What is Electronic Commerce?
Defining electronic commerce is problematic; its versatility and intuitiveness ensure that often it is not defined, or that it is defined in an all-encompassing fashion where it is taken to refer to any transaction conducted over a network. Illustrative definitions include:
Electronic commerce is a general term for the conduct of business with the assistance of telecommunications, and telecommunications-based tools (Clarke, 1993).
Electronic commerce refers generally to all forms of transactions relating to commercial activities, involving both organizations and individuals, that are based upon the processing and transmission of digitized data, including text, sound and visual images (Sacher Report, 1997).
Such definitions are broad, embodying in a single phrase a disparate range of activities which includes email as well as database access, trading support mechanisms for the online sale of goods and services, electronic data interchange and payment systems (Clarke, 1993). Consequently, it is possible to point to the large value of transactions conducted electronically, for instance, the level of business-to-business electronic commerce has been estimated at $114 billion for 1999 (Cohn, 2000).
Some, however, have been more parsimonious in their definition of electronic commerce. Definitions have been proposed which limit electronic commerce to business-to-business, or business-to consumer-transactions. The ITU (1997) distinguishes between those services that are ordered, billed and consumed entirely online, and those that are billed and consumed off-line. The OECD (1997) makes an additional distinction between those companies using electronic commerce as another sales channel for their existing activities, and those whose sole business is electronic commerce. The same report demonstrates that Internet-based electronic commerce, from business to consumer, is dominated by companies using the Internet as an additional sales channel and that these sales in comparison to the other forms of electronic commerce are relatively minor.
Figure 2 demonstrates the different focus of electronic commerce definitions. The broadest definitions are to be found at the base of the pyramid and include electronic funds transfer and credit card transactions. In the ‘electronic commerce infrastructure’ layer definitions make reference to the need for electronic commerce to occur over the underlying infrastructure, although ‘business to consumer’ oriented definitions include all electronic forms of business and customer interactions. The most restrictive definitions are located within the top layer of the pyramid, and are limited to only those instances where an electronic transaction occurs between a business and a customer.
Although most discussions acknowledge the diversity of electronic commerce, they subsequently limit their definition in some manner (The Economist, 1997; OECD, 1997; den Hertog, Holland, & Bouwman, 1999; Henry, Cooke, Buckley, Dumagen, Gill, Pastore, D., & LaPorte, 1999). The Economist (1997) provides an apt illustration of this in practice. After arguing that the scope of electronic commerce is in fact broad, containing within it a diverse array of business activities, The Economist (1997) then limits its focus:
[s]o why single out the Internet? Because it alone has the potential to deliver what the notion of electronic commerce has always implied. Computerized financial markets gave brokers equal and instant access to information and let them act on it there and then. The Internet promises to do the same for everyone from the individual investor to the casual shopper. Credit cards spurred home shopping by creating a virtual payment system that transcended national borders. The Internet extends this beyond the transaction itself to everything that comes before and after, from marketing and product display to order-tracking and sometimes even delivery. And unlike the commercial online services, which reserve their service for their subscribers and selected merchants, the Internet is open to everyone (The Economist, 1997: p. 5).
Interoperability: Openness and Interfaces
The growth of electronic commerce depends critically on one issue: interoperability between different systems. Oftel defines interoperability as:
… the technical features of a group of interconnected systems … which ensures [sic] end-to-end provision of a given service in a consistent and predictable way. (OFTEL, 1997)
Thus, interoperability links systems together. For example, it enables banks to link together their cash-point networks, extending the number of points at which service can be provided in a consistent and predictable fashion even though ownership remains split between the individual banks.
Central to interoperability are the two inter-related issues of interfaces and openness. Where two networks adjoin one another each must be able to understand the operation of the other if a service is to cross between the two. Interfaces, the technical and physical links between networks, are the mechanisms through which this understanding occurs. Network interfaces exist at the point where one network connects with another. However, as interoperability must also occur between the network and user, another set of interfaces, termed user interfaces, exists at the user's point of connection with the network. Both types of interfaces must co-operate with one another in the sending or receiving of a service.
Interfaces are intrinsic to interoperability, but if this is to occur the technical specifications of interfaces need to be open:
… an interface is open if its specifications are readily and non-discriminatorily available to all vendors, service providers, and users, and if such specifications are revised only with timely notice and public process. (Eurobit-ITIC-JEIDA Paper, January 1995, quoted in Band, 1997)
From this two characteristics of openness can be discerned; first, that interface specifications should be transparent, and secondly, that they should not be subject to unilateral alteration by one party to the detriment of others. Both of these are at odds with Microsoft's definition of openness, where voluntary licensing occurs, as this permits the possibility for unilateral alterations, as well as competitiveness enhancing hidden specifications. Moreover, the commercial imperative ensures that companies will be reluctant to license for fear of creating a competitor, as Band (1997) notes:
Microsoft … has not agreed to license its interface specifications to firms seeking to develop operating systems that compete directly with Microsoft operating systems. (Band, 1997)
Contrary to the notion of openness is the unilateral alteration of interface specifications by either a company, or a standards forum. Implicit within the notion of openness is the transparent and consensual determination of interface specifications, so that once set such specifications take on many of the characteristics of public goods. However, if one party is able, through whatever means, to alter interface specifications without the consent of others, it will be in a position to lock in customers, as well as extract higher returns from the market. Once undertaken the rest of the market must decide to either follow suit, or remain with existing specifications that would split the market into competing camps. Consequently, in a fragmenting market interoperability is dependent not only on interfaces within the same specifications group but also between different groups.
Fragmented but interoperable interface specifications are advantageous, as they provide a stimulus for technology-based innovation:
… developers of specifications for interfaces must be able to retain ownership of and benefit from the intellectual property that goes into the specifications, in order to maintain incentives to develop new technologies. (Computer Software Policy Project, quoted in Band, 1997)
Intellectual property rights provide the innovator with the ability to recoup its research and development costs. Furthermore, the lucrative nature of captive markets engenders competition through the entry of other companies wishing to extract income from the market for themselves. Eventually it is possible to imagine a competitive market, both within and between interface specification groups, where competition is vigorous and innovative. Possible benefits under such a scenario include reduced prices, enhanced innovative tendencies by companies, and improved choice.
A single set of open interface specifications are also capable of acting as a platform for competition while maintaining interoperability, once a distinction is drawn between specifications and implementations:
… interfaces specifications are pieces of paper; implementations are actual products or services. (Band, 1997)
Companies will be free to combine open specifications uniquely to create goods and services that are proprietary in nature. These then compete with one another:
… this combination of non-proprietary interface specifications and proprietary implementations meets the imperative of balancing the requirement of providing incentives to developers of new technology with the societal need for interoperability…. (Band, 1997)
At all times interoperability will be retained due to the open specifications on which products and services are based. Even so this may not encourage companies to enter the market. The market leader may be highly entrenched, and display sufficient market power so that companies are deterred from entering the market.
A Multi-Level Approach to Interoperability
Interoperability allows systems to be connected to one another, with services being delivered in a consistent and predictable fashion. As it has already been shown this may occur at either the network or user level, but these are not the only instances of interoperability in electronic commerce.
A multilevel approach to interoperability is depicted in Figure 3 below. In this multilevel approach interoperability will be influenced by two sets of factors. The decision as to whether a good or service should be made or bought influences interoperability. If a good or service is bought in from external sources interoperability becomes a more prominent issue. Furthermore, if the acquired good or service is to be incorporated into the existing activities of the firm interoperability is essential.
The scale of interoperability is shaped by whether or not the transaction is retail or wholesale in nature. In retail transactions interoperability is on a limited scale, whereas wholesale transactions require that interoperability is possible between a potentially much larger range of goods and services.
Interoperability allows one end user to communicate or transact with another in a consistent and predictable fashion. For all intents and purpose interoperability at this level occurs directly between the end users, even though communications and transactions pass from one to the other over one or more vendor networks. Moreover, interoperability occurs at the individual level, between each of the end users, and with a limited range of other goods and services. It is not the case that an infinite combination of goods and services is possible, as limited resources will be available at either end point.
•Business supplier to end user
Two interpretations of business supplier to end user exchange relations are imaginable. The first interpretation involves a business supplier and individual user while the second is business-to-business in nature. In the first of these the interoperability imperative rests with the individual. It is frequently suggested that one consequence of dis-intermediation is that users willing to trade seek out suppliers. If users cannot interoperate with the supplier's systems the initial stimulus, desire, is left unfulfilled. It is clear that in this instance interoperability is influenced by trade and retail considerations.
In the second imaginable scenario the individual end user is replaced by another business. Through conforming to purchaser behavior the interoperability imperative remains with the initiator of the transaction. However, if the balance of this arrangement alters so that it becomes more equitable both parties will see the benefits from interoperability. In this second scenario the wholesale model is more appropriate.
An instance of the first scenario would be the online purchase of goods and services from, say, Amazon.com, and the second is exemplified by financial institutions access to online services and databases.
The issue of interoperability on a network-to-network basis can be addressed through reference to interconnection. In the telecommunications industry interconnection is:
… the means by which the providers of one set of infrastructures are able to establish “any to any” connectivity across networks whether these are alternative suppliers or incumbents. (Scanlan, Williams & Whalley, 1998)
According to Louth (1997), the EU's interconnection directive is underpinned by three principles, similar to those for open interoperability interface standards, namely: 1. Transparency. The terms of interconnection are available to all as public documents. 2. Non-discrimination. Not only do the same rules apply in similar circumstances but the incumbent should not favor itself in any shape or form. 3. Cost-orientation. Charges must be cost-oriented as well as being sufficiently unbundled, but only if significant market power exists; otherwise, processes are determined by normal commercial considerations.
Without interconnection and interoperability where standards differ, international telephone calls would not be possible. For instance, with interoperability it is possible to originate a telephone call in London and terminate it in Chicago. In doing so operators are required to co-operate, as laid out in interconnection agreements, so that the call can be routed from one network to another so that termination is successful. Where the call moves from one network to the next points of interoperability will occur.
As it would be impractical for individuals to negotiate interconnection agreements whenever a network-to-network call is necessitated, arrangements are agreed on a company-to-company, or wholesale, basis.
Until now it has been assumed that interoperability is between two or more telecommunications networks, but it may be the case that interoperability is required between different types of networks. Though telecommunications networks are the principal way through which the Internet is accessed, other electronic commerce delivery media, such as broadcast, are also available. 1 The distribution of a given service through a combination of broadcasting and telecommunications networks requires interoperability at the point where connection occurs.
•Network service to network service
Within both telecommunications and electronic commerce a plurality of service providers exist. This enables the market to be characterized by on the one hand universal service providers, and on the other specialist boutiques offering niche or a limited number of services. Motivated by quality, pricing or convenience considerations, consumers may eschew the universal service providers in favor of the more specialized operators.
In such circumstances interoperability is necessary so that various services can be combined. This would enable, for example, the integration of voice with video images at the point of origin. Interoperability occurs at the point where the two services are combined. Without interoperability the two complimentary service components would remain separate. The combination of two or more services is an illustration of horizontal interoperability.
The ability to combine service components facilitates proprietary combinations. Depending on the scale at which these are conducted the combinations may involve either, involve trade between provider and individual user, or wholesale arrangements where service components are bought and then combined by companies before being sold to third parties. A telecommunications parallel to this latter case would be unbundled network elements bought by new entrants from incumbents and then subsequently resold to customers.
•Network to network application
Interoperability at this level occurs at the point where the underlying network and the network application connect. Consequently, interoperability is vertical in nature. Generic examples of network applications include voice as well as fax and email, while sectorial instances include credit card/financial payment mechanisms in the financial sector, and reservation systems in the travel industry. Airline reservations systems exemplify specialized network applications.
A network application which is generic needs to be interoperable as more likely than not it will be incorporated with other applications and services. Within electronic commerce, payment systems will fit into this category. Interoperability will be necessary whether or not the network application is at the retail (individual) or wholesale (mass) scale. One possible distinction is that in the former case interoperability may be more limited when compared to the latter; a whole network application with limited interoperability would reduce the possible combinations, and thus commercial opportunities, of this with other applications.