Return to ENVRI Community Home
To clarify the terminology in these pages, the box below describes the conditions and implications for each of the three different case types: science, test, and implementation.
Scientific use case
For example, the picture depicts the relationships between use/test/implementation cases with the tasks involved in ENVRIplus work packages. ENVRIplus development areas (WP6-8) would be based on Implementation cases identified together with RIs. WP9 will select test cases for deployment. The Test Cases concern software validation, software integration (how to integrate services into different e-Infrastructures), assessment of available resources, and release management. It involves of negotiation of required test-bed resources at different development levels and capability levels, deployment of the services and implementation cases, operational level managements. The deployed test cases shall be able to support valid and realistic scientific use cases that would benefit ENVRIplus RIs and users.
The driver in the process to define appropriate cases should be the implementation of ENVRI results in the relevant RIs. An action plan should start with describing the desired implementations in these RIs and successively to select some test cases that will assist in learning best practices for full implementation. It would be nice to have these embedded in one or more science cases with scientific users. This leads to the following proposed sequence of steps in an action plan.
Step 1: Description of Implementation cases. Theme 2 (WP5) started to work with the RIs in describing their portfolio of required new and/or enhanced services, resulting in the Implementation cases. This is essentially the “translation” of the WP/Task description to the needs of each RI. The current WP5 requirement collection phase is addressing explicitly each of the Theme -2 topics with questionnaires based on example relevant use case scenarios (within RI and between RIs). This will result in implementation plans for each topic and for each interested RI.
Step 2: Selection of the Test cases. These are meant to learn how to best prepare for the implementation. The most interesting Test cases are the ones that are related to more than a single WP or Theme and may cover some implementation cases. Test cases should be small and time limited, concrete but not necessarily have immediate nice stories (science case) from domain scientists. The step 1 work will help in a) the concrete definition of some test cases, associated to a few RI services, b) the identification of the current gaps in technologies, and c) action plans for delivery of results and “lessons learned’ at short time.
Step 3: Identification of scientific use cases. In an ideal world, the Test cases are embedded in one or more scientific use cases, research projects in need of new developments and services that ENVRIPLUS may provide. The RI domain leaders may suggest cross-cutting use cases serving as selling stories. However, to get the cooperation of a research project it is required to fund a postdoc willing to act as a guinea pig. Such funding is not likely, and ENVRIPLUS has probably to refrain from scientific use cases and not execute step 3. A more modest approach is to integrate selected test cases and find potentially interested scientists.