پشتیبانی از گردش کار برای کاربردهای تجارت الکترونیک

Internet-based electronic commerce is becoming the next frontier of new business opportunities. However, commerce on the Internet is seriously hindered by the lack of a common language for collaborative commercial activities. Although Extensible Markup Language (XML) allows trading partners to exchange semantic information electronically, it does not provide support for document routing. In this paper, we describe various inter-organizational electronic commerce applications and discuss their needs for workflow support. Then, we propose a blueprint for XRL, an Extensible Routing Language that enables routing of commercial documents over the Internet and helps in creating truly intelligent documents. This routing language is simple, yet powerful enough to support flexible routing of documents in the Internet environment.

The Internet and the World Wide Web (WWW) have changed the landscape of networked computing and have become the de facto environment for electronic commerce. However, the current electronic commerce technologies rely on much in-house programming activities and are inefficient and lack interoperability. The trend is to develop more standardized architectures and techniques for open electronic commerce services [4], [5], [11] and [17]. One important thrust for increased productivity and interoperability is to develop more homogeneous languages for various electronic commerce activities [9]. A second thrust is towards developing autonomous, cooperating agents that can communicate intelligently with one another [16]. In this paper, we focus on the development of a language that provides support for routing of workflow for Internet-based electronic commerce services. In the spirit of HTML which provides support for rendering documents in a platform independent manner, and XML (Extensible Markup Language) which allows exchange of semantic information, the proposed language called View the MathML sourcextensible View the MathML sourceouting View the MathML sourceanguage (XRL) provides support for routing documents and managing workflows across trading partners. The Web has evolved through various stages. It started as a way of accessing distributed information on the Internet using a GUI interface based on the HTTP protocol and the HTML language. It was a successor to Gopher, WAIS and Archie, which were also information access and knowledge discovery tools but were based on text-based interfaces [13]. The next important advancement was to make the HTML protocol more interactive using the FORMS feature so that commercial tasks such as buying and selling, filling in surveys, searching databases, etc., could also be performed. In 1997, about US$40 billion worth of transactions were performed over the Web. This figure has grown exponentially since then, and is now around several hundred billion dollars. Several architectures for Web commerce have been proposed, e.g. Commerce Net, etc. [10] and protocols like SSL and SET [6] have been developed to address security issues that arise on the Web. Table 1 summarizes the main stages through which the web has evolved. Stages 2 and 3 reflect the current stage of development of the web. XML [19] and [20] makes it possible to add semantic information to an HTML document so that trading partners can understand the meaning of various fields in the document. Fig. 1 shows a common model of how the web operates to provide interactive services such as shopping, database access, etc. The various steps are as follows: (1,2) A client connects to the server and downloads a form. (3) The client fills in information on the form and submits it. (4,5) A CGI program on the server processes the form. (6) The server sends a reply to the client. The features of this model are that all interaction is in synchronous (request, reply) mode. While it works very well for a variety of applications, there are other instances where it is not so effective. In particular, some of the problems are as follows. First, it is not always possible to establish a connection between the client and server either because the underlying network is unavailable or the server is overloaded. Secondly, this is not a very efficient way of communication when a large amount of information has to be exchanged. Thirdly, this form of communication is not very conducive to flow-type applications, which involve a flow of information/documents through multiple workers in several organizations. Thus, there is a need for a framework that exploits the advantages of the web, and yet provides better support for asynchronous, flow-type commercial applications. This kind of situation arises frequently in supply chain management [2], [3] and [15]. In such situations, asynchronous transmission of messages can be more reliable and efficient. For instance, it is considerably easier and more efficient for a customer to place large numbers of orders (for stocks, etc.) or make bids in an electronic market (for several products at the same time) asynchronously. This paper is organized as follows. Section 2 discusses the need for workflow support in Electronic Commerce. Section 3 gives an overview of the intelligent document architecture based on routing slips. Next, Section 4 describes the XRL language. Then, Section 5 presents a preliminary design of the document routing assistant based on the XRL language. Section 6 discusses various theoretical issues that arise in the context of this architecture, the language, and the system. Finally, Section 7 concludes the paper.

Support for routing is a major element missing from current architectures for open electronic commerce. We described a proposal for routing documents on the Internet based on routing slips. The routing slips are a convenient metaphor for specifying flexible routing schemas and running asynchronous (flow-type) transactions. They can be specified in XRL, a markup language for defining a routing schema. Routing slips would be used for routing documents on the Web and implementing inter-organizational workflow systems. XRL is an extension to XML and enables efficient routing. This is an essential step towards the goal of creating truly intelligent documents on the Web, which contain semantic, routing, and permission information. XRL has a syntax similar to HTML and XML, and is easy to use and flexible. It should be noted that XRL is a routing language and does not have a data model to accompany it. This can be viewed both as a strength and weakness. It is a strength in that it isolates the modeling of control flows from the data modeling issues. After all, workflow modeling primarily involves modeling the control flows associated with the performance of information tasks. On the other hand, lack of a data model is a weakness in that it does not allow processing of workflows based on specific contents of the data in the documents that are being routed. Note that we do allow documents to be attached to the routing slips, but do not impose any structure or restrictions on the contents or format of these documents. Compared to routing mechanisms in commercial workflow systems such as SAP [12], XRL offers a lightweight and more affordable alternative. A conventional workflow solution only works in a closed environment where the client and server must be able to speak a proprietary language and the system integration is usually through an Application Programming Interface (API). XRL follows the design philosophy of HTML and XML, and may be applied to production, administrative, and ad hoc workflows [7]. A detailed architecture for the document routing assistant required to support an implementation of XRL was given in Section 5. As a next step in this research, one of the authors has embarked on a project to develop a next version of XRL, which is entirely XML compliant [23]. The advantage of doing so is to be able to take advantage of existing XML parsers to validate and parse the routing slip. Moreover, the newer version will add expressive power and overcome several shortcomings of the current version. XML is a markup language that allows users to define a set of tags, which specify the structure of a document [19] and [20]. For example, a user may specify tags for name, age, department, email, etc. This logical structure may be stored either in the document file itself or in an associated file called the document type definition (or DTD) file. While XML can help in exchange of semantic information, it still lacks routing information. Such information is critical to enable proper routing of a document within an organization and across organizations. Finally, here is a list of future work we foresee in this area: •Support for negotiation described in Section 2 must also be added. •More work is required to develop a definition of routing completeness. How can we describe all routing scenarios? • A more formal approach for defining asynchronous transactions is required. • Implementation of the Document Routing Assistant (DRA).