The Semantic Web mission is to enable a better organization of the Web content to improve the searching, navigation and integration of the available information. Although the Semantic Web is intended for machines, the process of creating and maintaining it is a social one: only people, for example, have necessary skills to create and maintain ontologies. While most existing ontologies are designed by single individuals or small groups of experts, actual ontology users are not involved in the development process. Such an individual approach in creating ontologies, lead to a weak community grounding. On the other hand, Social Software is becoming increasingly popular among web users, giving opportunities to exploit the potential of collaboration within a community. Tools like wikis and folksonomies allow users to easily create new content and share contributions over a social network. Social Software tools can go beyond their current limits, by exploiting the power provided by semantic technologies. Conversely, Semantic Web tools can benefit from the ability of Social Software in fostering collaboration among users, by lowering entry barriers. In this paper we propose a new approach for ontology evolution, considering collaborative tagging systems as an opportunity to complement classic approaches used in maintaining ontologies.