A. Dieberger: Where did all the people go? A collaborative Web space with social navigation information

Where did all the people go? A collaborative Web space with social navigation information.

Andreas Dieberger
IBM Almaden Research Center
andreasd@us.ibm.com

A CoWeb is a collaborative Web space that allows people to modify Web content and create new pages in an easy fashion. Originally developed by Mark Guzdial as a tool for anchored discussions in an educational setting it is a flexible discussion space. Yet, as with most of the Web there is little support for showing what users did in this space, including where they have been etc.

We extended the CoWeb to provide several forms of social navigation information. Footprint markers next to links indicate the amount of traffic on pages they link to and newness markers indicate when pages were last modified. In addition, pages that have not been visited for a longer time show "lack of traffic" markers. These link markers are just one of several possible ways to convey interaction history information. Other possibilities include modifying link colors, background color and so on. Currently we are experimenting with sound cues to convey the information contained in the link markers.

In the present system social navigation information is available also as lists of recently modified and recently accessed pages, through access to the log file and through a page that indicates which pages are currently being modifed.

Our rationale for providing this information is that social information is ubiquitous in the physical world, be it in the form of footprints, loud party noise coming from an adjacent room, or verbal recommendations from friends and colleagues. These social cues may help us to make decisions, and to navigate the environment we live in [1]. Social navigation information does not always change our behavior, but it always increases our awareness of other people's activities. The importance of such awareness information is underscored by the growing interest in awareness support systems [2]. Yet, the Web is a poor conduit for this type of information.

There are two main types of social navigation. Direct social navigation takes the form of recommendations or of guiding people. It thus tends to be a one-on-one interaction. In contrast, indirect social navigation focuses on aggregated history information. An example of such aggregated information is read wear [3]. Read wear indicates the amount of traffic on an information object by simulating "wear" on the object. It scales well, typically avoids privacy issues and can be a useful tool to provide awareness of the overall activity in a community. As our goal was to increase awareness of interactions with the information space, the social navigation CoWeb focuses on indirect social navigation.

Using a CoWeb is simple. People use it like any other Web site, except that pages contain a button to "modify this page". Clicking this button presents a Web form with the current content of the page. People can contribute either plain text or HTML. Creation of links and new pages is simple as well: Enclose some text with asterisks and that text turns into a link to a new page with that text as page name. Similarly, links are created by enclosing an existing page name or any URL with asterisks.

The social navigation markers on the CoWeb instantly reflect modifications and aggregated accesses to pages. People describe the social navigation CoWeb as being more "alive" than ordinary CoWebs, which shows that the markers increase people's awareness of each other's activities and therefore their perception of the CoWeb as a social "place".

Like in standard CoWebs there are no fixed roles such as author, editor, or reader. Everybody can modify every page. The CoWeb keeps earlier versions of pages available so that a damaged page can be repaired easily. This flexibility also allows people to easily remove inapproprate content, or move off-topic discussions to new pages. Edwards points out in [4] that roles in collaboration often are very fluid. The lack of strict roles and strict permissions in the CoWeb allows such a fluid role change: I can fix a minor spelling mistake on somebody else's page when I happen to notice it or add a short comment -- depending on the social conventions in the system. Note: There are CoWebs with stricter types of access control.

In the current architecture of the social navigation CoWeb history information is kept inside the server. We soon discovered that keeping history information separately is a better approach. For example using intermediaries [5] social navigation information could be collected just for one user or for groups of users, thus providing much more specific interaction history.

Social navigation is not a system or tool, but rather a novel way of thinking of the design of collaborative information systems [1, 6, 7]. It presents a number of interesting research questions, like whether every user should cause interaction history and to what extent, how to aggreggate history information for information spaces with different "rythms" and how to cope with changing interests.

Acknowledgements

We thank Mark Guzdial for creating the original CoWeb (based on earlier work by Ward Cunningham) and Peter Lönnqvist for his assistance with the usability study. We also thank Nils Dahlbeck, Kia Höök, Colleen Kehoe, Lex Spoon, and Annika Waern for valuable discussions about social navigation, and Teenie Matlock for editing.

Appendix

This work was done while the author worked at Emory University and while visiting at the Swedish Institute for Computer Science. Several CoWebs are accessible through http://swiki.sics.se/

References

  1. Dieberger, A., Supporting Social Navigation on the World Wide Web. International Journal of Human-Computer Studies, 1997. 46: p. 805-825.
  2. Gutwin, C. and S. Greenberg, Effects of Awareness Support on Groupware Usability, in Proc. CHI'98, ACM Press: Los Angeles, CA. p. 511-518.
  3. Hill, W.C. and J.D. Hollan, Edit Wear and Read Wear, in CHI'92, ACM Press: Monterey. p. 3-9.
  4. Edwards, K., Policies and Roles in Collaborative Applications, in Proc. of CSCW'96, Boston. p. 11-20.
  5. Barrett, R. and P.P. Maglio. Intermediaries: New Places for Producing and Manipulating Web content, in WWW7.
  6. A.J. Munro, K. Höök, and D. Benyon, Editors: Social Navigation of Information Space, 1999, Springer: London.
  7. Wexelblat, A., Footprints: Interaction History for Digital Objects, in MIT Media Lab. 1999, MIT: Cambridge, MA.