A domain framework for representation of Web

A domain framework for representation of Web system impacts Norazlin Yusop, David Lowe, Didar Zowghi University of Technology, Sydney PO Box 123 Broadway NSW 2007, Australia {norazlin, didar}@it.uts.edu.au, [email protected]

Abstract. In web systems development, the business environment, processes and the related Web systems are interdependent. The business domain not only drives the identification of system needs, but the business is fundamentally changed by the introduction or evolution of the system. This results in Web systems that are highly volatile with complex inter-dependencies with the business domain. In this paper we describe a framework which allows us to describe and structure these interdependencies, particularly focusing on the concept of mutual constitution. The result is a consistent terminology and a basis for more effective reasoning about the business consequences of introducing and changing Web systems.

1 Introduction A key characteristic of many Web systems1 (as well as various other software systems) is that the introduction of the system has a fundamental impact on the nature of the business processes and business models which are being supported by the system. In other words, the business environment and processes not only drive the definition of the system needs, but are in turn fundamentally changed by the system. In effect, the business “problem” defines the nature of the technical system “solution” that is desired, but the solution itself changes the characteristics of the problem space, resulting in systems and business processes which co-evolve. This can be described as solutions and problems being mutually constituted - a concept well understood in the area of social informatics [1] and requirements engineering [2]. Whilst this concept of mutual constitution - or its expression as a technical solution leading to substantial changes in the problem space - is typical of almost all systems development, it is particularly significant with Web systems. This is due in large part to the scale and immediacy of the impacts that web systems can have on their domain. 1

Whilst we use the term Web system throughout this paper, we recognise that this terminology is still somewhat ill-defined within the literature (which includes terms such as Web applications, web-based systems, online systems, as well as a host of domain specific terms such as business-to-business (B2B), business-to-consumer (B2C), etc.). In this paper we use the term Web system to refer to those system which utilise web technologies as an integral element of a functionally complex system which typically incorporates interfaces beyond the organisational boundaries.

Whilst there has been a significant body of work looking at specific impacts of Web system, the results have not been drawn together into an overall framework. In this paper we describe a framework which allows us to describe and structure the impacts (and hence interdependencies) between Web systems and their organisational contexts. Such a framework can provide benefits such as ensuring a consistent terminology and a basis for more effective reasoning about the business consequences of introducing and changing Web systems.

2 Related Work There is a significant body of work that discusses specific impacts of Web systems and technologies. For example, [3] discusses ways in which e-commerce technology has a positive impact on the supply chain as it allows information visibility, which in turn allows companies to integrate their system components on production planning and scheduling and inventory control, and implement that through the supply chain thus getting into just-in-time service/product delivery. In [4], Osmonbekov, et al state that using electronic commerce tools such as Internet electronic data interchange (EDI), less costly extranets and intranets, web storefronts, web-based products, or service information can reduce the systems and transaction cost that the organisations will have to bear. These tools enable the buyers and sellers to get shared information on updated information pertaining to inventories, orders and shipments This is a positive impact for both parties as it eliminates the need to install and maintain expensive proprietary networks, (such as the old EDI networks) which leads to substantial cost savings which can effectively be passed on to the end user. Web systems development can pose negative impacts to the organisation and its systems. With new technology, such as web systems, new forms of threats are encountered regularly. These threats include intrusion and illegal access to proprietary data, virus attacks, denial of service attacks, and illegal acquisition or distribution of data that is copyrighted. Some of these problems are more critical with intranet system where there is lack of full implementation of firewalls and encryption technology [5]. The threats that the organisation face also impact on the organisation’s system, whereby the system needs to be safeguarded with sophisticated security measures that support mutual authentication, fine grain authentication, communication integrity, confidentiality, non-repudiation, and authorization. Security poses a high impact on the organisation and the system development because there is no security technology that can secure the system effectively [6] [7]. Another significant impact pertains to e-business interfaces. Common ebusiness interfaces refer to industry-level specifications for the interactions among IT applications at different enterprises which are needed to coordinate a business process involving multiple supply chain roles [8]. Some of the B2B interface specifications include Microsoft Biztalk, Ariba cXML, Open buying on the Internet (OBI), EDI/XML, OASIS ebXML, Common Business Library (CBL), and Information and Content Exchange (ICE) among others. These electronic interfaces impact on fundamental supply chain capabilities, which in turn would reduce misalignments between supply chain partners and increase the opportunities through flexible integration of supply chain processes. These supply chain capabilities are:

coordination capabilities; plug and play capabilities; and knowledge creation and sharing capabilities [8]. These e-business interfaces therefore have a positive impact on the organisation and its systems. The literature discussed above tends to highlight aspects such as the impact of electronic commerce tools on the reduction of business costs [4] and changes to business processes – particularly as they relate to aspects such as supply chains. These impacts can be both positive (such as increased customer satisfaction) and negative (such as increased vulnerability to security breaches). E-business interfaces are critical in any web system projects as it aligns the supply chain processes. However, it is also critical that organisations ensure that the different web applications and tools can be integrated seamlessly with the organisation’s system and the system of its trading partners. Any development of a new system, particularly web systems will pose a significant impact on the other system components that are already in place. Despite this, there has been little research on systemizing or structuring the nature of these impacts.

3 Dimensions of Impact From the web system impacts discussed in the literature (examples of which are shown in Table 1) it can be observed that different web systems have different impacts on the business environment. In this paper our focus will be on describing the characteristics of the domain of mutual influence – i.e. those characteristics of the system environment which affect the system, but are in turn affected by the system. We have identified six dimensions of these impacts, which are as follows: • Affecting Characteristics: The system characteristics which has an affect on its domain, or a domain characteristic which has an effect on the system; • Affected Characteristics: the characteristic of system or its domain which is affected. • Types of Web Systems: The types of system for which this impact applies. e.g. internet, intranet and extranet. • Facet of the Domain Impacted: Whether the characteristics that are impacting, and which are impacted upon, are part of the system, the organisation, or the external environment (e.g. O S indicates an organisational characteristic and a system characteristic which impact on each other). • Effect: whether the impact is viewed as having a positive or a negative effect on business operations; • Level: extent of the impact, e.g. high, medium, low.

Table 1: Example characteristics of the business environment that impact and/or are impacted by the web system Characteristics that affect

2

3

4

7

15

Supply chain intermediation and disintermediation in (6) Access levels differentiation (7) Customer value (8) Service personalization (10) Robots to monitor (11) promotion activities Multiple response generations (12) Real time access, easy to use interfaces, interactivity (14) Security threat to org. proprietary data (20) Internet infrastructure security threat (21) Management and improving organisation’s relationship with customer and supplier (38)

Characteristics that are affected

Value Exchange (9)

Scalability (13)

Usability (15)

Security (22)

Relationship Management (39)

Types of systems / Facet of domain / Effect / Level Extranet O S Positive Medium for all aspect EXCEPT High for value offering Internet/Extranet S S Negative High for System All S S Positive Medium (all aspects) All O S Negative High (all aspects) Extranet/ Internet O S Positive Medium (all aspects)

Sources

[11] [14] [16] [17]

[18] [19] [20] [21]

[22] [23] [24]

[4] [12] [25]

[15]

Key: Facets of the Domain Impacted: O S: Organisation characteristic that impacts on the System characteristic but which is in turn directly impacted by the system. Hence the organisation and system characteristics are mutually influenced, thus identified as being the characteristic in the Domain of Mutual Influence S S: System characteristic that are mutually influenced by another system characteristic Explanations for Level of Impact: High: Significant Impact. Impact that changes the business model, business process, customer-supplier relationships, value offering, organization structure and system Medium: Impact that changes some parts of the business model, business process, customer-supplier relationships, value offering, organization structure and system Low: Minimal impact. Impact that does not change the business model but causes a minimal change to the business process, customer-supplier relationships, value offering, organization structure and system

EXTERNAL ENVIRONMENT

ORGANISATION (Internal Environment)

DOMAIN OF CONTEXT: Environment characteristics which impact on the design of the system

DOMAIN OF MUTUAL INFLUENCE: Environment characteristics which impact on, and which are impacted by, the design of the system

SYSTEM

SYSTEM

DOMAIN OF EFFECT: Environment characteristics which are impacted by the introduction of, or change to, the system.

Fig. 1. Domain of Context, Domain of Influence, Domain of Effect Table 1 shows an example of data compiled using this framework, which specifically depicts the characteristics of the domain that are mutually influenced. The numbers in brackets ( ) displayed on each characteristic correspond to the numbers in Figure 2. The numbers displayed on this table is not in sequence as this table is extracted from a complete table, which is discussed in detail in [9] to illustrate characteristics of the domain which are mutually influenced (shown in the Facets of the Domain Impacted column). It is necessary to include the numbers this way for the purpose of crossreferencing in Figure 3.

4 Impact framework As discussed above, we have argued that a Web system and its problem domain are mutually constituted - i.e. they are inter-dependent such that they impact each other. This problem domain involves characteristics of both the organisation hosting the system and external environment that relates to it. These characteristics can be grouped into several sub-domains, which are useful in understanding and reasoning about the inter-dependencies. These are shown in Figure 1. The Domain of Context is made up of those characteristics of a systems domain, which provide a context for the system (i.e. affect the design of the system in some direct and immediate way). For example, both existing business processes and a legal regulatory framework may affect the design of the system. The Domain of Effect

is made up of those characteristics of a systems domain which are impacted upon in some direct and immediate way by the introduction of the system. For example, both existing workflows and customer expectations will often be impacted upon by the system, and are therefore part of the domain of effect. The Domain of Mutual Influence is the intersection of the domain of context and the domain of effect - i.e. it is made up of those characteristics of a systems domain which both impact the design of the system, and which are potentially impacted upon by the introduction of the system. Taking the previous example, existing business process are likely to be part of the domain of mutual influence (since they influence the system design, but will also be changed by the system) but legal frameworks will not (since, whilst they affect the system design, they are unlikely to be directly impacted upon by the system). 4.1 Domain of Mutual Influence From the Impact Framework described above, the characteristics of the Domain of Mutual Influence are those characteristics of the domain which affect the system, but which are in turn directly and immediately affected by the system (e.g. business processes, or – for most web systems – the external customers). Some of the key characteristics are discussed below. Value exchange processes and activities between the organisation and its intermediaries are identified as a characteristic of the organisation that impact on the system. In e-business, process of intermediation and disintermediation presents specific e-business risks for traditional sellers [11]. For example; intermediaries such as brokers might easily disappear or re-appear in an e-business environment. This is because buyers might decide to use a middle person (intermediary) to increase their combined buying power and seek better negotiations with the sellers. Due to this issue, it is important to show who is doing what with whom as e-business is more flexible [11]. This will have significant impacts on the e-business project if there is no differentiation in the value offered by the different entities. An important issue in ecommerce business design is the assignment of value activities to actors. Web systems offer the customers a distinct customer service value, compared to traditional business systems. The types of customer service offered via the web systems include answering customer enquiries, providing search and comparison activities, providing technical information to customers and allowing customers to track order status [10]. Most banks and financial institutions allow their customers to view their account balance, status of stock portfolio, loan application anywhere and at any time via an extranet. As shown in Table 1, row 2, the facets of the domain impacted are labelled as O S, which explains that the characteristics of the organisation facet and the characteristics of the system facet are mutually influenced. Integration of different web applications and tools can lead to complex problems for an organisation and its trading partners. It is therefore important that different systems within the organisations and with external parties are able to interoperate with each other. Poor integration of the web applications and tools will lead to negative impact for both the organisation and the system. Any development of a new system, particularly web systems will pose a significant impact on the other system components that are already in place. It is critical that the system components

are able to interoperate with each other. The issue of interoperability has shifted from the level of applications to the level of standards. A trading partner has to deal with several standards at the same time. In case where one trading partner exchanges messages across industries, the variety of standards is likely to increase even more. One solution to deal with such problem is through the use of B2B protocol and integration engines. These execute actual message exchanges according to various standards [6]. Bussler et al [13] state that B2B e-business integration is a complex issue and web services are used as a solution for many significant integration problems. There is currently little research or lessons learned that report on the deployments that could advance web services technology functionality. SOAP (Simple Object Access Protocol) is the implementation technology for web services. SOAP messages can be easily exchanged over the Internet. As such, SOAP can be proposed to be a superior way to implement B2B interactions between companies (inter-enterprise) [13].

5 Impacts and the Domain Boundary: Real world examples As discussed above (and shown in Figure 1) we can identify the domain of context, the domain of effect, and the intersection of these two – the domain of mutual influence. In order to identify where the various characteristics lie with respect to these domains we have developed the Impact and Domain Diagram (refer to Figure.2). Let us illustrate the diagram with two real world cases which were analysed from the literature. Figure 2 shows that characteristics 1, 2, 3 and 4 are characteristics of the organisation which impact on characteristic 5, which is a characteristic of the system domain.. In a real world example discussed in Turban et al [10] Adaptec, Inc. is a large microchip manufacturer supplying critical components to electronic equipment makers. The organisation needed to improve on its product delivery cycle time as it required up to 15 weeks delivering products to customer, while competitors took only 8 weeks. The longer delivery time was mainly caused by the need to coordinate design activities between Adaptec, Inc. headquarters in California and its three principal fabrication factories in other countries. In order to solve this problem, Adaptec, Inc. adopted extranet and enterprise-level supply chain integration software, which incorporates automated workflow and EC tools. E-mail was used to communicate with manufacturers across several time zones, thereby automatically starting the flow of raw materials. This in turn reduced invoicing and shipping times. Due to the organisation’s need to reduce design cycle time, the implementation of the extranet and web applications impact on the organisation’s system. The system developers needed to ensure seamless integration of these applications and tools. As the implementation of the extranet and web application tools were a success, it therefore impacted positively on the organisation as it benefited from the reduced time required to generate, transmit and confirm purchase orders [14]

Fig. 2. Impact and Domain Diagram In another example, Figure 2 shows that characteristic 38 is a characteristic of the organisation that impact on characteristic 39, which is a characteristic of the system. Table 1 shows that management and improvement of organisation’s relationship with customer and supplier is a characteristic of the organisation (characteristic 38) that impact on process of relationship of management (characteristic 39) which is conducted and facilitated in the web system. Web systems facilitate relationship management activities and operations between customers and suppliers over the Internet and extranet. In a case study extracted from a literature pertaining to relationship management, Dignum [15] cited an experience report, which is conducted by Eindhoven University of Technology in cooperation with 11 partners in the region of Eindhoven. In a customer relationship management project embarked by a trucking company in Eindhoven, Netherlands, DAF trucks use the

Internet to get feedback from their customers on the product and on the transaction of performance. This means that the feedback from a customer can be processed together with the customer’s profile by the departments of the company that are most affected by the feedback. This has significant impact on the system, whereby the organisation’s information systems that are used internally have to be connected and synchronised, due to the use of an enterprise resource planning system throughout the company.

6 Implications for organisations From the framework and illustrated examples, therefore it is critical for clients to know and articulate the characteristics of the business domain that are going to impact on their organisation. In order to be able to articulate these characteristics, organisations need to first be aware of these impacts. Table 1 presented in this paper prompts clients and web system developers to think and be aware of these impacts. Pertinently, the dimensions presented in the table depict the extent and level of impacts that the characteristics will have on the business domain. The extent and level of impacts dimensions will guide the web system developers and their clients in evaluating if the web system can have a positive or negative impact and also if they will pose a high, medium and low impact on the organisation. Before developing a web system, clients and web system developers may use the characteristics and the dimensions of impacts presented on Table 1 as a catalyst in predicting the impacts and outcome of the web system project. It is critical that clients and web system developers first evaluate the identified characteristics and based on the extent and level of impacts, analyse in what way these impacts can affect on the business domain. For example, in developing an extranet web system for a client’s organisation, web system developers need to analyse in what way the client’s business domain characteristics are impacted. Table 1 has highlighted several characteristics, which have mutual impacts on the client’s business domain. Integration of different web system applications particularly, has a high level of impact on the client’s system. Issues such as interoperability problems of different web applications and tools due to different level of standards, for example, can cause significant problem to the client’s business domain. It is therefore critical that web system developers investigate these issues before developing a web system development project. The framework presented will guide clients and web system developers to clearly visualise the different domain that can be impacted by the development of the web system. The dimensions of impact and framework presented in this paper can also guide clients and web system developers in mapping the characteristics that are most likely to impact on the client’s business domain. By mapping these characteristics, the clients and web system developers will be able to evaluate the aspects of the domain impacted by the system and imperatively if the domain is impacted mutually. Characteristics of the domain that are mutually impacted are the ones that need to be carefully examined as these characteristics have a reverse impact and it highlights that a web system project can impact on two domains.

7 Limitations of Research In identifying inter-dependencies between web systems and their environment, we undertook a broadly-focused analysis of literature from diverse disciplines including: marketing; management; logistics; engineering; information systems; information technology; and manufacturing. These findings are then used to formulate a framework which comprises of six dimensions of the web systems impacts. As the literature was sourced from diverse disciplines, it may be seen as a limitation to this research as the results obtained are broad. However, the literature that we have selectively surveyed is rich in information and represents various disciplines which are critical to the development of web system. The generic types of impacts that have been collected and characterised from the literature become the basis of questions to be used in the second stage of data gathering, which is the case study. The case study is conducted to triangulate the data collected from the literature. A series of industrial case studies will be conducted on organisations that have adopted web systems for their businesses in recent years. There is a limitation to the case study in which we acknowledge that most web system development projects are developed within a short time frame and web system developers need to complete the project within a limited time as well. In order to obtain a richer data, the case study needs to ideally be a longitudinal case study, particularly looking at several domains. We attempt to address this limitation by focusing on organisations that have recently developed web based enterprise systems which have had a direct impact on their trading partners, such as the suppliers and customers.

8 Conclusion and Future Work In this paper we have described a framework for structuring information about Web impacts. The application of this framework in collating specific impacts is described elsewhere [9]. This framework is designed based on key characteristics of these impacts, and can support specific reasoning about these impacts. For example, once a data repository has been populated with information on potential impacts, it becomes possible to address questions such as: “if we make a change to the security support in the design of an intranet, then what aspects of the business operations might be most significantly affected?” Ongoing work is focusing on commercial case studies which are providing us with significant data on business impacts which can be structured according to this framework. This can subsequently support organisations in developing strategies for the co-evolution of web systems and business processes. The nature, extent and dimensions of the impacts identified from the organisations will serve as value-added knowledge for the companies in understanding how to best manage their Web development projects. The purpose of the case studies is to identify and analyse specific impacts from the web system development projects. In particular the goal is to develop a clearer understanding of the mutual dependency between application solutions and the organisational context of the web applications. It is also the goal of the study to reaffirm from practice the type and level of impacts found in the literature. In addition,

there may be more specific and fine-grained impacts and characteristics that will be discovered from the organisation that could complement the literature survey and hence the framework that we have developed in this study. A prototype tool is currently being developed from the repository of the knowledge captured in this paper. This tool aims at allowing industry users (for example, web system developers and their clients) to query for information such as the web systems impacts and their domains with ease. The premises of the tool are to help identify and assess the impacts and their extent, and to identify characteristics typical to various web systems, such as the Internet, extranet and intranet. This tool will also increase awareness of risks involved in design. Pertinently, it will assist in strategic decision making of various stages of web systems development.

References 1.

Sawyer, S., Rosenbaum, H.: Social informatics in the information sciences. Informing Science 3 (2000) 89–96 2. Carroll, J.: Scenarios and design cognition. In: International Requirements Engineering Conference, Essen, Germany (2002) 3–5 3. Golicic, S., Davis, D., T.M., M., Mentzer, J.: The impact of e-commerce on supply chain relationships. International Journal of Physical Distribution and Logistics Management 32 (2002) 851–871 4. Osmonbekov, T., Bello, D.C., Gilliland, D.I.: Adoption of electronic commerce tools business procurement: enhanced buying center structure and processes. Journal of Business & Industrial Marketing 17 (2002) 151–166 5. Scott, G., Walter, Z. : Management problems of Internet systems development: System Sciences, HICSS. Proceedings of the 35th Annual Hawaii International Conference, 7-10 Jan. (2002) 6. Medjahed, B., Benatallah, B., Bouguettaya, A. Ngu, A.H.H., Elmagarmid, A.K.: Business-to-business interactions: issues and enabling technologies, The International Journal on Very Large Data Bases, 12(1) (2003) 59-85 7. Yang, J., Papazoglou, M.P.: Interoperation support for electronic business, Communication of the ACM, 43(6) (2000), 39-47 8. Gosain, S., Malhotra, A., El Sawy, O., and Chehade, F.: The impact of common e-business interfaces. Communications of the ACM, 46(12), December (2003), 186-195 9. Yusop, N., Zowghi, D., Lowe, D.: An analysis of e-business systems impacts on the business domain. In: The Third International Conference on Electronic Business (ICEB 2003)., Singapore (2003) 10. Turban, E., Lee, J. King, D & Chung, H.M.: Electronic Commerce: A managerial perspective, Prentice-Hall (2000) 11. Gordijn, J. and Akkermans, H.: Designing and Evaluating E-business models, Intelligent Systems, 16 (4) (2001) 11-17 12. Murray, J.: Choosing the right web services management platform, Fawcette Technical Publications, (2003) Available online: http://devresource.hp.com/drc/topics/web_services_mgt.jsp#techpapers

13. Bussler, C.,Maedche, A. & Fensel, D.: Web services: Quo Vadis, IEEE Intelligent Systems, (2003) 14. Turban, E., Lee, J. King, D & Chung, H.M.: Electronic Commerce: A managerial perspective (2nd ed), Prentice-Hall (2002) 15. Dignum, F.: E-commerce in production: some experiences, Integrated Manufacturing Systems, 13(5) (2002) 283- 294 16. Gordijn, J., Akkermans, H., Vliet, H.V.: What is an electronic Business Model, Proceeding Knowledge Engineering and Knowledge Management, 12th International Conference (EKAW 2000), Springer Verlag, (2000) 257-274 17. John. I., Muthig, D., Sody, P. & Tolzmann: Efficient and Systematic Software Evolution Through Domain Analysis, Proceedings of the IEEE Joint International Conference on Requirements Engineering (RE02) (2002) 18. Arlitt, M., Diwakar, K., Rolia, J.: Characterising the scalability of a large webbased shopping system", ACM Transactions on Internet Technology, 1(1) (2001) 44-69 19. Arlitt, M., Jin, T.: Workload characterisation of the 1988 World Cup Web site, IEEE Network, 14(3) (2000) 30-37 20. Challenger, J., Iyengar, A., Witting, K., Ferstat, C., Reed, P.: A publishing system for efficiently creating dynamic web content, Proceedings of IEEE INFOCOM 2000, (2000) 21. Fielding, R.T., Taylor, R.N.: Principled design of the modern Web architecture, In Proceedings of the 2000 International Conference on Software Engineering (ICSE 2000), Limerick, Ireland, June 2000, (2000) 407-416 22. Hansen, S.: Web information systems- The Changing landscape of management models and web applications", in Proceedings of Conference SEKE 02, July 15-19 (2002) 23. Teo, H.H., Oh, LB., Liu, C., Wei, K.K.: An empirical study of the effects of interactivity on web user attitude, International Journal of Human-Computer Studies, 58(3) (2003) 281-305 24. Morris, M.G., Turner, J.M.: Assessing users' subjective quality of experience with the world wide web: an exploratory examination of temporal changes in technology acceptance, International Journal of Human-Computer Studies, 54(6) (2001) 877-901 25. Chakrabarti, A., Manimaran, G.: Internet Infrastructure Security: A Taxonomy, IEEE Network, November/December (2002) 13-21 26. Ardagna, D. and Francalanci, C.: A cost-oriented methodology for the design of web based IT architectures, Proceedings of the 2002 ACM symposium on Applied computing, (2002) 1127 – 1133