Modeling and Managing Mobile Commerce Spaces using RESTful Data Services. S. McFaddin 1 , D. Coffman 1 , J.H. Han 2 , H.K. Jang 2 , J.H. Kim 2 , J.K Lee 2 , M.C. Lee 2 , Y.S. Moon 2 , C. Narayanaswami 1 , Y.S. Paik 2 , J.W. Park 2 , D. Soroker 1 1 IBM T.J. Watson Research Center, Hawthorne, NY 2 IBM Ubiquitous Computing Laboratory, Seoul, Korea mcfaddin@us.ibm.com Abstract This paper advocates the use of RESTful data services in mobile commerce information spaces. REST is a restriction of web services to a simple protocol centered on a limited set of access operations against data resources. This paper analyzes how REST may be used as a general design principle for mobile commerce spaces. This analysis is threefold. First, it provides an abstract component model that contains a minimal embodiment of the REST functionality needed to support mobile commerce. Second, it describes and analyzes the operation of a mobile commerce space, resulting in a reference REST-based data decomposition that may be reused across a variety of domains. Third, it provides and analyzes a set of usage cases in the mobile commerce domain, which can form the basis of future REST-oriented programming models and tools. This analysis is based on a substantial mobile commerce framework we have implemented, called Celadon. This framework is currently being used to demonstrate several ambitious mobile commerce applications. 1. Introduction Mobile computing envisions a world in which humans carry devices with them throughout their daily life and use those devices to perform key functions relating to their location and context. The first part of the vision is well underway, and has been driven by successful applications such as voice, messaging and e-mail. The second part of the vision will depend upon the success of “localized mobile commerce” applications, which include location and context sensitive functions such as shopping, payment and customized advertisement [1]. The success of mobile commerce applications will hinge upon the ability of developers to address a fundamental requirement: they must build and deliver applications which are capable of matching the needs of mobile users with the various local environments they interact with. This has not been a requirement of voice, messaging and e-mail, which are generally based on fixed interfaces against well known service points: though the device may move around, the services do not. In contrast, mobile commerce requires devices and local service environments to quickly recognize each other, do some business, and then say goodbye. This fundamental requirement presents a major challenge in managing complexity, since a single mobile commerce enabled device could travel through dozens of local environments within a single day of usage. Each environment could offer dozens of different services, each bearing dozens of object types, each having unique and generally dissimilar interfaces. This compounded complexity offers a challenge in the modeling, implementation and management of both mobile devices and the environments they interact with. This paper advocates a simplified approach to managing mobile commerce spaces based upon the design principles of “Representational State Transfer”, commonly called “REST” [3,4,13]. REST models an information space as a reduced set of simple and common operations against a well understood set of resources. REST is a restriction of the more general remote procedure call (RPC) design style. With RPC, an information space is analyzed into an arbitrary collection of object types, with each object type exposing customized (and possibly dissimilar) interfaces. The core weight of the design is carried by the interface analysis. In contrast, REST requires the same interface to be supported by each object type, throwing the weight of the design on the object-type decomposition itself. The contributions of this paper are as follows. First, the paper makes the REST concept specific by providing a domain independent abstract component model for data oriented REST-based services. This explains concretely what should be expected of REST and provides an analysis of the core and irreducible functionality which we feel is required of any component model implementation placed in service of the mobile commerce domain. Second, the paper describes our methodology for deriving REST-based data decompositions for the mobile