Each experience

and application research cycle consists of selecting

and defining the specific real-life scenario and later

the project to be used in the living lab. The

technology has then to be prepared for the

experimentation with e.g. the functionality modules,

data required, and user configuration. As

experiments and longitudinal cases only become

meaningful if users adopt new collaborative working

methods and know how to use the environment

productively, training becomes a crucial element.

5 CONCLUSIONS AND

FUTURE WORK

This paper describes the motivation and objectives

of one the central activities within ECOSPACE

project in order to define an augmented middleware

for a collaborative environment, to support not only

interoperability between existing platforms but also

to allow real shift in the collaborative integration

based on a semantic layer that abstract the patterns

of interactions and look for the most adequate

mechanism to implement them. The initial steps in

the development of this middleware have been

introduced in Section 3 of this paper.

This collaboration environment will enable

knowledge workers, and especially eProfessionals,

to easily network together, form groups and

professional virtual communities for stimulating

creativity and innovation while increasing

productivity.

Therefore, we have presented a collaboration

platform reference architecture and corresponding

upperware enabling the interplay and

interoperability of collaboration services and tools in

a collaboration environment. This architecture also

facilitates the business process management, mobile

and wearable computing. This result will be

contributed to standards and will foster the seamless

cooperation of users within and between

organizations, teams and communities.

As future work, there is a long way of research in

order to satisfy the full objectives of this project.

Thus, context rules requires a deeper research in

order to establish how these are going work with the

different semantic services and activities and taking

into account user’s preferences. It is needed a clear

definition of these context rules in collaboration

services. Finally, the models to define

communication between asynchronous and

synchronous applications and services are required.

This way will be able to cooperate between services

based on synchronous modes, such as RPC, and

others that are based on messages-event, such as

MOM.

ACKNOWLEDGEMENTS

This work has been partially funded by the project

ECOSPACE “Integrated Project on eProfessional

Collaboration Space” (FP6 IST-035208).

REFERENCES

Alves, A., Arkin, A., Askary, S., et als. 2006 "Web

Services Business Process Execution Language

Version 2.0", OASIS, August

Akkiraju, R., Farrell, J., Miller, J. et als. Web Service

Semantics - WSDL-S. W3C Member Submission 7

November 2005

Dean, M., Connolly, D. van Harmelen, F., et als. 2003

"OWL web ontology language reference". W3C

Working Draft.

Erl, T. 2004 “Service-Oriented Architecture: A Field

Guide to Integrating Xml and Web Services”. Prentice

Hall Ptr. March.

Kavantzas, N., Burdett, D., Ritzinger, G., et als. 2005

"Web Services Choreography Description Language

Version 1.0", W3C Candidate Recommednation.

November 2005.

Laso-Ballesteros, I. 2006 “Enabling Productivity and

Innovation Through Empowered and Motivated

People”. Proceedings of 1

Conference on

Collaborative Working Environments for Business and

Industry. Brussels, Belgium

Laso-Ballesteros, I. ed. 2006 “New Collaborative Working

Environments 2020. Report on industry-led FP7

consultations and 3rd Report of the Experts Group on

Collaboration@Work”. European Commission Society

and Media.

Martínez-Carreras, M.A., Gómez-Skarmeta, A.F. 2006

“Towards Interoperability in Collaborative

Environments". Proceeding of 2nd IEEE International

Conference On Collaborative Computing:

TOWARDS A SEMANTICALLY AUGMENTED COLLABORATIVE WORKING ENVIRONMENT - Designing a

Service Oriented Architecture Environment for eProfesionals

Networking, Applications And Worksharing And

Workshops - CollaborateCom 2006.

Motahari, H. R., Benatallah, B., Casati, F. and Toumani,

F. ”Web Services Interoperability Specifications!,

IEEE Computer, May, pp 24-32

Prinz, W., Löh, H., Pallot, M., Schaffers, H., Skarmeta, A.,

and Decker, S. 2006 "ECOSPACE – Towards an

Integrated Collaboration Space for eProfessionals".

Proc. Collaboratecom2006. Atlanta. 17-20th Nov.

Roman, D., Keller, U., Lausen, H., et als 2005 "Web

Service Modeling Ontology", Applied Ontology, 1(1):

77 - 106.

ICE-B 2007 - International Conference on e-Business