INTEGRATION ARCHITECTURES BASED ON SEMANTIC

WEB SERVICES: FAD OR MODEL FOR THE FUTURE?

Findings of a Comprehensive SWOT Analysis

Daniel Bachlechner

University of Innsbruck, Innrain 52, Innsbruck, Austria

Keywords: Semantic Web services, integration architectures, industrial adoption, SWOT analysis.

Abstract: Web services brought about a revolution by taking a remarkable step toward seamless integration of

distributed software components. The importance of Web services as a cornerstone of service-oriented

integration architectures is recognized and widely accepted by experts from industry and academia. Current

Web service technology, however, operates at the syntactic level and, hence, requires human interaction to a

large extent. Semantic Web services pledge the automation of core Web service tasks, such as discovery,

selection, composition, and execution, thus enabling interoperation between systems and keeping human

intervention to a minimum. Within the scope of this work, we discuss the capabilities of integration

architectures based on Semantic Web services as well as relevant environmental factors. The discourse is

based on the findings of a SWOT analysis that was conducted in early 2007. In order to best assess the

relevance and applicability of integration architectures based on Semantic Web services in an organisational

context, particular importance was attached to differences in the viewpoints of practitioners and researchers.

1 INTRODUCTION

Many enterprises employ multiple mission-critical,

best-of-breed application systems from different

vendors with different technologies and platforms

(Hohpe and Woolf, 2005). They chose the best

vendor for every operational area and connected the

products via the interfaces they provided. This

approach, however, normally leads to highly

complex systems. Nevertheless, until recently, this

strategy was considered a silver bullet when

assembling business software.

Together with mergers and acquisitions,

reorganizations, and leadership changes, which also

cause significant impact on IT infrastructures, best-

of-breed solutions lead to extreme heterogeneity.

Obviously, the operation of such patchworks is

extremely complex and costly. The maintenance of

numerous vendor relations and the necessity of

specific know-how usually are not justifiable.

However, the integration of application systems

within organizations and across organizational

boundaries is essential to realize competitive

advantages. Even if just a few critical systems

cannot share their data effectively, they create

information bottlenecks that often require human

intervention to be solved. Only with properly

deployed integration architectures can organisations

focus their efforts on their value-creating core

competencies.

Web services brought about a revolution by

taking a remarkable step toward seamless integration

of distributed software components. The importance

of Web services as a cornerstone of service-oriented

integration architectures is recognized and widely

accepted by experts from industry and academia.

Current Web service technology, however, operates

at the syntactic level and, hence, still requires human

interaction to a large extent. Semantic Web services

(SWS) pledge the automation of core Web service

tasks, such as discovery, selection, composition, and

execution, thus enabling interoperation between

systems and keeping human intervention to a

minimum (McIlraith et al., 2001; Fensel and

Bussler, 2002; Terziyan and Kononenko, 2003).

However, with respect to SWSs, there seems to

be a gap between research trends and industrial

needs (Sollazzo, 2002). Within the scope of this

work, we discuss the capabilities of integration

architectures based on SWSs as well as relevant

environmental factors. The discourse is based on the

findings of a SWOT analysis that was conducted

Bachlechner D. (2008).

INTEGRATION ARCHITECTURES BASED ON SEMANTIC WEB SERVICES: FAD OR MODEL FOR THE FUTURE? - Findings of a Comprehensive

SWOT Analysis.

In Proceedings of the Tenth International Conference on Enterprise Information Systems - SAIC, pages 45-52

DOI: 10.5220/0001718200450052

 SciTePress

within the scope of a Delphi study (Bachlechner,

2007). In order to best assess the relevance and

applicability of integration architectures based on

SWSs in an organisational context, particular

importance was attached to differences in the

viewpoints of practitioners and researchers.

In section 2, we describe the design of the

underlying analysis. While sections 3 and 4 present

and discuss the results of the SWOT analysis in

detail, we conclude with a summary of major

findings in section 5.

2 APPROACH

The underlying Delphi study was conducted at the

University of Innsbruck in early 2007. The main

goal of the study was to collect and quantify the

opinions of clearly defined groups of practitioners

and researchers on the potential of SWSs as basis for

integration architectures that enable organisations to

link their data processing systems efficiently. It was

expected that an understanding of the relevance and

applicability of SWS-based integration architectures

would help to align future research efforts with

industry needs effectively. Another goal was to

make participating experts from academia and

industry more sensitive to the progress and focus of

SWS research in general.

2.1 Survey Design

Within the scope of the study, the participating

experts were provided with two questionnaires. The

first questionnaire contained open-ended questions

designed to capture the experts’ views concerning

factors potentially affecting the relevance and

applicability of SWS-based integration architectures.

The responses from the first round were aggregated

into groups and classified by the unique issues that

best summarized their contents. The second

questionnaire was based on the responses from the

first round. The participants were asked to review

the aspects identified in the first round and rank

them on structured bipolar rating scales ranging

from 1 to 5, with 1 representing Strong

Disagreement and 5 representing Strong Agreement.

Two rounds were expected to be sufficient to attain a

first impression concerning the opinions of the

experts.

The Delphi study consisted of four parts,

structured and formalized in a way that allowed for

various analyses: a SWOT analysis, a requirements

analysis, an analysis of expected effects, and a

technology roadmap. Within the scope of this work,

we focus exclusively on the results of the SWOT

analysis. Its purpose was twofold: firstly, it was used

to compare SWS-based integration architectures and

traditional approaches with respect to their

capabilities; secondly, it was used to collect

information on relevant environmental factors.

2.2 Questionnaire

The SWOT analysis helped to assess the relevance

and applicability of integration architectures based

on SWSs by analyzing their strengths, weaknesses,

opportunities, and threats. With regard to SWS-

based integration architectures, we wanted to know

the strengths and weaknesses that people with

industrial and academic backgrounds associate with

them. What makes them better than or inferior to

traditional approaches? What external factors drive

or restrict their use?

The questions stated within the scope of the

SWOT analysis read as follows:

 Where do you see the strengths of integration

architectures based on SWSs?

 Where do you see the weaknesses of integration

architectures based on SWSs?

 What factors do you think will drive the use of

integration architectures based on SWSs in the

future?

 What factors do you think will restrict the use

of integration architectures based on SWSs in

the future?

2.3 Expert Panel

The candidates were selected from academia and

industry in similar proportion. Repeated

involvement at major conferences and publication in

at least one of the relevant fields were two of the

main criteria used to find suitable representatives of

the target population. The candidates were

exclusively people involved in at least one of the

major international conferences related to SWSs and

associated technologies, enterprise integration

architectures and middleware solutions, and book

authors or members of widely recognized initiatives

active in at least one of the related research fields.

The 38 experts who participated in both rounds

of the study were from all parts of the world and

affiliated with major universities and enterprises.

While 21 of the experts had academic backgrounds,

17 had industrial ones. These numbers correspond

well with Clayton’s (1997) recommendation for an

adequate panel size. The expertise of the participants

ICEIS 2008 - International Conference on Enterprise Information Systems

in the area of research was gathered to evaluate their

technical qualification for the study. The expertise

distribution grouped by backgrounds is shown in

Figure 1. The scale ranged from 1 to 5, with 1

representing Novice and 5 representing Expert. None

of the participants ranked in the lowest category.

Beginner Competent Proficient Expert

Industry

Academia

Figure 1: Expertise distribution.

2.4 Survey System

Web-based surveys provide capabilities far beyond

those available for any other type of self-

administered survey technique. They can be

designed in ways that facilitate a dynamic

interaction between respondents and the survey

system, which is of particular interest for Delphi

studies (Dillman, 2004). Web surveys offer

enormous potential for conducting research.

However, it always must be kept in mind that a

survey corresponds to a level of technical

sophistication that makes it possible for most users

to respond. Within the scope of the development of

the survey system, programming and design steps

were taken to minimize the differences across

respondents caused by different operating systems,

Web browsers, and screen configurations.

3 RESULTS

In the following, we discuss the results of the SWOT

analysis. The respondents rated up to 40 statements

with respect to each of the questions stated within

the scope of the SWOT analysis. The respondents

were free to leave statements unrated or to check a

No Comment box. The most important statements

from respondents with either academic or industrial

backgrounds are presented in tabular form.

Statements are defined as Most Important if the

mean of their ratings is greater than or equal to 4

(i.e., on average, the experts at least agreed to the

statement). The most controversial statements

comparing the two groups of respondents are

illustrated by means of net diagrams. The five

statements in which the difference of the means of

the two groups of respondents is maximal are

defined as Most Controversial.

3.1 Strengths

Tables 1 and 2 list the most important strengths of

integration architectures based on SWSs from either

an academic or an industrial perspective. Improved

service discovery capability and facilitated

interoperability are the most important strengths,

according to both groups of respondents.

From an academic point of view, improved

mediation between services, enhanced process and

term definitions, and the formalization of systems

also play key roles. From respondents with an

industrial viewpoint, improved mediation between

services and enhanced process and term definitions

got average ratings of only 3.86 and 3.67,

respectively. Interestingly, the formalization of

systems is also among the most controversial

statements with regard to strengths of integration

architectures based on SWSs. This is due to the fact

that respondents with industrial backgrounds gave

the statement an average rating of 3.40 which is

significantly lower than the one of respondents with

academic backgrounds.

Table 1: Most important strengths from an academic

perspective.

Mean

Improved service discovery capability 4.24

Facilitated interoperability 4.20

Improved mediation between services 4.10

Enhanced process and term definitions 4.06

Formalization of systems 4.00

Table 2: Most important strengths from an industrial

perspective.

Mean

Facilitated interoperability 4.36

Improved service discovery capability 4.33

Facilitated reuse of services 4.07

Improved service composition capability 4.00

The facilitated reuse of services and improved

service composition capability also are among the

most important strengths of SWS-based integration

INTEGRATION ARCHITECTURES BASED ON SEMANTIC WEB SERVICES: FAD OR MODEL FOR THE

FUTURE? - Findings of a Comprehensive SWOT Analysis

architectures from an industrial point of view. The

facilitated reuse of services, with an average rating

of 3.95, also is quite important from an academic

point of view.

Figure 2 shows the most controversial strengths

of SWS-based integration architectures comparing

the two groups of respondents. Respondents with

industrial backgrounds perceive the goal-based

paradigm as a clear strength, with an average rating

of 3.93, while respondents with academic

backgrounds consider the statement rather neutral,

with an average rating of 3.16. Conversely,

respondents with academic backgrounds perceive

the formalization of systems, compliance with

business and legal rules, improved service

choreography capability, and facilitated system

upgrades as key strengths, while respondents with

industrial backgrounds rate them only slightly above

average.

Figure 2: Most controversial strengths.

3.2 Weaknesses

Tables 3 and 4 list the most important weaknesses of

integration architectures based on SWSs from

academic and industrial perspectives. The use of

immature technologies and the description overhead

are the most important weaknesses, according to

both groups of respondents.

Respondents with academic backgrounds

perceive high initial start-up costs, the lack of

agreement on the description depth, and high system

complexity as particularly important weaknesses.

Interestingly, the lack of agreement on the

description depth and the high system complexity

both also are among the most controversial

statements. This is due to the fact that respondents

with industrial backgrounds rated them with average

scores of 3.31, significantly lower than the

respondents with academic backgrounds. The high

initial start-up costs also are not perceived as a very

serious weakness by the respondents with industrial

backgrounds. The statement got an average rating of

only 3.54 from practitioners. High service

development costs, unsatisfactory support of change

management, and the labour-intensive service

specification also are among the most important

weaknesses from an academic viewpoint. The

labour-intensive service specification with an

average rating of 3.92, also plays a key role for

experts with an industrial point of view. The

unsatisfactory support of change management and

the high service development costs with average

ratings of 3.58 and 3.17, respectively, play only

minor roles for practitioners.

For respondents with industrial backgrounds,

the facts that SWS-based integration architectures

have not yet been adopted and that software

engineers are not ontology experts represent

additional, and particularly important, weaknesses.

With average ratings of 3.68 and 3.89, respectively,

both also are relevant for researchers.

Table 3: Most important weaknesses from an academic

perspective.

Mean

High initial start-up costs 4.42

Lack of agreement on description depth 4.38

Use of immature technologies 4.26

High system complexity 4.21

Description overhead 4.16

High service development costs 4.05

Unsatisfactory support of change

management

4.05

Labour-intensive service specification 4.05

Table 4: Most important weaknesses from an industrial

perspective.

Mean

Use of immature technologies 4.23

Not yet adopted 4.15

Software engineers are not ontology

experts

4.08

Description overhead 4.00

While respondents from both groups agree on

weaknesses such as the use of immature

technologies, description overhead, and labour-

intensive service specification, there also are some

controversies. Figure 3 shows the most controversial

weaknesses of integration architectures based on

SWSs comparing the two groups of respondents.

ICEIS 2008 - International Conference on Enterprise Information Systems

Figure 3: Most controversial weaknesses.

Respondents with academic backgrounds rate the

high degree of formality, unintuitive concepts, and

high service development costs as rather important

weaknesses. From an industrial perspective, these

aspects are rated only slightly above average.

Instead, respondents with industrial backgrounds

perceive as a key weakness the fact that integration

architectures based on SWSs have not yet been

adopted. Furthermore, the lack of a dominant design

seems to be a serious issue from an industrial

viewpoint, while it does not seem to be relevant for

researchers. However, the latter controversies are

not among the top five.

3.3 Opportunities

Tables 5 and 6 list the most important factors driving

the use of integration architectures based on SWSs

from either an academic or an industrial perspective.

The need for service interoperability and the

availability of business cases, compliant middleware

implementations and effective tools are the most

important driving factors, according to both groups

of respondents.

Interestingly, the factors that are perceived to

drive the use of integration architectures based on

SWSs differ significantly between the two groups of

respondents. Proven cost-effectiveness, a compelling

value proposition, and increasing dynamics of

cooperation are among the most important factors

driving the use of integration architectures based on

SWSs, according to the respondents with academic

backgrounds. With an average rating of 3.69, proven

cost-effectiveness is a significantly less important

factor for respondents with industrial backgrounds.

The difference of the means is large enough to make

it the most controversial statement with respect to

factors driving the use of SWS-based integration

architectures. Increasing dynamics of cooperation

and the compelling value proposition, with average

ratings of 3.73 and 3.80, respectively, are only

slightly less important from an industrial point of

view than from an academic viewpoint. Researchers

also attach particular importance to such factors as

the increasing support from standardization bodies

and the proliferation of services. From the point of

view of practitioners, with average scores of 3.69

and 3.77, respectively, both are less important.

Preceding agreement on standards, buy-ins from

large integration players, potential savings, and

increasing dynamics of systems are among the most

important factors driving the use of SWS-based

integration architectures from the point of view of

respondents with industrial backgrounds.

Interestingly, increasing dynamics of systems also is

among the most controversial statements with regard

to driving factors. This is due to the fact that

respondents with academic backgrounds rated the

statement with an average score of 3.58,

significantly lower than the respondents with

academic backgrounds. The other factors, preceding

agreement on standards, buy-ins from large

integration players, and potential savings, with

average ratings of 3.95, 3.68, and 3.74, respectively,

are significantly less important from the perspective

of a researcher as compared to that of a practitioner.

Table 5: Most important opportunities from an academic

perspective.

Mean

Availability of business cases 4.42

Proven cost-effectiveness 4.33

Availability of compliant middleware

implementations

4.21

Increasing dynamics of cooperation 4.16

Availability of best practices 4.16

Need for service interoperability 4.05

Compelling value proposition 4.05

Increasing support from standardization

bodies

4.00

Availability of effective tools 4.00

Proliferation of services 4.00

Factors such as the availability of integrated

development environments and methodologies also

play key roles for SWS-based integration

architectures from an industrial viewpoint. However,

while the availability of methodologies, with an

average rating of 3.84, is only slightly less important

for respondents with academic backgrounds, the

availability of integrated development environments

got an average rating of only 3.68.

INTEGRATION ARCHITECTURES BASED ON SEMANTIC WEB SERVICES: FAD OR MODEL FOR THE

FUTURE? - Findings of a Comprehensive SWOT Analysis

Table 6: Most important opportunities from an industrial

perspective.

Mean

Need for service interoperability 4.46

Preceding agreement on standards 4.33

Availability of effective tools 4.31

Buy-in from large integration players 4.25

Potential savings 4.18

Increasing dynamics of systems 4.18

Availability of business cases 4.15

Availability of compliant middleware

implementations

4.15

Availability of integrated development

environments

4.15

Need for effective collaboration 4.08

Need for flexible integration 4.08

Availability of methodologies 4.00

Practitioners also perceive the needs for effective

collaboration and flexible integration to be important

drivers regarding the use of integration architectures

based on SWSs. Researchers, however, merely are

in accord with respect to the need for flexible

integration. It got an average rating of 3.95. With

respect to the importance of the need for effective

collaboration, participants with academic

backgrounds are less confident. The average rating

is only 3.68.

Proven cost effectiveness.

Increasing

dynamics of

systems.

Cooperation

across

industries,

academia and

interest

organizations.

Preceding

globalization.

Consolidated

pattern algebra.

Industry

Academia

Figure 4: Most controversial opportunities.

Figure 4 shows the most controversial factors

driving the use of integration architectures based on

SWSs comparing the two groups of respondents.

From an academic point of view, cooperation across

industries, academia and other interest organizations,

as well as preceding globalization, play much more

important roles than they do for participants with an

industrial viewpoint. In contrast, the consolidated

pattern algebra is perceived as an important

opportunity only from practitioners.

3.4 Threats

Tables 7 and 8 list the most important factors

restricting the use of integration architectures based

on SWSs from academic and industrial perspectives.

Interestingly, not a single factor is among the most

important ones for both groups.

The difficulty in describing semantics and the

unavailability of convincing case studies are by far

the most important restricting factors for researchers.

With average ratings of 3.69 and 3.58, respectively,

both are significantly less important for

practitioners. However, despite the clear difference

regarding the average means, neither statement is

among the most controversial. From an academic

point of view, the increasing complexity, unproven

cost-effectiveness, high costs, failure to reach

critical mass, and limited consideration of business

needs play key roles with respect to factors

restricting the use of SWS-based integration

architectures. With scores of 3.85 and 3.77,

respectively, the average ratings of practitioners

regarding the limited consideration of business

needs and unproven cost-effectiveness are quite in

accord with those of the researchers. The other

factors, increasing complexity, failure to reach

critical mass, and high costs, with average ratings of

3.62, 3.54, and 3.33, respectively, are significantly

less important from the perspective of a practitioner

as compared to that of a researcher. Finally,

respondents with academic backgrounds attach

particular importance to the lack of skilled

developers as well as the lack of integration into

middleware technologies. From respondents with an

industrial viewpoint, they got average ratings of 3.69

and 3.83, respectively.

By far, the lack of effective tools is the most

important factor restricting the use of integration

architectures based on SWSs for practitioners. With

an average rating of 3.94, the lack of effective tools

also is quite important from an academic point of

view. Still, quite important factors from an industrial

perspective are the lack of industrial commitment

and the limited interest of vendors. Interestingly,

respondents with academic backgrounds rate both

statements with average scores of 3.35 and 3.17,

respectively, only slightly above average. Due to the

large differences of the means, both statements are

among the most controversial with respect to factors

restricting the use of SWS-based integration

architectures. The difficulty to catalyze the market

and the fact that the dominant vendors use their own

technology also play major roles with respect to

restricting factors from an industrial point of view.

ICEIS 2008 - International Conference on Enterprise Information Systems

From the perspective of researchers, both are of

lower importance than for practitioners, with

average scores of 3.77 and 3.61, respectively.

Table 7: Most important threats from an academic

perspective.

Mean

Difficulty of describing semantics 4.39

Unavailability of convincing case studies 4.22

Increasing complexity 4.11

Unproven cost-effectiveness 4.11

High costs 4.06

Failure to reach critical mass 4.06

Limited consideration of business needs 4.06

Lack of skilled developers 4.00

Lack of integration into middleware

technologies

4.00

Table 8: Most important threats from an industrial

perspective.

Mean

Lack of effective tools 4.23

Lack of industrial commitment 4.08

Limited interest of vendors 4.08

Market does not understand values and

capabilities

4.00

Difficulty of catalyzing the market 4.00

Dominant vendors use own technology 4.00

The two groups of respondents regard quite

different aspects of the threats to integration

architectures based on SWSs as important.

Figure 5: Most controversial threats.

Figure 5 shows the most controversial factors

restricting the use of SWS-based integration

architectures comparing the two groups of

respondents. From an academic point of view, the

lack of semantic annotations and the heterogeneity

of workflows and business processes play much

more important roles than they do for experts with

an industrial point of view. In contrast, the lack of

funding is perceived as an important threat only

from respondents with industrial backgrounds.

4 DISCUSSION

So far, we have shown that there are several more or

less agreed capabilities of integration architectures

based on SWSs. They were called strengths and

weaknesses. We also described environmental

factors relevant to the use of SWS-based integration

architectures. The factors that drive the use of SWS-

based integration architectures were called

opportunities. We also mentioned threats that

describe factors restricting the use of integration

architectures based on SWSs. Throughout this work,

we assigned opinions either to people with academic

or with industrial backgrounds. The former also

were called researchers and the latter practitioners.

The analysis was characterized by a continuous

comparison of hypothetical SWS-based integration

architectures and traditional integration approaches.

At the end of the day, it is the practitioners who

decide about the adoption of integration

architectures in industry. Researchers are supposed

to deliver technologies meeting the requirements of

the practitioners. The perceived strengths of SWS-

based integration architectures are worth knowing,

but the weaknesses are what researchers have to

focus on. The weaknesses are the limitations, faults,

or defects that keep an approach, such as an

integration architecture, from achieving its purpose.

The same applies to the environmental factors.

Exploiting opportunities is desirable, but countering

threats is essential. Focusing on the weaknesses and

threats ultimately helps to find an answer to the

question whether SWS-based integration

architectures are a fad or a model for the future.

Researchers and practitioners agree that SWS-

based integration architectures reduce

interoperability problems and facilitate enterprise

integration by improving such selected SWS usage

activities as discovery, mediation, and composition.

Enhanced process and term definitions, as well as

explicit definitions of service conditions and

functionalities based on ontologies, allow for a

better understanding of the used systems. It also is

agreed that the use of immature technologies

currently is one of the most important weaknesses of

integration architectures based on SWSs. One of the

key problems regarding evolving or immature

technologies, particularly within highly complex

INTEGRATION ARCHITECTURES BASED ON SEMANTIC WEB SERVICES: FAD OR MODEL FOR THE

FUTURE? - Findings of a Comprehensive SWOT Analysis

systems, is that no one understands the risks that

come with their use.

Weaknesses that are perceived as important, such

as the lack of agreement on description depth,

description overhead, and the labour-intensive

service specification, allude to one and the same

issue. There is a need to focus on this within the

scope of the research on SWS frameworks. It may

be reasonable to depart from the idea that generic

frameworks work for all SWS applications. It seems

important to find the right balance between the

satisfaction of high knowledge requirements and the

avoidance of description overhead. Sivashanmugam

et al. (2003) discuss the description of Web services

based on shared ontologies.

Interestingly, the costs incurred by setting up and

maintaining SWS-based integration architectures are

perceived as a weakness by researchers rather than

by practitioners. Nevertheless, there is no doubt that

the cost-effectiveness of SWS-based integration

architectures needs to be proved. Making disparate

systems share information cost-effectively is a key

problem for companies, and represents billions of

euros in technology spending, with a high

percentage of worldwide IT budgets dedicated to

enterprise integration projects. De Brujin et al.

(2005) describe a SWS framework that aims at

enabling flexible and cost-effective integration.

Unlike researchers, practitioners perceive a lack

of industrial commitment and only limited interest of

vendors. Furthermore, practitioners are aware that

the market principally does not understand the

values and capabilities of integration architectures

based on SWSs. This explains why it is difficult to

catalyze the market and why dominant vendors use

their own technologies. Interestingly, researchers

also are aware of this situation and know that their

consideration of business needs is limited. The

unavailability of convincing case studies and best

practices can be viewed as a direct consequence of

the lack of target group orientation.

5 CONCLUSIONS

With respect to many aspects, the picture of

integration architectures based on SWSs looks quite

different from an academic point of view than from

an industrial viewpoint. We hope that this discourse

helps to take a first step toward closing the gap

between research trends and industrial needs and,

subsequently, to exploit the full potential of SWSs

within the scope of integration architectures.

However, more could be done in this area.

Additional studies that address further issues would

be valuable to make more accurate conclusions. For

instance, it would be interesting to evaluate specific

SWS frameworks with respect to their relevance and

applicability for integration architectures.

Furthermore, based on the results of this work, best

practices that define the configuration of an SWS-

based integration architecture, could be formulated

for specific organisational environments.

ACKNOWLEDGEMENTS

Thanks go to the experts who contributed to the

success of the underlying study. Without the insight

they shared, it would not have been possible to

assess the potential of SWSs satisfactorily.

REFERENCES

Bachlechner, D., 2007. Relevance and applicability of

Semantic Web services in electronic business: a

Delphi study. University of Innsbruck. Innsbruck.

Clayton, M.J., 1997. Delphi: a technique to harness expert

opinion for critical decision-making tasks in

education. In Educational Psychology. 17(4): 373-386.

de Bruijn, J., Fensel, D., Keller, U., Lara, R., 2005. Using

the Web service modeling ontology to enable semantic

e-business. In Comm. of the ACM. 48(12): 43-47.

Dillman, D., 2000. Mail and Internet surveys. John

Wiley & Sons Inc. New York.

Fensel, D., Bussler, C., 2002. The Web Service Modeling

Framework WSMF. In Electronic Commerce

Research and Applications, 1(2): 113-137.

Hohpe, G., Woolf, B., 2005. Enterprise integration

patterns: designing, building, and deploying

messaging solutions. Addison-Wesley. Boston.

McIlraith, S.A., Son, T.C., Zeng, H., 2001. Semantic Web

services. In IEEE Intelligent Systems. 16(2): 46-52.

Sivashanmugam, K., Verma, K., Sheth, A., Miller, J.,

2003. Adding semantics to Web services standards. In

Proc. of the 1st International Conference on WS.

Sollazzo, T., Handschuh, S., Staab, S., Frank, M., 2002.

Semantic Web service architecture - evolving Web

service standards toward the Semantic Web. In Proc.

of the 15th International FLAIRS Conference.

Terziyan, V., Kononenko, O., 2003. Semantic Web

enabled Web services: state-of-Art and Industrial

Challenges. In Web Services - ICWS-Europe 2003.

ICEIS 2008 - International Conference on Enterprise Information Systems