Our investigation showed broad and enthusiastic support for a single point of contact for G2G work in New York State as a "better way of working." However, the demonstrated complexity of implementing this concept suggests that incremental and modular approaches make the most sense for future development. We believe the following ideas represent the best near term opportunities for moving in this direction.

The first and easiest opportunity is to create and maintain organized access to Web resources relevant to state and local professionals. The Resources section of the Prototype represents a good start. Users appreciated the opportunity to have many resources -- including NYS and federal government information, professional organizations, legal resources, and data resources -- categorized and summarized for them. These resources were easy to access, relevant, and timely. Users liked the selection and arrangement of topics and made useful suggestions for description, searching, and customization. This kind of resource does not need special security and could be associated with the New York State Home Page in the absence of a specific G2G portal.

The Gateway Prototype demonstrated how multiple job functions associated with a particular professional role could be brought together in one Web interface accessible by a single sign-on process. Although the Prototype incorporated only a few applications, the logic and appeal of this approach was evident to the participants, who could readily envision how all their own business applications could be brought together in a single interface. The next logical extension of this idea is not to try to develop a G2G Gateway that encompasses every possible function, but to begin by co-locating access to programmatically related functions through a single Web interface. New York's CentraPort project is already doing this for county-level access to social services programs. Doing the same for all town clerk functions, all real property management functions, all public health functions, etc., promises an incremental progresssion toward a substantial G2G enterprise, with each increment bringing its own benefits to a sizable policy domain. After related applications are co-located, careful consideration of more integrated applications can follow.

The application which generated the most excitement and unanimous desire to be made fully functional was the Contact Directory and its associated Contact Repository Application. The need for such a resource is pressing and evident: no single, authentic source of contact information exists, yet every state agency and local government needs this information to do its work. The development of this resource would have several benefits: (1) it would replace an uncoordinated, duplicative, expensive, and largely manual workload with a streamlined and standardized electronic resource, (2) it would provide the opportunity to test data ownership and stewardship rules across the entire G2G spectrum, and (3) it would provide a realistically limited opportunity to test single sign-on and role-based identity. Access to the data management application would need to be controlled, but access to the directory information could be provided without user authentication. Note, however, that this application represented a high degree of data integration and therefore poses much more significant challenges in terms of difficulty, complexity, and cost than either of our other recommendations.