In blueprint planning, learning is accomplished largely through surveys and pilot projects. The former have important drawbacks. Although microcomputers promise speedier data processing, surveys remain complex and time-consuming to organize. The large amount of data they generate is often turned into usable information too late for practical planning. Even when they are timely, large-scale surveys are better suited to quantitative analysis of people,s assets and production, for example, than to the qualitative assessment of social relations and attitudes required for programme design.
Pilot projects, learning by doing, but on an initially small scale, anticipate implementation problems in a way surveys cannot. Provided they are done by the agencies that will later be responsible for operations, and use procedures that are replicable on a large scale, they can greatly increase responsiveness to local needs and bureaucratic realities. Pilot projects make sense particularly for agencies (including donors) required to justify future spending with firm and tested proposals.
They do not solve all the problems of running big programmes, however. These programmes need to be adjusted over time in response to varying local needs and environmental conditions. More recent planning methods have therefore moved toward a continuous learning and design process during implementation. Rather than design one initial replicable plan, the aim is to develop a flexible system for producing local plans in conjunction with local people. Managing the link between-bureaucracy and people therefore becomes centrally important.
For governments, greater participation of local communities has both drawbacks and advantages. On the one hand, it adds to the management burden of government at the project design stage. Also, where communities are highly stratified, participation may bring to the surface tensions between local interest groups: local elites, which dominate "community" organizations, may oppose the involvement of disadvantaged groups and take over programmes to serve their own ends. In addition, field staff may not be enthusiastic if working for local communities is thought to incur loss of authority and status, as well as inconvenience.
On the other hand, some form of partnership with local communities in the design of programmes is essential, if plans are to meet beneficiaries, needs and encourage contributions of money and labour. Without such contributions, programmes may never be viable or, once started, may not be sustained. The widespread success of "sites and services", schemes during the 1970s indicates that communities are prepared to shoulder much of the financial and managerial burden of development if design technologies and levels of service are appropriate and affordable.