This paper presents the result of a practical study comparing how domain expertise readers tag documents with different descriptions. We conducted an experiment comparing tagging behavior of experts and novices groups and intended to discover which group would generate more reliable and more representative sets of tags when they were asked to provide tags for document bookmarks in a Mozilla Firefox browser. In this analysis, we defined two measures, similarity and relevance, to describe the differences between the two groups. Result from this experiment showed that tags chosen by experts yielded better similarity and relevance in both analyses. Tags chosen by the expert group had higher commonality in pairwise similarity analysis; further, the relevance analysis showed that tags chosen by experts reflected better understanding of the content. Tagging behavior has become highly popular on the web; implications on the design of future social information system are discussed.