We have described some of the problems we are facing in constructing a repository for a department in our Laboratory. The nature of our environment forces us to use natural languages to describe and/or specify components. We see the domain analysis as the important step in the construction of an effective reuse repository. The process to arrive at a repository will be an adaptation of the process described in [5]. The process for domain analysis is still being developed; our starting point was an approach presented in [6].
We have just started a case study and we have a long way to go before software reuse is institutionalised software reuse.