Cornell University, HCI Group 301 College Avenue Ithaca, NY USA 14850 +1 607 255 7826
Michael J. Muller
IBM Research One Rogers Street Cambridge, MA USA 02142 +1 617 693 4235
Social tagging systems allow users to share resources categorized according to community-generated tags. These systems serve to organize personal information, provide opportunities for users to express their identities and to allow for social information seeking. In this paper, we examine the motivations behind tag selection, specifically focusing on the social aspects of choosing tags for an audience. We describe initial results from a qualitative study of users of tagging systems deployed within a large enterprise. Exploratory coding suggests that users remain cognizant that their tags play a social role and that users’ tag selection strategies are managed with respect to this awareness.
encompasses both organization of resources and communication, while sociality describes tagging for oneself versus other recipients (i.e. viewers of the content). Tag typology has also been studied with respect to whether or not tags are helpful in completing a task  or whether they fulfill a specific function such as information retrieval . It is not yet clear from a priori categorization of tags what motivations exist for selecting specific tags. In this poster, we focus primarily on the social motivations behind the tag-selection process, specifically the awareness of one’s audience. We then describe initial findings from a qualitative study of users of tagging systems that are deployed within a large enterprise.
Categories and Subject Descriptors
H.5.3 [Information Interfaces and Presentation]: Group and Organization Interfaces
We are conducting a field study in which we observe members of two communities of practice within an enterprise: project managers and user experience designers. We focus on these two specific communities to more closely investigate tagging within teams and within communities of practice. Semi-structured interviews are being conducted with a sample of 40 individuals (20 from each community), stratified by tagging activity as determined by log analysis of the Dogear internal social bookmarking website . Interviews include a series of open-ended questions regarding informants’ roles in the organization, the communities with which they identify and general discussion regarding their tagging activities, concluding with specific probes in which we ask informants to discuss specific tags that they have created in different tagging systems. Coding of the interviews is influenced by grounded theory (e.g. ): there is no a priori hypothesis; theory emerges from the data; and there are several rounds of interviewing in which the questions for later round are influenced by the themes emerging from analysis of the preceding rounds.
Tagging, social software, community, audience
Social tagging systems have emerged as a compelling technological framework in which to study the development of distributed communities [4, 6, 7, 9, 10, 11]. One common approach, social bookmarking, allows users to categorize shared resources using tags . Social bookmarking systems inject sociality into information seeking; that is, users search for resources via community browsing, such as looking for bookmarks according to popularity or by who’s creating the tag . Tagging motivation within these social systems has been characterized in a largely dichotomous manner. Marlow et al.  employ the categories of organizational vs. social, where organizational motivations help users seek individual benefit for re-finding personal resources. Social motivations are expressive of the tagger’s identity and therefore...