The Web has developed as an enormous information resource. Especially in the last few years with the thriving of the socalled web2.0 services allowing user-generated content easily to be entered in -and shared by- typical web 2.0 databases. Bookmarking systems enable users to label their resources with tags, which in aggregation give rise to dynamic categorization schemes, i.e. folksonomies. Web querying through folksonomies presents an interesting potential in contrast to traditional search engines, such as the suggestion of relevant topics that may refine or even define the original search term. Both approaches are in wide use, each appreciated for their own qualities. However, it is possible for the methods to be used complementary, making use of the specific advantages of each of them. In this paper we specifically introduce the notions of tags and folksonomies and we present a method of using these notions in a tag based search technique. We discuss and elaborate results on the basis of preliminary experiments with our tag-based search engine.