Systematic reuse of large-scale software components promises rapid, low cost development of high-quality software through the straightforward integration of existing software assets. To date this promise remains largely unrealized, owing to technical, managerial, cultural, and legal barriers. One important technical barrier is architectural mismatch. Recently, several component integration architectures have been developed that purport to promote large-scale reuse. Microsoft's OLE technology and associated applications are representative of this trend. To understand the potential of these architectures to enable large-scale reuse, we evaluated OLE by using it to develop a novel fault-tree analysis tool. Although difficulties remain, the approach appears to overcome architectural impediments that have hindered some previous large-scale reuse attempts, to be practical for use in many domains, and to represent significant progress towards realizing the promise of large-scale systematic reuse.