1、中文 2020 字， 1187 单词 外文原稿二： The production and use of semantically rich accounting reports on the Internet: XML and XBRL ByRoger Debreceny， Glen L. Gray Business reporting on the Web Corporations have been posting financial information on the Web since the mid-1990s. Indeed, the corporate finance and
2、investor relation functions were “early adopters” of Web technology. Use of the Web for financial reporting is now globally pervasive, as reported in the following studies: 99% of the top 100 Fortune 500 companies have Web sites, and 94% include financial information (FASB, 2000)； 69% of 10,000 corp
3、orations that trade on the NYSE, NASDAQ, and TSE have Web sites, and 35% include financial information (Trites, 1999)； 86% of the top 30 corporations from each of 22 countries (660 companies) have Web sites, and 62% include financial information (Lymer et al., 1999). Reporting by corporations on the
4、 Web has been accompanied by a variety of “portals,” such as（ ） and （ ） that provide information on corporations including pointers to corporations “home pages” and reporting information. Recently, the issue of reporting on the Web has been of considerable interest to standard setters and accounting
5、 bodies. The Financial Accounting Standards Board (FASB) has published a study on electronic reporting (FASB, 2000) as part of its ongoing study into expanded forms of reporting, flowing from the 1994 report of the Special Committee on Financial Reporting (the “Jenkins report”； AICPA, 1994). The Int
6、ernational Accounting Standards Committee (IASC) commissioned and published a report on Web-based business reporting (Lymer et al., 1999). The report discussed the wide variety in both the extent and the nature of reporting on the Web around the world, and recommended (1) the development of a “code
7、of conduct” to guide preparers on best reporting practice and (2) the development of a so-called “Business Reporting Language” to harmonize and integrate reporting in the IASC domain. Finally, the Canadian Institute of Chartered Accountants published a report on the nature and extent of reporting on
8、 the Web in North America (Trites, 1999). Meeting stakeholder information needs Stakeholder needs With electronic dissemination of financial information on the Web, users, both in person and supported by intelligent software agents, can directly monitor the performance of corporations. At the lowest
9、 level of functionality, a user can manually navigate to the Web site of a corporation to view the desired information. At a higher level of functionality, the user can request particular information from a search engine (e.g., “what were the total 1999 sales for General Motors?”). The highest level
10、 of functionality would be the automation of the whole mechanical process of locating, compiling, and analyzing financial information. For example, as envisaged by commentators, such as Elliott (1994) and Wallman (1997), intelligent software agents 2 could retrieve accounting and financial data from
11、 Web sites to use in investment models. An intelligent software agent3 employs techniques drawn from autonomous programming, information retrieval, artificial intelligence, and inferential statistics.4 The agent undertakes information search, retrieval, and decision making on behalf of human princip
12、al, and it may act in an autonomous or semiautonomous fashion. Intelligent software agents are beginning to move from the laboratory to the electronic commerce environment, particularly in support of business-to-consumer electronic commerce (B2C EC) (Maes； Maes and McCubbrey) and supply chain integr
13、ation (Khoo； Thibadeau and Nissen, 2000). In a financial context, the intelligent software agent could rank alternative investments or build theoretical portfolios. The agent could automatically notify investors when investment rankings or portfolio values have moved beyond the bounds of designated
14、parameters. The users own Web-browsing patterns can also provide the intelligent software agents with inputs upon which to act (Cheung et al., 1997). The collective intelligence of the users and their agents can also provide agglomerated analysis of the Web-based financial information to rank invest
15、ments, to raise green or red flags, or as a tool for enhanced shareholder democracy. 5 In short, intelligent software agents could be used to monitor and respond to the financial and nonfinancial condition of a corporation. The requirements for users of financial information are that information be
16、available over time to facilitate longitudinal analyses and, at least within an industry, be reported in a consistent fashion to make possible the analyses of peers (AIMR and Palepu). 6 Unfortunately, as the following discussion demonstrates, the current state of the Web would make widespread deploy
17、ment of intelligent software agents to meet these twin requirements nearly impossible. Incentives for corporations to provide information with underlying semantic value Nothing in the reports of the FASB, IASC, or CICA discussed in Section 2 suggest that reporting on the Web be made legally required
18、 in any of the jurisdictions. It seems that any move to provide information on the Web with higher semantic value than that provided in the current form of HTML and Adobe Acrobat will be voluntary. The incentives that will drive corporations to such report will depend on the perceived benefits to be
19、 derived from the disclosures and the cost of making the disclosures. It should be noted that at this stage in the development of Web disclosure, there is consideration only of changing the mode and not the substance of disclosure. In considering the potential demand for and cost of Web-based financ
20、ial reporting with high semantic value disclosures, network economics is at play (Economides, 1996). Even the most casual empiricism demonstrates that over the last few years, a number of new information intermediaries, such as（ FreeE） , have come into existence to leverage the widespread availabili
21、ty of Edgar filings. It is reasonable to expect that current and new intermediaries would similarly take advantage of widespread availability of high quality and deep information on corporate financial performance. Further, it is likely that existing information intermediaries would quickly take adv
22、antage of high quality information in digital form to feed the databases mentioned above. If the cost were at an acceptable level and a clear market demand for Web-based financial reporting with higher semantic value established, it is entirely likely that corporations will meet the perceived demand
23、. If there was widespread adoption of a standard for Web-based disclosure that was in use by key stakeholder communities, such as analysts, presumably, those companies who choose not to make such disclosures may consider that they would be penalized in the capital markets in terms of higher cost of
24、capital compared to their peers who do make such disclosures (Botosan and Sengupta). If a standard can be agreed that has widespread acceptance in both the preparer and user communities, it is arguable that the cost for corporations and for their auditors in preparing the disclosures will be reduced
25、. The manufacturers of accounting, consolidation, and audit workpaper software may be induced to integrate high semantic value Web-disclosures into their packages. It can be argued that the early adopters of such a reporting technology may gain a first mover advantage (Arthur, 1989). The reaching of the “take off” point at which stage the demand for high semantic value reporting drives most corporations to make such disclosures and which at the same time
