Folder vs. Tag Preference in Personal Information Management

Users' preference for folders vs. tags was studied in two working environments where both options were available to them. In the Gmail study we informed 75 participants about both folder-labeling and tag-labeling, observed their storage behavior after a month, and asked them to estimate the proportions of different retrieval options in their behavior. In the Windows 7 study, we informed 23 participants about tags and asked them to tag all their files for two weeks, followed by a period of 5 weeks of free choice between the two methods. Their storage and retrieval habits were tested prior to the learning session and after 7 weeks, using a designated classification recording software and a retrieval habits questionnaire. A controlled retrieval task and an in-depth interview were conducted. Results of both studies show a strong preference for folders over tags for both storage and retrieval. In the minority of cases where tags were used for storage, participants typically used a single tag per information item. Moreover, when multiple classification was used for storage, it was only marginal used for retrieval. The controlled retrieval task showed lower success rates and slower retrieval speeds for tag use. Possible reasons for participants’ preferences are discussed. Personal Information Management (PIM) is a basic human-computer behavior in which the user stores his/her information items (e.g. files, email, and Web favorites) in order to retrieve them later. Traditionally, PIM systems provided folders for information storage and retrieval; however, as a consequence of the popularity of Web 2.0, tags also defused into PIM systems. It is widely claimed that tags have two fundamental advantages over folders; tags enable multiple classification and eliminate the need for hierarchies. (a) Multiple Classification: In the folders method, an information item can only be stored in a single folder; however, the user may have a number of possible classifications related to that item (Dourish et al., 2000). For example, pictures from a conference in Copenhagen can be stored under "Pictures," "Trips," "Conferences," or "Copenhagen." As time passes, users may forget the choice they initially made, making retrieval difficult. In contrast, the tagging method enables users to apply any number of tags to their information item, and use any of these tags to retrieve it. (b) No Hierarchical Location: Folders may hide information items from view because they do not show items contained in subfolders (Malone, 1983). The tagging method consciously refutes hierarchies and locations. Instead, all information items are stored in a single repository and are retrieved via nonhierarchical means, such as tag search, tag selection or tag clouds. Which option is preferable when both are available and users are familiar with the use of both? To the best of our knowledge, this is the first research to test this question. To do so, we conducted two studies: In one, the Gmail study, we first introduced 75 participants to the folder-labeling and tag-labeling options of the Gmail interface, waited a month, and then observed their mailbox to analyze their actual behavior. In the second study, the Windows 7 study, we asked 23 participants to tag all files they used for a period of two weeks, then returned after five weeks to observe, using a designated software, the amount of tagging performed on new files, and conducted in-depth interviews regarding the users’ behavior. In the Windows 7 study, we also compared the retrieval efficiency of the two methods using a controlled test. Theoretical Background