Working from the initial framing for the project and our experience in Phase 1, we see opportunities for substantial refocusing for Phase 2. Our partners are each in distinct circumstances with different needs and opportunities related to their immediate tasks and demand for training opportunities. Distance education modules alone are not sufficient. Local benefits from use of indicators require closer integration with devising, choosing, implementing, and monitoring particular urban development strategies. Indicators focused on cross sectional snapshots are useful in allocation formulas for funding; they are of less use in figuring out what to do and whether it is working. We, therefore, suggest that Phase 2 funding focus on 1) close collaborative enhancement of tools and capabilities in use rather than on creation of fixed training modules, and 2) using indicators to assess infrastructure services as evolving capabilities during urbanization.

Our partners in Phase 1 were in very different stages of implementation of GIS and of use of urban indicators. Capetown and Kathmandu have well-established databases, which they have begun to use for planning and management of development. They continue to have difficulty in attracting and keeping high quality GIS staff, but a few specialized modules of distance based GIS training may have little impact.

Distance based modules are insufficient and seen as less credible in credentialing and enhancing careers or employee capabilities. Many providers of GIS training already exist, including services with highly developed on location and distance learning modules. Organizations that provide such training expend significant sums creating and maintaining course materials and we cannot match their economies of scale. Our partners, however, lack funding to provide these educational materials to their staff, including web access for distance education materials.

Funds might better focus on training placements and task specific "over-the-shoulder" consulting, rather than development of course materials. The mix of UCGIS project teams will be more effective by working as partners on specific tasks with local colleagues. Skills and knowledge would be transferred by mentoring and mutual learning rather than by training materials. Such opportunities would complement the already available training opportunities, which partners could be funded to attend.

We need to move beyond indicators as static measures. Even if we measure current levels of services, and could measure whether these improve, we cannot devise strategies for achieving improvements without considering how situation change. We need to consider the dynamics rather than comparative statics. In considering dynamics, indicators will have to take into account changes in—and therefore attributes of—institutional technology and physical technology, supply and its costs, demand and its price and ability to pay, and realized behaviors.

Serving an informal settlement with water is likely to involve combinations of household connections, shared taps, and vendors. Similarly an area of new urban development is likely to occur on a landscape that already includes rural and village systems (institutions and physical systems) of water supply, waste disposal, and transportation access. These existing systems will persist, perhaps for a long time, while new systems are created, sustained, and achieve complete coverage. In Kathmandu, for example, even after several attempts over more than 25 years, there is still no functioning sewage treatment plant, much less general coverage for the urban area. The actual system in place is a combination of legacy systems and failed components of newly introduced technology. It will be more useful to formulate scenarios of change from existing circumstances rather than to imagine, completely transformed future states.

The water supply system could be described as a piping network, but much of the network carries no water. Further, distinct populations in the same spatial neighborhoods rely on very different modes of services for water supply, sanitary waste, and travel. Thus spatial disaggregation alone is insufficient. Disaggregation by population groups and change over time is essential. Changes are likely to occur in stages. Even if a piping system is installed, some households will not be able to afford or establish legal status to connect to it. Cultural norms may determine that different groups access services in distinct ways. We cannot simply assume that supply is or will be used if available, much less used by everyone with spatial accessibility.

Our partners in Kathmandu are interested in transportation questions. We might look at describing accessibility changes for various population groups based on scenarios of changing vehicle types, pricing, public sector supply, and road capacity networks. Or, we might look at scenarios of water supply for informal settlements based on strategies for regulating such settlements. Or, we might assess strategies for sanitary waste disposal in newly urbanizing areas given realistic expectations based on the record of sustaining externally designed sewage collection and treatment systems.

Tools should be developed to assess changes in the capabilities of impoverished persons to access water, sanitary waste disposal, solid waste disposal, and transportation (or communication) access to jobs, schools, health and other services. These tools will help to devise, choose, implement, and monitor urban development strategies that may reduce poverty by increasing the capabilities of the impoverished.

The following suggests possibilities for a continuation proposal. We identify our primary interests and what we see as opportunities, rather than a scope for the entire UCGIS effort. We assume that an overall proposal would be prepared by UCGIS and then subcontracted as in the first round.

The three objectives of a continuation grant would be:

1. Provide funding so that partners can take advantage of existing training opportunities, including online and programs in country and in the US. For example, the University of Illinois already runs a summer training program focusing on information technology for urban development professionals in developing countries. Participants must have a funding source to be able to afford to attend. NGOs and universities in Nepal run GIS training programs aimed at agency personnel; we have participated as instructors in these programs. Funds to acquire up to date software and hardware for these training programs rely almost entirely on external funding of attendance fees.

2. Support over the shoulder collaboration of staff and US partners to work on specific tasks and problems that arise in trying to put GIS to work for human development. This would include remote interaction through email and internet and onsite interaction by bringing people to the US and sending people to partner sites for short periods of focused work.

3. Focus on developing analyses and tools to devise and assess strategies for improving 1) accessibility and 2) water supply and its relationships to drainage. We will have to negotiate this focus with partners so that we can sustain both a focus and relevance to partners' needs. Such focus is essential, however, to developing sufficient expertise to contribute significantly to the work of partner agencies.