Electronic Business: Concepts, Methodologies, Tools, and Applications (4-Volumes) P19 pptx

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¿UPFROODER-
UDWLRQWKURXJKLQWHU¿UPQHWZRUNVInformation
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
DQGWKHVPDOO¿UPMarketing 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

WRZRUNWRJHWKHUIRUPXWXDOEHQH¿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
IURP³EULFNDQGPRUWDU´EDVHGRSHUDWLRQWRZDUGV
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
DOVREHQH¿WWKURXJKLQFUHDVHGEXVLQHVV
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-
HUVDGGFOLHQWVSHFL¿HGIXQFWLRQDOLW\WRVHUYHUV
and introduce concurrency. These features help
lower computing costs of modern businesses as
ZHOODVEHWWHUPDQDJHQHWZRUNWUDI¿FDVLOOXV-
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.
$VVXFKLQIRUPDWLRQVHDUFKDQG¿OWHULQJDSSOLFD-
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,
YLVLWYDULRXVVHOOHUV¿QGWKHEHVWGHDOEDVHGRQ
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-
WLRQVFRQ¿JXUDWLRQVDYDLODELOLW\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
³FRQVXPHUJXLGH´RURWKHUDGYLVRUEHIRUH
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
EDVHDQGEHQH¿WVDOOSDUWLHVLQYROYHG)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´WKURXJKWKHVHUYHUVWRUHWULHYH
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
LQYHQWRULHVDQGWLHXSSUHFLRXVFDVKÀRZ
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
VKRSNHHSHUWR¿QGWKHDYDLODELOLW\RIFHUWDLQ
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
ORDQDSSURYLQJRI¿FHUVWRPDNHLQIRUPHG
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
KLJKHUOHYHOLQWHUPVRIDSSOLFDWLRQGH¿QHGDJHQW
data. The agent code can then direct the control
ÀRZDSSURSULDWHO\ZKHQWKHVWDWHLVUHVWRUHGDW
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
EHUHDGDEOHRUPRGL¿DEOHE\RWKHUV
$VHFXULW\EUHDFKFRXOGUHVXOWLQWKHPRGL¿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
UHTXLUHVDNH\FHUWL¿FDWLRQLQIUDVWUXFWXUH3XEOLF
NH\VFHUWL¿HGE\WUXVWHGDJHQFLHVFDQEHSRVWHG
in network-wide directories that can be accessed
by agents and servers.
Authorization and Access Control
$XWKRUL]DWLRQLVWKHJUDQWLQJRIVSHFL¿FUHVRXUFH
DFFHVVULJKWVWRVSHFL¿FSULQFLSOHVVXFKDVRZQ-
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]DWLRQRUWKHLUVHFXULW\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¿QHGWRDQDO\VH$JOHWV¶DQG+3:HE6HUYLFHV¶
FDSDELOLWLHV WR IXO¿OO WKHP 7KH HFRPPHUFH
requirements range from simple information
H[FKDQJHDQGEXONGDWDWUDQVIHUWRVHFXUH¿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
FKDQJHVDQGPD\FRQ¿JXUHRQWKHÀ\
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
³PHVVDJHW\SH´LQIRUPRIDVWULQJDQGD³PHV-
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 GHPDQGVRI
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¿WVFORVHO\LQWRGLVWULEXWHGFRPSXWLQJDQGLV
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¿FRQWRORJLHV+HZOHWW3DFNDUGHDUH
used. Ontology refers to the common vocabulary
DQGDJUHHGVHPDQWLFVVSHFL¿FIRUDVXEMHFWGRPDLQ
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'HDFKVHFWRUFDQFUHDWHLWVRZQ
VHPDQWLFWKDW¿WVLQGLYLGXDOQHHGVDQG\HWUHPDLQV
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
LQWKHIRUPRIKDQG\PRGXOHVWREHFRQ¿JXUHG
WRJHWKHUWR¿WLQGLYLGXDOQHHGV)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-
WHUSULVHVVHSDUDWHGE\¿UHZDOOVIntra-enterprise
process management differs from inter-enterprise
SURFHVV PDQDJHPHQW VLJQL¿FDQWO\ 'LIIHUHQW
HQWHUSULVHVDUHQRWRQO\VHSDUDWHGE\¿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
RQDFHQWUDOL]HGZRUNÀRZVHUYHU5DWKHUWKH\
need support for peer-to-peer interactions. This
has become the major impendence for using the
FRQYHQWLRQDOFHQWUDOL]HGZRUNÀRZV\VWHPVIRU
inter-enterprise e-commerce automation.
2QHGLI¿FXOW\IRUWKH$JOHWWHFKQRORJ\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,
LWDOVRKDVDQLQWHJUDWHGVXSSRUWIRU¿QHJUDLQHG
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-
FHVVWRVHUYLFHVLQVLGHKLVQHWZRUNDQG¿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.