Abstract: Technical interoperability is often
considered as a pre-requisite for a global and open e-commerce
marketplace. The proliferation of e-commerce architectures encourages
market fragmentation. The commodity view of technologies seriously
undermines the viability of individual e-commerce businesses.
The disconnect between technologies and business practices creates
short termism in business model creation and funding of the e-commerce
industries. A seamless value chain that underpins an economic
structure is ultimately dependent upon a shared approach to and
a shared view of business processes and business semantics. Fragmentation
at the basic level of architectures is not only commercially
detrimental to individual businesses, it also decreases the long-term
benefits of e-commerce to the market as whole.
1. Introduction
Interoperability is widely
regarded as one of the most critical and urgent issues for the
future development and growth of e-commerce. While the concept
of interoperability is not new, how it needs to be applied, interpreted
and implemented at a practical level is however subject to continuous
discussions and occasional controversies. This paper discusses
the different approaches to interoperability from the technical
as well as business perspectives. In particular, it seeks to
highlight how and why considerations of interoperability are
deeply related to the activities of the various groups of stakeholders
in e-commerce. It raises the question of whether a uniform approach
to interoperability is possible and feasible.
Secondly, this paper reviews how interoperability relates to
development of business models in the e-commerce space. Drawing
on practical experiences of creating and then implementing business
models for e-commerce, it illustrates the business problems surrounding
and arising from the various interpretations of interoperability,
and the considerable gap between interoperability as a technical
concept and interoperability as a building block for an e-commerce
commercial venture.
2. Technical Approaches to
Interoperability in e-commerce
Technical interoperability
is often considered as a pre-requisite for a global and open
e-commerce marketplace. It has been variously interpreted, such
as:
- Using standards or "standards-based"
protocols
- Adopting a common technical
framework at a basic level
- Providing a "core set"
of services or service feature sets
- Implementing "complementary
products" or a basic suite of software tools
- Defining a stable system architecture
that has the support of the largest possible number of entities.
Technical interoperability
therefore is closely linked to standardization. Indeed, many
have argued that without standardization, technical interoperability
is not achievable. However, those who advocate this view do not
often agree on what needs to be standardized and secondly, how
it is to be - or should be - achieved.
In recent years, a large number
e-commerce frameworks, architectures and models (these three
terminology is often used interchangeably and the distinction
between them is a continuous source of debate in technical environments)
have been developed within the standardization community and
elsewhere, in an attempt to advance interoperability. The CEN
Workshop Agreement (CWA) on this subject, for example, provides
an analysis of over twenty of such efforts [1]. As the CWA acknowledges,
its coverage is by no means exhaustive. The sponsors of the individual
efforts include standardization committees, industry consortia,
vendor communities, individual vendors, as well as user communities.
Table 1 e-commerce Frameworks,
Architectures and Models and Their Sponsors
SPONSORS |
FRAMEWORKS, ARCHITECUTRES
AND MODELS |
General Frameworks |
Microsoft |
BizTalk |
CEN/ISSS Electronic Commerce
Workshop |
Building Blocks |
UN/CEFACT and OASIS ebXML |
Technical Architecture |
CommerceNet |
eCo Framework |
Esprit IMPRIMATUR Project |
IMPRIMATUR Business Model |
ISO TC184 and related groups |
Industrial Data Framework
(STEP) |
Sun Microsystems |
Java Electronic Commerce
Framework |
OMG Electronic Commerce
Domain Task Force |
OMG Electronic Commerce
Domain Specifications |
ISO/IEC JTC1/SC32 |
Open-edi Reference Model |
Trading Models |
Various |
Ad Hoc Functional and Process
Models |
IETF TRADE Working Group |
IOTP - Internet Open Trading
Protocol |
Open Applications Group |
Open Applications Group
XML Framework |
CommerceNet |
OBI - Open Buying on the
Internet |
RosettaNet |
RosettaNet |
ACTS SEMPER Project |
SEMPER - SECURE ELECTRONIC
MARKET PLACE FOR EUROPE |
Payment Models |
Various |
Electronic Payment Technologies |
VISA and MasterCard |
SET - Secure Electronic
Transaction |
CEN/ISSS TC 224 and ISO/TC68/SC6 |
Trading and Payment Model
in joint report on card-related secure commercial and financial
transactions |
Security Models |
IETF PKIX Working Group |
PKIX |
CEN/ISSS TC 224 and ISO/TC68/SC6 |
Security Model in joint
report on card-related secure commercial and financial transactions |
Mobile Commerce Models |
Ericsson, Motorola, Nokia,
Panasonic, Siemens and Sony |
MeT - Mobile electronic
Transactions |
As the categorisation in the table shows, these frameworks/architectures/models
have different focuses. Moreover, within an individual category,
they often centre on particular themes or approaches. For example,
BizTalk positions interoperability primarily in terms of data
representation and formatting, Building Blocks in terms of business
processes/sub-processes, ebXML in terms of messaging, semantics,
processes and increasingly modelling, eCo in terms of systems,
Java EC Framework in terms of software components, and so on.
Some of them also have a notably sector flavour, such as IMPRIMATUR
(content industries) STEP (manufacturing), RosettaNet (semiconductor),
MeT (telecoms and IT). On the other hand, while some of these
architectures overlap in certain areas, the bounds of their specifications
rarely coincide. In other words, it is exceedingly difficult,
if not impossible, to make a meaningful comparison between these
available architectures. Moreover, the possibility of interoperability
between two systems that have implemented different architectures
is close to nil, at least without extensive and expensive interfacing
exercises which are specific (i.e. proprietary) to the two systems
concerned. The CEN/ISSS Electronic Commerce Workshop, therefore,
has established a project to develop a meta-framework that provides
a modelling language, methodology and open source tools for facilitating
de facto interoperation, or mapping, between two incompatible
systems [2].
The proliferation of architectures
is often accompanied by exhortations to converge efforts on a
single architecture, or to pool resources to develop a common
"baseline" or "reference". However, in so
far as the vast majority of these architectures do not share
the same foundations, especially at the level of business process
and semantic definitions, it is difficult to see how convergence
can be realised, even on technical grounds alone.
3. The Context for Interoperability
in e-commerce
The problems of interoperability
have not generally enhanced the reputation of standardization.
Formal standardization is perceived to be slow, unresponsive
to market needs, unrepresentative of the various interests involved
etc. - these arguments are well known and well rehearsed. At
the same time, the past decade has witnessed a mushrooming of
industry consortium activities, in e-commerce architectural development
and other areas. This trend shows no sign of abating.
To put it simply, interoperability is about the use of a common
set of technical specifications. These specifications need to
be implemented into products (by vendors). The products need
to be implemented (by users) for delivering particular services.
Often, products are bundled together and made available as specific
applications (by vendors or users). In this value chain of transforming
basic IT products into e-commerce services, many parties or intermediaries
could be involved, including service providers, integrators/aggregators,
consultants etc. While e-commerce has "dis-intermediated"
many parties in the more traditional value chain, it has equally
spawned a new generation of intermediaries along the (re-structured)
value chain spanning producers of the raw material both tangible
and intangible to the end consumer. Many predict the increasingly
prevalence of the dynamic value chain, which is constituted and
de-constituted on the fly, elastic to market demand in almost
real time.
The proliferation of e-commerce
services is in principle not dependent on interoperability as
defined above. Arguably, the reverse may be true. However,
the proliferation of e-commerce services that also enables easy
access to the market by those who wish to offer services and
by those who wish to purchase services is dependent upon an open
marketplace. Openness in this context can be defined in terms
of entry barrier in respect of cost of entry, and for the provider,
also time to market, the degree of market fragmentation, and
the availability of innovation opportunities. Interoperability
is a necessary, though not sufficient, condition for market openness.
Conversely, lock-in regarding a particular technical architecture
that is implemented into specific product(s) potentially jeopardises
market openness. As software becomes increasingly sophisticated
and an e-commerce service, let alone services, needs to rely
on multiple IT products in its delivery, the architectural basis
has simultaneously become the more complex, and the technical
openness of the architecture the more difficult to achieve.
Technical openness is, like
interoperability, a concept that often confuses rather than elucidates.
A technical specification that is open is generally defined
as one that is publicly available in terms of accessibility.
However, accessibility is not the same as implementation or
use. An open architecture may comprise a myriad of technical
specifications, all of which are publicly accessible, but with
different licensing agreements attached. Second, an open architecture
does not guarantee uniform implementation. Lack of interoperability
between different implementations of the same specification is
fairly common in the industry. There are also cases of lack
of interoperability between different versions of the same product
that implements the same specification. Third, the individual
specifications that comprise an architecture do evolve over time
and maybe at different points in time. Versioning of an architecture
therefore may not be a straightforward matter which has ramifications
for implementation compliance. Fourth, compliance itself is
a malleable term, depending upon the compliance criteria, which
may not ensure objectivity, neutrality or completeness.
E-commerce interoperability presents an enormous challenge because
of the range of technical requirements in supporting a typical
e-commerce business, which as experience shows, rarely succeeds
without careful consideration of the business processes which
span the value chain. Outsourcing may be a practical solution
for some businesses, but it is not a substitute for value chain
management. The execution of business processes requires not
only technologies and specifications in the more "visible"
areas such as architectures, semantics, electronic data interchange,
information security and payment mechanisms. Depending upon
the particular services offered, it typically also requires technologies
and specifications in support of information management, data
representation, communications, and specific sectorial needs.
The Diffuse Standards and Specifications List, part of the IST
Diffuse Project's online services, for example, provides a classification
scheme of some thirty individual areas that may apply to offering
e-commerce services [3]. The hype surrounding a particular technology,
therefore, does not remove the fundamental need for meeting a
wider range of technical requirements. As a general observation,
given the large variety of technologies involved in delivering
particular services, it is difficult to see how an individual
technology - no matter how topical - can solve all the problems,
or can guarantee interoperability, or openness. Indeed, it is
unclear what these terms really mean in the context of such a
claim.
4. Interoperability from a Business Perspective
Interoperability is unlikely
to be a familiar term in the language of an e-commerce entrepreneur,
who is typically more concerned with the business model, the
bottom line, resourcing, staffing etc. Even in the "new
economy" companies who deliver "pure" e-commerce
services, technologies are often dealt with separately from "the
business", though there are notable exceptions. While technologies
lie at the heart of an e-commerce business, the gulf between
the "tech" and "non-tech" parts of the business
is often immense, in the corporate culture, organisation, and
strategy setting.
There is a pervasive view among
senior business managers in e-commerce service companies that
technologies are a commodity and a company's technology capability
is a matter of financial resourcing - a view that is not discouraged
by the IT providers to these companies. IT projects are not always
aligned with business targets and objectives. The expectations
from IT are not always realistic, leading to disappointment and
even failed businesses.
The widespread availability
of commercial off-the-shelf IT products and their generally falling
prices is also partly responsible for the view of IT as a commodity.
Modularisation of IT components encourages the perception that
applications can be easily added or adapted to meet new business
needs. The assumption is that the underlying architecture is
scalable, robust and "open". Companies often find
out belatedly, and to their cost, the complexity of integrating
apparently trivial changes to "the system". Industry
statistics indicate that a staggering 90% of IT projects do not
deliver on time or are abandoned.
The disconnect between technologies
and business models has a profound impact on the development
of e-commerce. A common way of segmenting the e-commerce market
is in accordance with the nature of the entities involved and
how transactions are organised between the entities.
The compartmentalisation of
technology has not only impacted on the organisation of e-commerce
businesses, but also on the funding structure and evaluation
models in venture capital and institutional financing. As market
fashion has shifted from B2C to B2B and more recently to C2C
or P2P etc, the continuous quest and re-invention of business
models that meet market approval has detracted focus from the
fundamental business processes and the technology infrastructure
for supporting these processes. It also encourages a short termism
in creating business models that are contingent upon the emergence
of the "killer app" or the "killer adoption".
More fundamentally, it encourages the view that there are separate
universes of business operations, distinct technology paradigms,
and islands of value chains. It glosses over the fact that an
individual who purchases books online is the same person who
orders goods via a B2B hub and who is the "customer"
of e-government services. From a technology perspective, a seamless
value chain that underpins an economic structure is ultimately
dependent upon a shared approach to and a shared view of business
processes and business semantics. Fragmentation at the basic
level of architectures is not only commercially detrimental to
individual businesses, it also decreases the long-term benefits
of e-commerce to the market as whole. Commercial differentiation
needs not be equated with market fragmentation. A diverse and
a dynamic e-commerce marketplace is dependent upon a commonly
shared framework in different contexts.
5. Conclusions
This paper advocates the need
to re-examine the concept of interoperability as it is applied
in the technical and business contexts. It argues for a better
understanding of the complex issues surrounding interoperability
and how some of them impinge directly on the e-commerce marketplace
in practice. Ultimately, how these issues are to be approached
and tackled will determine the degree of openness of the e-commerce
market in future, and the extent to which the benefits of e-commerce
will be realised by the different sections of the community.
References
[1] CEN/ISSS Electronic Commerce
Workshop, CEN Workshop Agreement - Summaries of some Frameworks,
Architectures and Models for Electronic Commerce,
CWA 14228, published in June 2001 (available from CEN Member
National Standards Bodies)
[2] CEN/ISSS Electronic Commerce Workshop Project Group on E-Commerce
Integration Meta-Framework (ECIMF), 2001-2002, http://www.cenorm.be/isss/workshop/ec
[3] IST Diffuse Project, http://www.diffuse.org |