Activity-centric computing

An idea you already use

When you set aside a project on a workbench, you can leave its parts, plans, and tools together. Come back tomorrow and the things you need are already where you left them.

A client proposal has a similar setup. You might keep the draft on one screen, the cost estimate beside it, and the client brief open in a browser. Those pieces belong together because they serve the same purpose. An activity-centric desktop gives that whole setup a place you can return to.

The problem it solves

Open your browser and you’ll see pages from several projects. Open your project folder and you’ll see the files stored there. To get back to the proposal, you still have to find the right pages and documents and put the windows where you need them.

Activity-centric systems give the proposal its own place alongside your other work. The activity can hold the resources you need across applications and bring them back together when you return (Bardram et al., 2019).

Where the idea came from

The idea began with a simple observation about computer use. People leave work unfinished and come back to it. Over four decades, researchers have tried different ways to give that work a place to live.

1. 1983

   The workspace as a way back into a task

   At UC San Diego, Liam Bannon and colleagues studied command histories annotated by the people who wrote them. A single session could contain several unfinished tasks. They proposed a workspace that could keep the information needed to pick each one up again (Bannon et al., 1983).

2. 1986

   Rooms for different kinds of work

   At Xerox PARC, Austin Henderson and Stuart Card built Rooms. Windows used together could live in a virtual room, and a window needed in several rooms could appear in each of them (Henderson & Card, 1986).

3. 2000–2004

   A task becomes something you can select

   Microsoft Research’s Task Gallery put the current task’s windows on a stage and let people choose another task from a 3D gallery (Robertson et al., 2000). Scalable Fabric used a flat desktop with named groups of windows around the edges. Selecting a group brought its windows into the center (Robertson et al., 2004).

4. 2003–2005

   The activity reaches across applications

   UMEA and TaskTracer recorded the resources people used while working on a task (Kaptelinin, 2003; Dragunov et al., 2005). TaskMaster grouped email and related material by task (Bellotti et al., 2003). IBM’s ActivityExplorer organized shared material around a team’s activity (Muller et al., 2004).

5. 2006–2009

   The activity can travel with the person

   Carnegie Mellon’s earlier Project Aura had explored reassembling a task on another computer (Garlan et al., 2002). Jakob Bardram and colleagues built an Activity Bar for Windows XP that let people switch between groups of applications (Bardram et al., 2006). Related hospital research examined how clinicians could resume and share activities as they moved between locations (Bardram, 2009).

6. 2016–2021

   Dedicated desktops and task-based tabs

   Researchers tested what happens when each task gets a dedicated desktop (Jeuris & Bardram, 2016). The Tabs.do study brought the same idea into the browser by saving related tabs as a group that people could reopen together (Chang et al., 2021).

7. Today

   OpenWorkspace Workspaces

   A named Workspace records the supported windows, content, and layout you assembled for an activity. You can reopen that arrangement the next time you work on it.

What the research found

• A single project is scattered across tools

  A 2006 study found that related material was split among documents, email, and bookmarks. A project folder could hold the files while other parts of the same project lived in separate collections. (Bergman et al., 2006)

• Returning often follows several switches

  Field observations found that people moved between related sets of work throughout the day. Computer logs from a separate study showed how one interruption could lead to several intervening activities before people returned to what they were doing. (González & Mark, 2004; Iqbal & Horvitz, 2007)

• A dedicated desktop can help people resume

  In a 16-person Windows 7 experiment, participants resumed tasks about 10 seconds faster on average and reported lower cognitive load when each task had a dedicated virtual desktop. (Jeuris & Bardram, 2016)

• People group work at different scales

  In a field study of Giornata, some people used activities for broad areas of work while others made them for individual projects or short-lived tasks. People also changed how they organized their activities over time. (Voida & Mynatt, 2009)

A 2019 review counted 68 distinct technologies across 101 peer-reviewed papers (Bardram et al., 2019). Researchers have tested the idea at different scales, from the code and editor views needed to fix a bug in Eclipse (Kersten & Murphy, 2006) to a group of browser tabs needed for an online task (Chang et al., 2021).

Where Workspaces fit

In a 2004 diary study, people returning to earlier work often had several documents to find again. The researchers proposed a way to save the applications, documents, and window layout for a task and recall them together (Czerwinski et al., 2004).

OpenWorkspace applies that idea to the desktop. The arrangement you already use becomes a named Workspace you can return to as the work continues.

• Compose. Arrange the windows, documents, and browser tabs you need for the activity.
• Capture. Save the participating elements and their arrangement as a named Workspace.
• Resume. Reopen the Workspace to restore supported windows, content, and their saved positions.
• Revise. Update the saved arrangement as the work changes.

OpenWorkspace also applies focus + context through FocalContextual^™. Workspaces preserve the working arrangement over time. FocalContextual gives primary and supporting windows distinct places within that arrangement.

Sources

Twenty works selected for citation influence, historical importance, and direct relevance to organizing and resuming desktop work.

1. Bannon, L., Cypher, A., Greenspan, S., & Monty, M. L. (1983). Evaluation and analysis of users’ activity organization. Proceedings of CHI 1983, 54–57.Observed interleaved computer work and proposed workspaces to help people recover their place when switching tasks.
2. Henderson, D. A., Jr., & Card, S. K. (1986). Rooms: The use of multiple virtual workspaces to reduce space contention in a window-based graphical user interface. ACM Transactions on Graphics, 5(3), 211–243.Grouped frequently used windows into virtual rooms and allowed some windows to appear in more than one room.
3. Robertson, G., van Dantzich, M., Robbins, D., Czerwinski, M., Hinckley, K., Risden, K., Thiel, D., & Gorokhovsky, V. (2000). The Task Gallery: A 3D window manager. Proceedings of CHI 2000, 494–501.Put the windows for a task on a central stage and made whole tasks recognizable and selectable in a 3D gallery.
4. Garlan, D., Siewiorek, D. P., Smailagic, A., & Steenkiste, P. (2002). Project Aura: Toward distraction-free pervasive computing. IEEE Pervasive Computing, 1(2), 22–31.Designed a system that could reassemble the services needed for a task as a person moved between computing environments.
5. Bellotti, V., Ducheneaut, N., Howard, M., & Smith, I. (2003). Taking email to task: The design and evaluation of a task management centered email tool. Proceedings of CHI 2003, 345–352.Designed and evaluated TaskMaster, an email client that grouped messages and related resources around tasks.
6. Kaptelinin, V. (2003). UMEA: Translating interaction histories into project contexts. Proceedings of CHI 2003, 353–360.Collected files, websites, and other resources from the user’s interactions while a project was active.
7. Czerwinski, M., Horvitz, E., & Wilhite, S. (2004). A diary study of task switching and interruptions. Proceedings of CHI 2004, 175–182.Used week-long diaries to examine task switching and proposed saving sets of applications, documents, and their layout.
8. González, V. M., & Mark, G. (2004). “Constant, constant, multi-tasking craziness”: Managing multiple working spheres. Proceedings of CHI 2004, 113–120.Observed information workers switching between working spheres, each spanning related tasks, tools, and people.
9. Muller, M. J., Geyer, W., Brownholtz, B., Wilcox, E., & Millen, D. R. (2004). One-hundred days in an activity-centric collaboration environment based on shared objects. Proceedings of CHI 2004, 375–382.Studied the first 100 days of a system where collaborators assembled different kinds of shared resources into activities.
10. Robertson, G., Horvitz, E., Czerwinski, M., Baudisch, P., Hutchings, D., Meyers, B., Robbins, D., & Smith, G. (2004). Scalable Fabric: Flexible task management. Proceedings of AVI 2004, 85–89.Let users name groups of windows and bring an entire group from the periphery into the focus area.
11. Dragunov, A. N., Dietterich, T. G., Johnsrude, K., McLaughlin, M., Li, L., & Herlocker, J. L. (2005). TaskTracer: A desktop environment to support multi-tasking knowledge workers. Proceedings of IUI 2005, 75–82.Associated desktop actions and resources with tasks so that users could retrieve the relevant materials later.
12. Bardram, J. E., Bunde-Pedersen, J., & Søgaard, M. (2006). Support for activity-based computing in a personal computing operating system. Proceedings of CHI 2006, 211–220.Extended Windows XP with an Activity Bar for grouping applications and resources and switching between activities.
13. Bergman, O., Beyth-Marom, R., & Nachmias, R. (2006). The project fragmentation problem in personal information management. Proceedings of CHI 2006, 271–274.Documented how information belonging to one project ends up in separate file, email, and bookmark collections.
14. Kersten, M., & Murphy, G. C. (2006). Using task context to improve programmer productivity. Proceedings of SIGSOFT FSE 2006, 1–11.Captured the code relevant to each task in Eclipse and evaluated how task-specific views affected programmers’ work.
15. Iqbal, S. T., & Horvitz, E. (2007). Disruption and recovery of computing tasks: Field study, analysis, and directions. Proceedings of CHI 2007, 677–686.Traced the intervening activities that can follow a computer interruption before a person returns to the original task.
16. Bardram, J. E. (2009). Activity-based computing for medical work in hospitals. ACM Transactions on Computer-Human Interaction, 16(2), Article 10.Developed and studied activity-based access to clinical resources, including support for suspension, resumption, roaming, and sharing.
17. Voida, S., & Mynatt, E. D. (2009). “It feels better than filing”: Everyday work experiences in an activity-based computing system. Proceedings of CHI 2009, 259–268.Followed a small group of knowledge workers using Giornata in their everyday work, revealing different ways of dividing work into activities.
18. Jeuris, S., & Bardram, J. E. (2016). Dedicated workspaces: Faster resumption times and reduced cognitive load in sequential multitasking. Computers in Human Behavior, 62, 404–414.Compared task-dedicated virtual desktops with a single desktop in a 16-person controlled study.
19. Bardram, J. E., Jeuris, S., Tell, P., Houben, S., & Voida, S. (2019). Activity-centric computing systems. Communications of the ACM, 62(8), 72–81.Systematic review of 101 papers covering 68 distinct activity-centric technologies.
20. Chang, J. C., Kim, Y., Miller, V., Liu, M. X., Myers, B. A., & Kittur, A. (2021). Tabs.do: Task-centric browser tab management. Proceedings of UIST 2021, 663–676.Built a browser extension that saved and reopened groups of tabs as tasks and studied its use in everyday browsing.