FRAMEWORK FOR ANALYZING THE IMPLICATIONS OF
ARCHITECTURAL SOLUTIONS TO MARKET STRUCTURES
Nonna Kostamo, Otso Kassinen, Timo Koskela and Mika Ylianttila
MediaTeam Oulu, University of Oulu, P.O. Box 4500, 90014 University of Oulu, Finland
Keywords: Market structure, mash-up services, peer-to-peer, context distribution, value engineering.
Abstract: Understanding the economical effects of a technology has become more and more important. This has
created a need for a conceptual framework to analyze the economic implications of technologies already in
their design stage. So far, most of the research has been done in retrospect. Our framework is an organized
approach to analyze this complex multidisciplinary task. The framework is experimented by analysing a
case of a personalized mobile context-aware community web calendar. The market structure analysis
concentrates on three solutions designed for the collection and distribution of user context and community
information, an essential functionality for the service being developed.
1 INTRODUCTION
Understanding the economical effects of techno-
logies is more and more important. Research done in
this field has shown in retrospect that many chosen
technologies have not been the most efficient or
profitable ones. The problem is the lack of
understanding the economic effects of a technology
during its design phase. Also market competition has
led into the use of current or outdated technology.
Understanding the economic effects already at
the design stage is a complex task. The research in
this area concentrates on the market effects of
technology entries (Choi 1998), (Vega-Redondo
1996). Our approach tries to see the economic
effects of the technological solutions under research
and design. We experiment it in the case of a mobile
context-aware web community calendar, and present
and evaluate three solutions for the case.
2 CONCEPTS
Here we present the basic concepts of this paper.
2.1 Economics of Imperfect
Competition
Economics of imperfect competition, better known
as industrial organization, is a field of economics
that studies companies’ strategic behavior, market
structure, and their interactions. It is also referred to
as industrial economics. This paper deals with
market structure and different strategic possibilities.
2.2 Enablers
In 2004, O’Reilly Media introduced the concept of
Web 2.0; a new way of architecting software and
businesses. It treats the web as a platform (O’Reilly
2005). It also encourages the users to contribute as
service co-developers. Data from various sources is
combined into mash-up services. An example mash-
up service adds enriching information to an online
map service, e.g. Google Maps.
Context information is any information that can
be used to characterize an entity’s situation (Dey
2000). Context-awareness is powerful for automatic
service configuration in limited mobile environment.
3 FRAMEWORK STUDY
Table 1 presents the steps of our analysis
framework. In the first step, information gathering,
one defines the service requirements. In the second
step, benchmarking, a market survey is carried out to
build up a list of existing technologies and solutions
to validate the service idea. In criteria for
evaluation, one defines the criteria used to evaluate
alternative solutions.
196
Kostamo N., Kassinen O., Koskela T. and Ylianttila M. (2007).
FRAMEWORK FOR ANALYZING THE IMPLICATIONS OF ARCHITECTURAL SOLUTIONS TO MARKET STRUCTURES.
In Proceedings of the Second International Conference on e-Business, pages 196-199
DOI: 10.5220/0002112101960199
Copyright
c
 SciTePress

Table 1: Description of steps in analysis framework.
Step Task
Information gathering Defining service requirements.
Benchmarking
Defining current markets
and used technologies.
Criteria for evaluation
Defining the evaluation criteria
and how they are measured.
Analyzing
The initial technical solution
and market structure analysis.
Creation
Defining alternative ways to
meet the defined requirements.
Evaluation
Selecting the best alternative
based on the selected criteria.
Analyzing includes the analysis of the initial
technical solution and market structure. In creation
alternative solutions corresponding to the
requirements are created. Finally, in evaluation, the
alternative solutions are compared, and the best
solution is selected.
3.1 Information Gathering and
Benchmarking
Applying our framework to a problem case, we have
to first define the service requirements (information
gathering). The analysis of the current markets is not
presented in this paper.
3.1.1 Context-Aware Mobile Community
Calendar
As an example mash-up service scenario, we
propose a context-aware community calendar. The
existing commercial services, e.g., Google Calendar,
work only if the users are members of the same
community. Our calendar mash-up would share
users’ information while: 1) being independent of
calendar type; 2) being independent of terminal type;
3) having private and public information; 4) having
context-awareness if available; 5) not having all
users from the same calendar-service provider. Next,
we present the service architecture for the calendar.
3.1.2 Elements of the Service Architecture
User context, community and authentication related
information can be collected, refined, and provided
to web services by a mobile middleware component.
The middleware may even apply group information
from existing P2P communities to web services.
To ensure security, storage, and delivery of user
information, it should be controlled by a trusted
entity, which provides this information to authorized
web services through a control service. Using the
control service as a lightweight portal, the user may
modify her community information and access the
mash-ups (service discovery). Mash-ups consume
user context and community information. Figure 1
shows the architecture.
Mash-up
Service
Mash-up
Service
Mash-up
Service
Existing
Services
Providing User Context and
Community Information
Secure storage
and Delivery of
User Information
Authenticaton
(and Billing)
Subscribing to User Context
and Community Information
Service Discovery
Control Service
Middleware component
End-User Terminal
Middleware component
End-User Terminal
Middleware component
End-User Terminal
Service
Personalization
Figure 1: Elements of the service architecture.
3.2 Criteria for Evaluation
After the initial requirement analysis and
benchmarking, criteria for the evaluation of the
service are created and divided into two categories:
(1) mobile device specific and (2) service
architecture specific. The first one includes:
terminal’s processing power, terminal bandwidth,
and metadata needs. The second one: control service
load, centralized management need, system
scalability, search efficiency (coverage/availability),
information security, and standardization need.
3.3 Analyzing and Creation
In these phases different technical solutions and their
market structure analysis are presented based on the
previously defined requirements. We present three
different solutions, centralized, hybrid and peer-to-
peer (P2P), for implementing the delivery of user
context and community information.
3.3.1 Technical Solutions
The first approach, management of data is fully
centralized as seen in Figure 2 a). The mobile
middleware would update the information to the
control service (CS) at appropriate intervals and the
mash-ups (MuS) could then access the user (peer, P)
information. SP stands for superpeer.
FRAMEWORK FOR ANALYZING THE IMPLICATIONS OF ARCHITECTURAL SOLUTIONS TO MARKET
STRUCTURES
197

Figure 2: Technical solutions for context and community
information delivery a) centralized, b) hybrid and c) P2P.
The hybrid solution utilizes the resources of
decentralized P2P networks. The middleware could
publish user information in them. Mash-ups query
the network through the control service(s) that act as
gateways to user information and bind together P2P
networks of various providers; see Figure 2 b).
The third approach, seen in Figure 2 c), is
organized completely in a P2P manner, the web
servers also being part of the P2P networks. An
instant messenger (IM) type client could operate as a
service portal, launching the service directly to the
browser, since service discovery and distributed
rights control capabilities can be added to the client.
The browser cannot locally manage the rights of
user information, because they are scattered around
the network. Service discovery queries could be
made in P2P networks using the IM client, or service
links could be passed on among users.
3.3.2 Market Analysis
As we can see from Figure 3, centralized
architecture leads to an almost monopolistic market
structure. However, an oligopoly is more likely; the
service bundles offered by each retailer-lever
company could be differentiated.
Figure 3: Market structures of proposed solutions.
Hybrid solution leads to monopolistic
competition: there are many producers and many
consumers in a given market. In addition, the
consumers have clearly defined preferences, and the
sellers attempt to differentiate their products from
those of their competitors; the goods and services
are still heterogeneous since the user group of each
retailer differs from each other. The entry barriers of
new retailers lie in attracting the users to your
control service. Depending on the market share, the
retailers do have some control over prices.
From Figure 3, we can see that P2P solution leads
us closer to a perfect competition structure. We can
see that the control service retailer-level is marginal
as an authenticator. Information flow between
Aggregator/Retailers and consumers is unrestricted
as is necessary for the architecture to function well.
All services from manufacturers could be available
to everyone. This could lead to homogeneity or
personalization of the service bundles.
All this said, we must remember that our study of
these market structures is quite simplified. We do
understand that in reality the perfect competition is a
hypothetical market structure. To analyze the
technology’s effects to market structures better in
the design phase, we would need more precise
information on the market itself and, for example,
competitor strategies.
3.4 Evaluation
In this final phase, we compare and evaluate the
three distinct mobile community calendar solutions.
Their technical features are compared in Table 2.
Table 2: Comparison of differences in technical features of
proposed technological solutions.
Central’d Hybrid P2P
Control service load High Medium Low
System scalability Low Medium High
Metadata need Low High High
Terminal processing
power need
Low Low Medium
Terminal bandwidth
need
Medium Medium High
Information security High Medium Medium
Search efficiency
coverage/availability High Medium Medium
Centralized
management need High Low Very low
Standardization need Low Medium High
The control service load depends on the amount
of context and community information being
transferred through the control service; the load
decreases when moving from centralized to P2P
solution. It is quite obvious that system scalability is
inversely proportional to the control service load.
The metadata need affects the terminal bandwidth
need. The P2P solution needs more bandwidth
ICE-B 2007 - International Conference on e-Business
198

because the metadata includes the usage rights of the
delivered context information. Furthermore, the
processing power and bandwidth needs correlate
with power consumption. Thus, the P2P model is the
most demanding for the mobile device.
Information security and search efficiency are
best handled by a trusted central actor. In context
dependent services, search efficiency and data
availability are critical, due to the dynamic nature of
the data (e.g. location, presence). In P2P and hybrid
solutions, efficient querying is achieved by dividing
the load considering peers’ capabilities. The peers
with great processing power and bandwidth are set
as superpeers, handling the message routing and the
storage of user information (Gehlen 2005). Skype
has proved a hierarchical P2P approach feasible for
messaging and presence. Presence, which could also
cover other context types, is delivered efficiently
and securely in P2P manner; only the login server is
centralized. (Baset and Schulzrinne 2004)
The centralized management need is the greatest
in the centralized solution. The hybrid and P2P
solutions allow the user to choose more freely his
service and control service provider, and there is no
urgent need for the operator even though it might
maintain the P2P network and authentication
services. However, the more different controllers,
the greater is the standardization need, since
interoperability of the players must be guaranteed.
Table 3: Differences in market structures.
Central’d Hybrid P2P
Nr. of producers Low Medium High
Network size n/m n/m n
Type of services Heterog. Heterog. Homog.
Service freedom Low Medium High
Table 3 presents the main differences in the
market structures. In centralized solution, the
number of producers might be more than “low”, but
as in mobile operator markets, it is likely that the
majority of the market share would be divided
between 3 or 4 companies. The network size of a
company is all the consumers using this kind of
service (n) divided by the market share (m) of the
company in question. In P2P, n is the number of
consumers. The type of services is heterogeneous in
the centralized and hybrid solutions and
homogeneous in P2P, because of service freedom
i.e. service selection available for consumer use.
Considering the three propositions, P2P might
seem a good choice but its downsides are need for
greater amount of metadata and processing power in
mobile devices. Hybrid model, even though little
restricted on service offerings, might be a good
solution for mobile operator market implementation.
4 DISCUSSION
We have now presented a framework for analyzing
the economic implications of technological
solutions. We tested this framework for evaluating
the information-distribution technologies for a
community calendar service. Through utilizing this
framework, we discovered three possible solutions
for the delivery of user context and community
information in this service. Finally, we analyzed the
possible market structures of each solution and
evaluated the differences between these alternatives.
More research is needed to understand the
different economic implications of technology
during its design phase. Also a study of co-existing
technologies and solutions should be conducted.
REFERENCES
O’Reilly Media Inc. http://www.oreillynet.com/pub/a/
oreilly/tim/news/2005/09/30/what-is-web-20.html.
Dey, A., 2000, Providing Architectural Support for
Building Context-Aware Applications. PhD thesis.
College of Computing. Georgia Institute of
Technology, Atlanta, GA, USA, 170 p.
Gehlen, G., Pham, L., 2005. Mobile Web services for
peer-to-peer applications. In Proc. of the Consumer
Communications and Networking Conference, pp.
427-433.
Baset, S., Schulzrinne, H., 2004. An Analysis of the Skype
Peer-to-Peer Internet Telephony Protocol. Technical
Report CUCS-039-04. Department of Computer
Science, Columbia University, New York, NY, USA.
Choi, J., Thum, M., 1998. Market Structure and the
Timing of Technology Adoption with Network
Externalities. European Economic Review, Vol. 42,
pp. 225-244.
Vega-Redondo, F., 1996. Technological change and
market structure: An evolutionary approach.
International Journal of Industrial Organization, Vol.
14, Issue 2, pp. 203-226.
FRAMEWORK FOR ANALYZING THE IMPLICATIONS OF ARCHITECTURAL SOLUTIONS TO MARKET
STRUCTURES
199