ABSTRACT:

Although computing technology has been used in many organizations to increase the efficiency, accuracy, scale, and scope of accounting-related information, use of computing has not changed the basic system of debit-and-credit-based income measurement. The proposals in this article suggest a way for accountants to apply the power of computing technology to a fundamental step in the income measurement process, namely, interpreting the reality defined by the firm's recorded history of past economic events.

In essence, it is proposed that firms should continue to use their own different special-purpose transaction processing systems (TPSS) for transaction capture. However, rather than building specifications of general ledger account numbers into the transaction capture process, programs called interpreters could extract data from the TPS files or databases for use in generating formula accounting (IA) journal entries (Seddon 1992). Interpreters with different sets of rules can be used to generate different accounting interpretations of the underlying economic events. This article identifies the main data requirements of the proposed architecture and discusses the rules that need to be encoded in the interpreter programs if they are to generate IA journal entries from those data. Data requirements are presented in the form of an accounting data model called the resources and exchange events (REE) model. To demonstrate the feasibility of the architecture, a prototype system is constructed.