Same topic in earlier issue
Issue
Previous article
Article
SIGCHI Bulletin
Vol.29 No.1, January 1997
Next article
Article
No later issue with same topic
Issue

"Finding and Reminding" Revisited: Appropriate Metaphors for File Organization at the Desktop

Bonnie Nardi and Deborah Barreau

"[With writing], you have invented not an elixir of memory, but of reminding..." (Phaedrus in Plato)

Introduction

In the January 1996 SIGCHI Bulletin, Scott Fertig, Eric Freeman, and David Gelernter offered an interesting challenge to our 1995 paper on the information organization and retrieval behaviors of individuals in the work place. While they agreed with our categorization of information use in working environments as ephemeral, working, and archived, they took issue with our findings concerning the importance of location-based searching over logical retrieval, with our suggestion that the user's perception of their information space and the location of information within that space serves a reminding function, and with the evidence given that archival information has limited value to users in their work. At the core of their criticisms is the charge that our studies were constrained by the narrow scope of the desktop metaphor which "favors certain types of interaction over others." They suggest that users only need better tools to find the value in archiving.

While we applaud their call for better tools and recognize that archived information is critical to researchers, librarians and a few other professions, our studies suggest that the major information management problem facing most individuals at work today is the volume and variety of ephemeral information. Like Phaedrus, we wish to call the reader's attention to the crucial difference between memory and reminding. While Phaedrus was concerned that writing would make people forgetful and viewed the reminding function of writing as degenerate, in this literate age, we are fully accustomed to using inscriptions to remind us of things. (And indeed how would we know what Plato thought if he hadn't written it down?) We tend to think of an electronic file system as a repository of stored information -- and indeed it is -- but as our research shows, it is also a space upson which people inscribe things they wish to be reminded of. The use of spatial cues to manage this difficult but important reminding process extends from the physical world of the paper office to the virtual space of the electronic desktop.

User Preference for Location-Based Search

One of the problems facing organizations as they become more information intensive is that workers experience a sense of a loss of control over work and feel a greater burden from having more data to consider before making decisions (Zuboff, 1988). Carroll and Rosson (1987) describe a similar situation in their observations of workers who are reluctant to experiment or try new software features because of the investment of time required -- time they need to spend on attending to the work at hand.

Our observations suggest that workers have spatial perceptions about their work space and about how information is organized within it. They want to be able to get what they need quickly and they do not want to manage or maintain more information that they need to do their jobs. Our research has demonstrated that workers have the capacity to manage vast amounts of their "working information," the information they use routinely to carry out their work, in this manner. They have a sense of control over the environment and confidence that they can find what they need with minimal effort.

Location-based searching is not one-dimensional as Fertig and his colleagues suggest. It is often based upon association with documents of similar type, on a similar topic, created about the same time, or in a similar stage of completion being located at the same place. For individuals engaged in a variety of projects and activities, the location-based storage and retrieval strategy reduces the overhead for what they have to remember to successfully organize and retrieve a document.

Logical retrieval based upon richer categories may require more work for the user than the location-based find. Users have demonstrated a reluctance to engage in complex search strategies to retrieve their own documents. While the size of our study was limited, this hypothesis is worth further examination and testing before abandoning the desktop metaphor for a new approach.

Location of Files Serves a Critical Reminding Function

Our studies found that users often placed files where they could serve a reminding function. This included placing icons where they would be noticed and leaving messages in electronic mailboxes where they would serve as reminders of things to be done. Fertig et al. correctly point out that location-based reminders do not insure that one will actually be reminded of anything, and they suggest there is insufficient screen real estate to handle the volume and variety of reminders needed. While this may be true, there is evidence that location-based reminding has meaning for us -- we place things in obvious places to avoid having to try to remember them later, e.g., we put a report into the briefcase as a reminder to review it, attach a dental appointment card by magnet to the refrigerator door and place handouts for a meeting on a corner of the desk so we will be reminded by them when we depart. The placement of these items does not insure we will be reminded, but it does dramatically increase the probability of our remembering.

We are less likely to overlook these locational reminders, when they are part of a normal routine, and in our study we found secretaries checking a designated area of their desktop on a daily basis, and other similar routinized behaviors that made the reminding function of icon placement more effective. The current crop of desktop tools is not rich enough, but nonetheless, there is strong evidence that users make use of location as a reminding tool, even using their electronic mailboxes and their desktops for this purpose, though they were not designed to be used this way.

Our research suggests that better tools for organizing ephemeral information are very much needed. Because of the time-sensitive nature of this information, reminders are needed, and in the absence of other tools, users make use of location to help them remember the things they need to deal with. Since users tend to make use of locational reminders, designers of tools for managing ephemeral information may choose to take advantage of this fact.

Users Avoid Elaborate Filing Schemes and Archive Relatively Few Documents

Our studies focused on users in working environments. Unlike researchers who have reason to archive and use large collections of documents, we found that archiving was much less important for the users in our studies. Users may rely upon libraries or individuals who serve as information gatekeepers, boundary-spanners, or gardeners (see Nardi, 1993) to provide both the historical perspective and information about the newest technologies, developments, and changes that impact their work.

Organizations are dynamic institutions, with structures, roles, and scope of work changing with new markets, technologies, regulations, and initiatives. It is significant that corporate memory is constrained by the individuals who comprise the organization. While information may be collected and stored indefinitely, its utility will be limited by the memories of those in the organization who can supply the related context, history, and consequences implied by the information, or by management's willingness to invest the required resources for the research effort. Some users reported they would recreate a memorandum, paper, or study rather than store, locate, and edit an older version for a new purpose. Others reported initially developing elaborate filing schemes for documents, but abandoning them because of the amount of overhead required to maintain them. Organizing documents according to such rules amounts to maintaining a parallel system, one for the organizational scheme itself, the equivalent of a thesaurus or classification system, and the other for the documents that fit into it.

There are situations when old information is essential, and users must locate the information quickly. One cannot equate "old" information with "archival" information, however, as old information that users may require to do their work will likely be kept with their working information, even if used infrequently. Users also depend upon the memories and expertise of others, and upon organizational repositories, libraries, and data banks when archival information is required. Possibly the kind of system proposed by Fertig et al. could help with these issues of "institutional memory" but there's a whole set of challenges involving the collaborative management of information not addressed in their system which is oriented to an individual user, on the "diary" model.

For most people old information is not, in general, useful information. Big schemes to archive it are just not worth it. The average user has too much to do in the present, too many pressing needs with ephemeral and working information. They don't wonder what happened five years ago; too much is going on in the present. Old marketing data, last month's requests from the boss, lists of students' grades from years ago, an announcement about the last meditation group meeting are just not useful. It's not that people are prevented from keeping old information around -- they just don't need it! You have to ask yourself: What would someone do with all that old information even if they could find it quickly and easily?

Summary

The alternatives offered by many developers of personal information management systems seem to view documents in the work space as a collection that can be easily characterized, ordered, and retrieved based upon common characteristics, or based upon full text retrieval. These approaches ignore the complexity and variety of information in personal electronic environments. Documents in the electronic work space may be created by someone other than the user, employing different vocabularies and document characteristics, may include user-defined macros, executable programs, spreadsheets, graphics, signature files, and software as well as personal documents. While location-based searching has limitations, schemes that automatically characterize information may not provide enough flexibility to consider the richness of these environments, and schemes that allow characterization for visual retrieval may not easily accommodate all of the desired dimensions. All of the approaches deserve further study.

We observed real people performing work in real environments. Undoubtedly, they were constrained by their computer systems and the desktop metaphor, yet we found users exercising creativity and variety in the use of these systems to perform their work, extending the metaphor to accommodate their needs. We must also consider that Cole (1982) found a similar office organization ten years ago in the paper office. The burden of proof is on researchers to say why having a lot of archived information will become useful to people when it never has been in the past. If the dramatic move from paper to electronic information doesn't obliterate location-based finding, then what will? Purveyors of logic-based systems need to tell a good story as to what's changed in the new world that will lead to a preference for logic-based systems. So far we haven't heard that story. If it only involves new tools, but does not establish a true user need and speak to the nature of the actual work people are doing and want to do, then it will not be a compelling story. In theory of course logic-based finding could become useful, but given the overall nature of work, we don't see inklings that it will. While we recognize that researchers have need for organizing and retrieving information and building archives to support their research, the results of our studies suggest that users in most work environments have different needs and priorities. A greater concern for such workers is the management of the growing volume and variety of ephemeral information that must be managed and utilized. Zippy tools that help with these pressing problems would be very much appreciated by today's workers.

References

Barreau, D.K. (1995). Context as a factor in personal information management systems. Journal of the American Society for Information Science, 46 (5), 327-339.

Barreau, D.K. and Nardi, B. (1995). Finding and reminding: File organization from the desktop. ACM SIGCHI Bulletin, 27 (3), 39-43.

Carroll, J.M. and Rosson, M.B. (1987). The paradox of the active user. In J. Carroll (Ed.), Interfacing Thought: Cognitive Aspects of Human-Computer Interaction. Cambridge MA: MIT Press.

Cole, I. (1982). Human aspects of office filing: Implications for the electronic office. Proceedings Human Factors Society, Seattle, Washington.

Fertig, S., Freeman, E. and Gelernter, D. (1996). "Finding and reminding" reconsidered. ACM SIGCHI Bulletin, 28 (1), 66-69.

Malone, T. W. (1983). How do people organize their desks? Implications for the design of office information systems. ACM Transactions on Office Information Systems, 1, 99-112.

Nardi, B., Anderson, K. and Erickson, T. (1995). Filing and finding computer files. Proceedings East-West Conference on Human Computer Interaction. Moscow, Russia. July 4-8. Pp. 162-179.

Nardi, B. (1993). A Small Matter of Programming: Perspectives on End User Programming. Cambridge, MA: MIT Press.

Zuboff, Shoshana (1988). In the Age of the Smart Machine. New York: Basic Books.

Authors' Addresses

Deborah Barreau
College of Library and Information Services
University of Maryland
4105 Hornbake
College Park, Maryland, USA
barreau@wam.umd.edu

Bonnie A. Nardi
Advanced Technology Group
Apple Computer, Inc.
1 Infinite Loop
Cupertino, California 95014, USA
nardi@apple.com

Same topic in earlier issue
Issue
Previous article
Article
SIGCHI Bulletin
Vol.29 No.1, January 1997
Next article
Article
No later issue with same topic
Issue