NIST - ATP Meeting March 9 and 10, 1998 Defining the Advanced Technology Notes from presentation by: Dwight Koop Track 2 - Security and Payment Technologies Automated Adjudication I would like to limit my opening comments to just one of the many challenges that must be overcome before Internet based electronic commerce can reach national and global scale. The simple challenge I present to you is one of extending one of today's most common commercial mechanisms to better fit the needs of Internet based commerce. I would like to place before my colleagues, for further discussion, the issue of NON-REPUDIATION. Actually, the need for a public Internet dispute infrastructure, or "Automated Adjudication." Looking forward several Internet years we all expect open networks (like today's Internet) to be the transport for most of the world's commercial risk transfer and payment transactions. Not just Internet based catalog shopping; but, fast forming virtual markets. The Internet is the perfect medium to support fast forming virtual markets (Internet markets for serendipitous online trading between many previously mutually unknown, yet equal trading counterparties). This level of ad-hoc interoperability seems to be the intended goal and the likely result of the work we are all doing today to assure security, merge payment systems, define commercial languages, and integrate software environments. In these markets, repudiation, of both the legitimate and fraudulent kind, will continue to exist. Today, Internet credit card payment applications are expected to assure "non-repudiation." The intent is for these systems to include identity, transaction processing, and payment technologies that preclude capricious or fraudulent after-the-fact denial of a purchase. As the stakes and complexity of electronic commerce grow, so will the legitimate reasons for repudiation of a transaction. Knowing this, we should be developing low cost and efficient problem detection, correction, and dispute resolution systems in addition to implementing highly secure fraud prevention techniques. I believe the key to being the world's preferred electronic marketplace will be a sophisticated and cost effective social, legal, and digital system for automated adjudication of disputes, or: Public Internet Dispute Infrastructure("PDI") -- Generally Applicable Software System Including Automated Procedures and Standardized Protocols for Asserting, Documenting, Arbitrating, and Resolving Disputes. Let us assume success of well tested holistic approaches for assuring both privacy and identity. In addition, assume the internet commerce and banking industries successfully integrate their clearing, settlement, payment, and delivery infrastructures. Furthermore, please assume that we have evolved the technology to allow networked communities of well behaved secure digital agents who interact in virtual markets using standardized dictionaries, cross referenced taxonomies of digital commercial language, and a comprehensive system of international digital contracts. Even with all of the above, in complex markets there are always going to be unanticipated legitimate causes for, innovative justifications for, and outright fraudulent methods for repudiating trades. This means, the benefits of electronic commerce for innovative business-to-business and globally ubiquitous use is dependent on the creation of a universal commercial negotiation process for resolving disputes. At the high-stakes end of the spectrum, the value of automated adjudication will be the ability to fix the risk of an open transaction so the dispute process can go forward in time without open-ended exposure. At the other extreme of the electronic commerce spectrum (small transactions), the value of automated adjudication will be minimization of the processing costs of disputes. The necessary core software technologies for building a generalized dispute negotiation system already exist. What is missing is the impetus to bring the technologies together into a viable software solution so that companies, associations, and regulatory bodies can begin to offer technology assisted resolution of electronic commerce disputes as viable and trusted service. Supply chains; local, state, and federal governments; exchanges; settlement, payment, and credit clearinghouses; and other associations have as one of their principal responsibilities the structuring of processes for expediting error handling and resolving disputes. How will they carry out these responsibilities efficiently in the world of electronic commerce? The Issue: Most of the development cost of electronic commerce applications and markets is building the error handling logic. Additionally, a major portion of the operational costs of these systems arises from the need agree upon and manage the business processes of preventing, detecting, and correcting of errors and disputes. The Problem: High risk, complex, and/or very large scale automated markets require software solutions for gathering information about and resolving disputes. However, building error handling, repudiation, and automated adjudication infrastructure software on a market-by-market basis is cost prohibitive. The Solution: Create a generalized system (PDI) that provides the common elements of error and dispute resolution for reuse and customization by developers who then build vertical market focused electronic commerce applications. This system would be designed as a universally acceptable standardized software object framework for electronic commerce exception handling. This system would include a human language protocol for communications between counterparties (directly or to automatically); methods for outsourcing exception handling to an agent (human or digital); methods for agreeing on and hiring an arbitrator; and methods to insure contingent liabilities. Technical Approach: Encode a top-down decomposition of the business processes of dispute resolution used in commercial markets. The goal would NOT be to completely determine the details of each possible exception; this would be an endlessly complex bottom-up approach. Rather, the approach will be to focus on structuring the generalizable components the dispute communication process. Examples: The high level components of a standardized dispute resolution system would include software object classes, methods, and protocols integrated into a software system for representing and managing each of the following dispute processes components Dialogue: A will defined protocol for explicit message interchange used to assert claims, resolve disputes, and discharge obligations. Parties: Representations and attestations about the principles to a transaction. Obligation: Methods for creating and recovering cryptographically signed and dated digital copies of contracts, agreements, expected behaviors, and/or representations of anticipated failure modes. History: A standard method for telling (by each party) of the sequence of events that led to the dispute. Neutralization: Details for specifying who is responsible for, and how, open-ended market risks are being (or have been) mitigated. Dispute: The difference between the history and the obligations. Protest: A demand for proper discharge of an alleged obligation. Bifurcation: A method for agreeing to split the parties or components of a dispute into separate negotiations. Resolution: The proposed or desired action in order to discharge the obligation. Logging: Auditing and management accounting of the exception processing flow. In summary, the goal of my opening comments today has been to introduce you to the need for further research and development into a field I am calling "Automated Adjudication," or PDI. Standardizing and simplifying the common and generalized aspects of the dispute resolution process will serve as a catalyst to speed the adoption of Internet based electronic commerce, and in so doing, improve the quality and efficiency of our economy. I am looking forward to your comments, and can be reached via e-Mail at this address: dwight_koop@sac.net. <Return to Workshop Presentations Online > |