114 A Knowledge Management Approach to Improving E-Business Collaboration increasingly automated), organisations need to seek ways to harness and exploit the knowledge of their more precious resource, that is, employees. The human resource strategy needs to provide opportunities for staff to develop ‘know-why’ (Evans, 2003) by facilitating communication between teams by creating the time, space, and reward systems to develop a ‘care-why’ culture. CONCLUSION Collaborative systems are a vital component of modern business. Technology provides a means of communication, facilitating the transmission of information and the integration of processes between organisations. However, the value from e-relationships is dependent upon the ability to share, create, and implement knowledge em- bedded within the information communicated. Knowledge is contextual and situated within communities of practice. Collaborative relation- ships require a new community to be established which bridges the cultures of the workers in the partnership organisations. Collaboration requires trust; in the absence of face-to-face interaction trust emerges from recurrent actions. If e-business systems are developed on espoused requirements rather than practice, problems are likely to occur which may generate mistrust and limit the col- laboration. An approach to improving the ability of collaborative information systems to support authentic work practice is proposed. This places the recurring activities that comprise professional practice and work culture at the centre of an en- quiry about distributed knowledge in e-business collaboration and provides a multilevel approach to begin an exploration into socially embedded knowledge-based collaborative work processes. REFERENCES Barratt, M. (2004). Understanding the mean- ing of collaboration in the supply chain. Supply Chain Management: An International Journal, 9(1), 30-42. Blackler, F. (1995). Knowledge, knowledge work and organizations: An overview and interpreta- tion. Organization Studies, 16(6), 1021-1046. Cox, S. A., Krasniewicz, J. A., Perkins, J. S., & Cox, J. A. (2006, September). Modeling the organizational transformation associated with implementing e-business collaborative systems in the supply chain. In Proceedings of the British Academy of Management Conference (BAM2006) in Association with the University of Ulster and Queen’s University, Belfast. Davenport, T. H., DeLong, D. W., & Beers, M. D. (1998). Successful knowledge management proj- ects. Sloan Management Review, 39(2), 43-57. DeLong, D. W., & Fahey, L. (2000). Diagnosing cultural barriers to knowledge management. Academy of Management Executive, 14(4), 113- 27. Dingley, S., Shah, H., & Golder, P. (2000). Tribes of users and systems developers. Australian Journal of Information Systems, 7(2), 20-31. (QJHVWUĘP<Learning by expanding: An activity-theoretical approach to developmental research. Helsinki: Oreinta-Konsulti. (QJHVWUĘP < ([SDQVLYH OHDUQLQJ DW work: Toward an activity theoretical reconceptu- alisation. Journal of Education and Work, 14(1), 133-156. Evans, C. (2003). Structures, roles and responsi- bilities in a knowledge-centric culture. In C. Evans (Ed.), Managing for knowledge: HR’s strategic role. Oxford: Butterworth-Heinemann. Fahey, F., Srivastava, R., Sharon, J. S., & Smith, D. E. (2001). Linking e-business and operating processes: The role of knowledge management. IBM Systems Journal, 40(4), 889-907. Graham, G., & Hardaker, G. (2000). Supply-chain management across the Internet. International 115 A Knowledge Management Approach to Improving E-Business Collaboration Journal of Physical Distribution and Logistics, 30(34), 286-296. Grover, V., & Davenport, T. H. (2001). General perspective on knowledge management: Foster- ing a research agenda. Journal of Management Information Systems, 18(1), 5-21. Handy, C. (1995, May). Trust and the virtual or- ganization. Harvard Business Review, 40-49. Hofstede, G. (1991). Cultures and organizations: Software of the mind. Intercultural cooperation and its importance for survival. Maidenhead: McGraw-Hill. John-Steiner, V., Weber. R. J., & Minnis, M. (1998). The challenge of studying collaboration. American Educational Research Journal, 35(4), 773-783. Kirchmer, M. (2004). E-Business process net- works: Successful value chains through standards. Journal of Enterprise Information Management, 17(1), 20-30. Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge University Press. Lee, L. L. (2005). Balancing business process with business practice for organizational advan- tage. Journal of Knowledge Management, 9(1), 29-41. /L):LOOLDPV+,QWHU¿UPFROODER- UDWLRQWKURXJKLQWHU¿UPQHWZRUNVInformation Systems Journal, 9, 103-115. Martin, C. (1999). Net future. New York: Mc- Graw-Hill. Mohamed, M., Stankosky, M., & Murray, A. (2006). Knowledge management and information technology: Can they work in perfect harmony? Journal of Knowledge Management, 10(3), 103- 116. Nahapiet, J., Gratton, L., & Rocha, H. O. (2005). Knowledge and relationships: When cooperation is the norm. European Management Review, 2, 3-14. Nonaka, I., & Takeuchi, H. (1995). The knowledge- creating company: How Japanese companies create the dynamics of innovation. New York: Oxford University Press. Oliver, S., & Kandadi, K. R. (2006). How to develop knowledge culture in organizations? A multiple case study of large distributed organi- zations. Journal of Knowledge Management, 10(4), 6-24. O’Toole, T. (2003). E-relationships: Emergence DQGWKHVPDOO¿UPMarketing Intelligence and Planning, 21(2), 115-122. Pearlson, K. E., & Saunders, C. S. (2004). Manag- ing and using information systems: A strategic approach. Chichester: John Wiley. Sahay, B. S. (2003). Supply chain collaboration: The key to value creation. Work Study, 52(2), 76-83. Snowdon, D. (2003). Knowing what we know and tools for knowledge mapping. In C. Evans (Ed.), Managing for knowledge: HR’s strategic role. Oxford: Butterworth-Heinemann. Tan, K. C. (2001). A framework for supply chain management literature. European Journal of Pur- chasing and Supply Management, 7(1), 39-48. :DUG - *ULI¿WKV 3 :KLWPRUH 3 Strategic planning for information systems (2 nd ed.). Chichester: John Wiley. Wenger, H. (1998). Communities of practice. Cambridge University Press. KEY TERMS Collaborative Practice: The ways of working which emerge between communities of practice as they work towards the achievement of com- mon goals. 116 A Knowledge Management Approach to Improving E-Business Collaboration Collaborative Systems: A computer-based system that is accessed and used by more than one organisation to support business transactions in the supply chain. The system allows data to be automatically updated in a partner organisation’s systems during the processing of a transaction. Community of Practice: The emergent pro- cess of social learning as a group of people with shared values, beliefs, and goals, work together towards a common aim. E-Business Processes: The explicit proce- dures and routine decisions supported by data and information communicated electronically ZLWKLQ LQWHU¿UP QHWZRUNV WR VXSSRUW EXVLQHVV transactions and value added activities. E-Collaboration: The use of information technology to establish, facilitate, and sustain co- operation between two geographically dispersed parties, who have common goals, to enable them WRZRUNWRJHWKHUIRUPXWXDOEHQH¿W E-Community: The development of shared purpose, values, and experience resulting in the formation of trust between a group of people who may be geographically dispersed and communi- cate mainly via electronic means. Knowledge Workers: A person who works with data, information, and knowledge rather than physical components. Practice-Centric: The focus on the way in which knowledge workers carry out tasks based on their learning and experience of a situation, addressing anomalies or omissions that may exist in formalised business processes. Situated Knowledge: Knowledge that HPHUJHV IURP H[SHULHQFH UHODWLQJ WR VSHFL¿F business practices. Social Activity Theory: The study of authentic practice arising from communities of knowledge workers engaged in business practice. This work was previously published in the Encyclopedia of Human Resources Information Systems: Challenges in e-HRM, edited by T. Torres-Coronas and M. Arias-Oliva, pp. 598-604, copyright 2009 by Information Science Reference (an imprint of IGI Global). 117 Copyright © 2009, IGI Global, distributing in print or electronic forms without written permission of IGI Global is prohibited. Chapter 1.9 Linking Businesses for Competitive Advantage: A Mobile Agent-Based Approach Tong-Seng Quah Nanyang Technological University, Republic of Singapore Chye-Huang Leow Singapore Polytechnic, Republic of Singapore ABSTRACT In a highly competitive business environment, every organization is trying to achieve more using fewer resources. This is particularly true in this Internet era, where many businesses are moving IURP³EULFNDQGPRUWDU´EDVHGRSHUDWLRQWRZDUGV having at least an Internet presence, where e-com- merce is fast gaining acceptance. Recent advances in mobile agent technology promise yet another powerful tool to gain competitive advantage—to deliver cost-effective services through utilizing Internet infrastructure. Such a development helps both individuals and organizations achieve higher productivity at lower cost. In this chapter, the authors describe an intelligent mobile agent-based system that links hotels and restaurants to provide gourmet goers with a convenient way of searching for their choice restaurants. The system sends off intelligent mobile agents to automatically roam the Internet, gather the relevant information about food and services from participating restaurants, and provide the most optimized selection as sug- gestions to help the users make their meals deci- sion. This greatly reduces information overload for the users. Participating business establishments DOVREHQH¿WWKURXJKLQFUHDVHGEXVLQHVV INTRODUCTION Agent-based systems have gained prominence over the last few years. One of the most interest- ing categories of agents is mobile agents (Lange & Oshima, 1998). Unlike static agents, which are restricted to operate within a single machine or address space, mobile agents have the ability to migrate over the network, execute tasks at each location and potentially interact with other agents that cross their paths. Advantages of mobile agents 118 Linking Businesses for Competitive Advantage include their ability to reduce network usage, increase asynchrony between clients and serv- HUVDGGFOLHQWVSHFL¿HGIXQFWLRQDOLW\WRVHUYHUV and introduce concurrency. These features help lower computing costs of modern businesses as ZHOODVEHWWHUPDQDJHQHWZRUNWUDI¿FDVLOOXV- trated below. Many online business transactions involve processes that require extensive database searches and matches. For example, users of an online bookstore are likely to view various catalogs, matching descriptions with preferences they have in mind before deciding which books to purchase. $VVXFKLQIRUPDWLRQVHDUFKDQG¿OWHULQJDSSOLFD- tions often download and process large amounts of server-resident information and generate comparatively small amounts of result data. The scenario is greatly different with a mobile agent- based system, where mobile agents move to and execute on server machines and access server data without using the network, reducing band- width requirements. Many of today’s applications involve repeated client-server interactions, which require either maintaining a network connection over an extended period or making several sepa- rate requests. If mobile agents are used instead, the client system does not have to maintain a network connection when its agents access and process information. This permits increased asynchrony between the client and server. This feature is especially useful for mobile computers (such as laptops and PDAs), which typically have low-bandwidth, unreliable connections to the network and are often switched off to save power consumption. Also, the repeated client-server interactions are reduced to two agent-transfer operations, reducing the frequency of network usage, as well. An example of a user-level application would be an electronic marketplace. Vendors can set up online shops, with products, services or informa- tion for sale. A customer’s agent would carry a shopping list along with a set of preferences, YLVLWYDULRXVVHOOHUV¿QGWKHEHVWGHDOEDVHGRQ user preferences and purchase the product using digital forms of cash. An added advantage of such a system is that businesses may also be linked up to form a chain, such that mobile agents may move between stores within a business chain to make their purchases. Such a setup will enhance the competitive advantages of participating online stores. Apart from mobility, a mobile agent-based sys- tem will need mechanisms for restricted resource access, secure electronic commerce, protection of agent data, robustness and user control over roaming agents. These will be discussed in later sections. MOBILE AGENTS: ISSUES AND DEVELOPMENTS Agents-Enabled Electronic Commerce Mobile agents offer a number of useful possibili- ties: • The agent can express the application-level protocol required to perform a transaction. This includes dialogs on choices and op- WLRQVFRQ¿JXUDWLRQVDYDLODELOLW\GHOLYHU\ methods and opportunities for setting up, as well as complete and accurate capture of information required by the vendor in a particular format. Mobile agents technology is a plausible method for vendors to distribute the client end of a transaction protocol in a device-independent way. • Alternatively, the mobile agent may be able to present the consumer’s desire as a query to a number of potential vendors to determine degree of match, price, availability and so forth. • The agent may also be able to consult a ³FRQVXPHUJXLGH´RURWKHUDGYLVRUEHIRUH making a purchase. 119 Linking Businesses for Competitive Advantage • The agent can provide a secure vehicle for the transaction, providing bilateral authen- tication and privacy. • The agent can provide a transaction cur - rency for settlement. The agent’s account is presumably reconciled periodically against ³UHDO´PRQH\ To facilitate the development of mobile agents distributed applications, and to overcome some problems and issues that arise from this approach, some requirements must be addressed. Systems that support the use of the mobile agent paradigm have to provide a basic set of services and char- acteristics. These will be discussed later. Achievable Competitive Advantages Using Mobile Agent-Based E-Commerce Platforms While many potential competitive advantages can be achieved using a mobile agent-based e-commerce system, the following are being highlighted: 1. Efficient supply chain management: Enterprises may link up to provide a wider range of products and services to customers. This will likely attract a larger customer EDVHDQGEHQH¿WVDOOSDUWLHVLQYROYHG)RU example, by linking up a hotel server with restaurant chain servers, guests in the hotel get the impression of a wider range of cutlery service available. The restaurants, on the other hand, make their presence noticed and are likely to get more business. All these can be achieved by the mobile agent applications ³ZHDYLQJ´WKURXJKWKHVHUYHUVWRUHWULHYH and recommend cutlery establishments that match the hotel guests’ preferences. 2. Effective inventory control: Many busi- nesses have stores and shop-fronts at multiple locations. To minimize overstocking of LQYHQWRULHVDQGWLHXSSUHFLRXVFDVKÀRZ many businesses keep their inventory low. However, this risks loss of sales when a customer wants goods that are out of stock at a certain branch store. The situation can be saved if such businesses link up their store-front computers using a mobile agent- based system. Such a system will enable a VKRSNHHSHUWR¿QGWKHDYDLODELOLW\RIFHUWDLQ stock that matches a customer’s request, thus capturing sales instead of letting a customer walk out of the shop and be disappointed. 3. Powerful information searches for deci- sion making: Accessibility to information are crucial for important decision making such as loan approval—especially if the quantum is big. Using this example, credit providers may join a bureau which captures the ‘worthiness’ of private individuals. An agent-based system may be deployed to consolidate the credit situations of a loan applicant with various banks by utilizing the restricted access rights to the bureau controlled databases. This will help the ORDQDSSURYLQJRI¿FHUVWRPDNHLQIRUPHG decision and hence reduces bad debt for the lenders. Agent Mobility The primary identifying characteristics of mobile agents is their ability to autonomously migrate from host to host. Thus, support for agent mo- bility is a fundamental requirement of the agent infrastructure. An agent can request its host server to transport it to some remote destination. The agent server must then deactivate the agent, capture its state and transmit it to the server at the remote host. The destination server must restore the agent state and reactivate it at the remote host, thus completing the migration. The state of an agent includes all its data, as well as the execution state of its thread. At the lowest level, this is represented by its execution context and call-stack. If this can be captured 120 Linking Businesses for Competitive Advantage and transmitted along with the agent, the destina- tion server can reactivate the thread at precisely the point where it requested the migration. An alternative is to capture the execution state at a KLJKHUOHYHOLQWHUPVRIDSSOLFDWLRQGH¿QHGDJHQW data. The agent code can then direct the control ÀRZDSSURSULDWHO\ZKHQWKHVWDWHLVUHVWRUHGDW the destination. Security Issues The introduction of mobile agent code in a network raises several security issues. In an open network, such as the Internet, it is entirely possible that the agent and server belong to different administra- tion domains. In such cases, they will have much lower levels of mutual trust. Servers are exposed to the risk of system penetration by malicious agents, analogous to viruses and Trojan horses. Security-related requirements are discussed in the following sections. Privacy and Integrity Agents carry their own code and data with them as they traverse the network. Parts of their state may be sensitive and need to be kept secret when the agent travels on the network. For example, a shopper agent may carry its owner’s credit card number or personal preferences. The agent transport protocol needs to provide privacy to prevent eavesdroppers from acquiring sensitive information. Also, an agent may not trust all serv- ers equally. We need a mechanism to selectively reveal different portions of the agent state to different servers. For example, a shopping agent may solicit quotations from various vendors. To ensure fairness, one vendor’s quotation must not EHUHDGDEOHRUPRGL¿DEOHE\RWKHUV $VHFXULW\EUHDFKFRXOGUHVXOWLQWKHPRGL¿FD- tion of the agent’s code as it traverses the network. We need some means of verifying that an agent’s code is unaltered during transit across a distrusted network or after visiting a distrusted server. An agent’s state typically needs to be updated dur- ing its journey so it can collect information from servers. While we cannot assume that all servers visited are benign, we can provide mechanisms that allow such tampering to be detected. Cryptographic mechanisms can be used to provide a secure communication facility, which an agent can use to communicate with its home site, or servers can use to transport agents safely across distrusted networks. Selective revealing of state can be accomplished by encrypting dif- ferent parts of the state with different public keys belonging to the servers allowed to access those parts of the state. Mechanisms such as seals can be used to detect any tampering of agent code. Authentication When an agent attempts to transport itself to a remote server, the server needs to ascertain the identity of the agent’s owner to decide what rights and privileges the agent will be given in the server’s environment. A vendor’s server needs to know the visiting agent’s identity to determine which user to charge for service rendered. Con- versely, when an agent migrates to a server, it needs some assurance of the identity of the server itself before it reveals any of its sensitive data to the server. Digital signature systems have been used to develop mutual authentication schemes. To verify signatures, agents and servers need to r el ia bl y k now t he sig n i n g e nt i t y’s pu bl i c ke y. T h i s UHTXLUHVDNH\FHUWL¿FDWLRQLQIUDVWUXFWXUH3XEOLF NH\VFHUWL¿HGE\WUXVWHGDJHQFLHVFDQEHSRVWHG in network-wide directories that can be accessed by agents and servers. Authorization and Access Control $XWKRUL]DWLRQLVWKHJUDQWLQJRIVSHFL¿FUHVRXUFH DFFHVVULJKWVWRVSHFL¿FSULQFLSOHVVXFKDVRZQ- ers of agents). Some principals are more trusted than others, and thus, their agents can be granted less-restrictive access. This involves specifying 121 Linking Businesses for Competitive Advantage policies for granting access to resources based either on identities of principals, their roles in an RUJDQL]DWLRQRUWKHLUVHFXULW\FODVVL¿FDWLRQ Metering and Charging Mechanisms When agents travel on a network, they consume resources, such as CPU time, disk space and so forth at different servers. These servers may legitimately expect to be reimbursed monetarily for providing such resources. Also, agents may access value-added services—information and so forth—provided by other agents, which could also expect payment in return. For example, in a marketplace, users can send agents to conduct purchases on their behalf. Thus, mechanisms are needed so that an agent can carry digital cash and use it to pay for resources used by it. Operating system-level support may be needed for meter- ing of resource usage, such as the CPU time used by an agent or the amount of disk space needed during its visit. Agent Monitoring and Control An agent’s parent application may need to monitor the agent’s status while it executes on a remote host. If exceptions or errors occur during the agent’s execution, the application may need to terminate the agent. This involves tracking the current location of the agent and requesting its host server to kill it. Similarly, the agent owner may simply recall its agent back to its home site and allow it to continue executing there. This is equivalent to forcing the agent to execute a migrate call to its home site. The owner can use an event mechanism to signal the agent or raise an exception remotely. The agent’s event/exception handler can respond by migrating home. This capability of remotely terminating and recalling agents raises security issues—only an agent’s owner should have the authority to terminate it. Thus, some authentication functions need to be built into these primitives; that is, the system must ensure that the entity attempting to control the agent is indeed its owner, or has been authorized by the owner to do so. COMPARATIVE STUDY OF E-COMMERCE REQUIREMENTS, AGLETS, AND HP WEB SERVICES Finally, a set of e-commerce requirements will be GH¿QHGWRDQDO\VH$JOHWV¶DQG+3:HE6HUYLFHV¶ FDSDELOLWLHV WR IXO¿OO WKHP 7KH HFRPPHUFH requirements range from simple information H[FKDQJHDQGEXONGDWDWUDQVIHUWRVHFXUH¿UH- wall traversal, close collaboration and dynamic relationship requirements. It will be shown where each technology has its advantages and domains. This comparison also shows how the combina- tion of both technologies can provide combined advantages and strengths. Information Exchange in E-Commerce, Aglets, and HP Web Services Many of today’s e-commerce applications include complex business processes with a large number of concurrent tasks. These tasks may persist for a long duration; they may require long waiting times and could be nested within other tasks. Additionally, they are highly asynchronous, expose continues FKDQJHVDQGPD\FRQ¿JXUHRQWKHÀ\ 7KXV DQ\ ÀDW FRQYHUVDWLRQ PDQDJHPHQW like message exchange, lacks the scalability for handling and tracking such sizable applications. Unfortunately, message exchange is the way Aglets interact. These messages always follow the same basic scheme. They are composed of a ³PHVVDJHW\SH´LQIRUPRIDVWULQJDQGD³PHV- sage content”, which can be any type of object. However, they do not support the demands of modern e-commerce. Any more complex transactions in Aglets are usually implemented through a centralized sched- 122 Linking Businesses for Competitive Advantage uling architecture, where one Aglet host serves a coordination unit and does the scheduling, moni- toring and execution control. This may work well within one single enterprise, but causes serious problems for inter-enterprise transactions. HP Web Services, on the other hand, evolved from the Distributed Computing paradigm, which is primarily involved in handling such transac- tions. The e-brokering system was added on top of that, and it closely follows the e-commerce model. Business tasks are modeled as services and can be composed through other lower-level nested services. A typical complex HP Web Services request is broken down into simpler requests. The set of service providers for each of these simple requests is then dynamically discovered. Subsequently, the best match is invoked, and its execution mediated. This model used by HP :HE 6HUYLFHV ¿WV H[DFWO\ LQWR WKH GHPDQGVRI e-commerce. Bulk Data Transfer in E-Commerce, Aglets, and HP Web Services As personalized, continuously running and semi-autonomous entities, Aglets can be used to mediate users and servers to automate a number of time-consuming tasks in e-commerce. How- ever, again, Aglets communicate via message exchange, which may not be suitable for bulk data exchange. Routing and caching a large amount of data imposes a considerable burden for Aglets. For example, moving data between an operational database and a data warehouse via an Aglet is very unlikely. HP Web Services can provide asynchronous and synchronous communication in the same en- vironment. Bulk data transfer is an easy task for HP Web Services, as well as for other distributed computing environments, like CORBA and RMI. ,W¿WVFORVHO\LQWRGLVWULEXWHGFRPSXWLQJDQGLV a direct extension from Networking Transport Protocols (like TCP/IP). Extensible Mark-up Language (XML) as Joint Communication Language in Aglets and HP Web Services In today’s technical world, many different domain VSHFL¿FRQWRORJLHV+HZOHWW3DFNDUGHDUH used. Ontology refers to the common vocabulary DQGDJUHHGVHPDQWLFVVSHFL¿FIRUDVXEMHFWGRPDLQ Both HP Web Services and Aglets mainly focus on establishment of collaboration, mediation and providing services. They thereby aim at generic solutions to be applied across many different sectors of businesses. However, a banking institu- tion may use an entirely different ontology than a CD retailer. Currently, XML is in the process of solving this problem. Through the use of Document Type 'H¿QLWLRQ'7'HDFKVHFWRUFDQFUHDWHLWVRZQ VHPDQWLFWKDW¿WVLQGLYLGXDOQHHGVDQG\HWUHPDLQV generally usable across sector boundaries. The power of XML and its role in e-commerce have been widely recognized. Consequently, HP Web Services provides support for XML in its Ap- plication Programming Interface (API). The software developed during this project enables communication between Aglets and HP Web Services. The software can receive and send Aglet messages as well as deploy HP Web Services. And it exports all these functionalities LQWKHIRUPRIKDQG\PRGXOHVWREHFRQ¿JXUHG WRJHWKHUWR¿WLQGLYLGXDOQHHGV)XUWKHUPRUH reuse was one of the major design considerations for this project. The software could be easily extended with additional modules to implement a proxy between the Aglet world, HP Web Ser- vices and the Internet. A DTD-based interpreter VKRXOG FORVHO\ IXO¿OO WKHVH UHTXLUHPHQWV 7KLV would enable document-driven Aglet cooperation. Moreover, it would allow Aglets to share ontology (Hewlett Packard, 2000e) for multiple or even dynamic domains. In this way, the cooperation of dynamic Aglets would support plug-and-play commerce—mediating businesses that are built on 123 Linking Businesses for Competitive Advantage one another’s service. Aglets would acquire some of the key functionalities of HP Web Services. Firewalls in Aglets and HP Web Services Internet-based e-commerce involves multiple en- WHUSULVHVVHSDUDWHGE\¿UHZDOOVIntra-enterprise process management differs from inter-enterprise SURFHVV PDQDJHPHQW VLJQL¿FDQWO\ 'LIIHUHQW HQWHUSULVHVDUHQRWRQO\VHSDUDWHGE\¿UHZDOOV but also have self-interests and individual data sharing scopes. When they are involved in a busi- ness process, they are unlikely to trust and rely RQDFHQWUDOL]HGZRUNÀRZVHUYHU5DWKHUWKH\ need support for peer-to-peer interactions. This has become the major impendence for using the FRQYHQWLRQDOFHQWUDOL]HGZRUNÀRZV\VWHPVIRU inter-enterprise e-commerce automation. 2QHGLI¿FXOW\IRUWKH$JOHWWHFKQRORJ\WR¿W into this picture consists in the limitation of its coordination model. HP Web Services, on the other hand, has Firewall Traversal as one of its standard services. Since HP Web Services has its roots in distributed operating systems research, LWDOVRKDVDQLQWHJUDWHGVXSSRUWIRU¿QHJUDLQHG access control. The HP Web Services Engine can be inserted at multiple points in the chain between clients and remote services. These remote services will act and look just like a local service, since the HP Web Services Engine acts like a kernel. Thus, the administrator can see and control ac- FHVVWRVHUYLFHVLQVLGHKLVQHWZRUNDQG¿UHZDOO traversal is supported. Collaboration in E-Commerce, Aglets and HP Web Services An e-commerce scenario typically involves the following activities: identifying requirements, brokering products, brokering vendors, negoti- ating deals, or making purchase and payment transactions. Today, these activities are initiated and executed by humans. Using Aglets or, in general, Mobile Agents technology, to support e-commerce automation is a promising direction. Aglets could be personal- ized, continuously running and semi-autonomous, driven by a set of beliefs, desires and intentions (BDI). They could be used to mediate users and servers to automate a number of the most time- consuming tasks in e-commerce with enhanced parallelism. HP Web Services was primarily designed for enabling the creation of dynamic, Internet-based business relationships through the ad hoc discov- ery and interaction of e-services. E-services in- clude applications, computing resources, business processes and information, delivered securely over the Internet. The HP Web Services Framework 6SHFL¿FDWLRQ 6)6 GH¿QHV VWDQGDUG EXVLQHVV interactions and conventions as XML documents that allow e-services to dynamically discover and negotiate with each other and compose themselves into more complex services. Dynamics in E-Commerce, Aglets and HP Web Services E-commerce applications operate in a distributed computing environment, involving multiple par- ties with dynamic availability and a large number of heterogeneous information sources with evolv- ing contents. Dynamic relationships among a large number of autonomous service requesters, brokers and providers is common. A business partner- ship (e.g., between suppliers, resellers, brokers and customers) is often created dynamically and maintained only for the required duration, such as a single transaction. E-commerce activities typically rely on distributed and autonomous tasks for dealing with such operational dynamics. Thus, e-commerce is a plug-and-play environment. Ser- vices need to be provided on demand. To support such dynamics, an e-commerce infrastructure must support the cooperation of loosely coupled e-business systems. . values, beliefs, and goals, work together towards a common aim. E-Business Processes: The explicit proce- dures and routine decisions supported by data and information communicated electronically. where mobile agents move to and execute on server machines and access server data without using the network, reducing band- width requirements. Many of today’s applications involve repeated. other hand, make their presence noticed and are likely to get more business. All these can be achieved by the mobile agent applications ³ZHDYLQJ´WKURXJKWKHVHUYHUVWRUHWULHYH and recommend