Abstract
To be able to use or reuse an open source software, we must be aware of its existence and find it first. Existing search engines on the Web do not allow finding open source software satisfying given requirements while taking into account special semantics. In this paper, we propose a semantic search system for open source software allowing to identify and localize them. It relies on an ontology providing a mechanism to describe knowledge about open source software. Moreover, it can infer knowledge for semantic identification through some defined rules. Preliminary results are encouraging. The evaluation results compared to a non-semantic version of the same search engine confirm the added value of the consideration of the semantic aspect in search.