The EGAD (Electronic Grant Application Development) project is reaching a stage in its implementation where it is important to have some understanding of how information will flow. The information and process flow diagrams done in earlier fiscal years have been taken up by specialized contractors who are charged with completing and refining the definition of the information flow and implementation processes. The institutionalization of the electronic grant application process means that ideas and issues which were only conjecture or 'research' several years ago are being turned into standard authoritative approaches. Contractors are good at working out a task which has been defined for them but they normally do not have the authority to explore all possibilities. Over the years we have found that we constantly tread the boundary between what the system will allow us to think or do and the areas which we feel we must explore if the NIH is to be a viable institution in the future. We have been looking at new computer network tools for possible application in the grant review process. The Gopher network tool emerging from the University of Minnesota provides the possibility of accessing any database in its style anywhere on the InterNet. We have been thinking about how Gopher could be used to access material not included in a grant application but which nevertheless provides backup or support for an application. The WAIS (Wide Area Information System) emerging from Thinking Systems Inc., provides a network mechanism for searching through many loosely associated files for specific items of information. We have been experimenting with the possibility of using WAIS to scan through collections of applications to find new patterns of research ideas. Instead of being lodged on a central computer, these loosely associated files could be kept on the workstations of Scientific Research Administrators in the DRG. WAIS also provides an alerting function in the sense that once a question is asked by a grant administrator, it can be automatically re-asked periodically by the network system. Now that universal network connectivity has been achieved, we have been trying to understand what new methods of use for scientific and administrative work might be possible.
