Photos Spare Cycles MythBusters

Paper: ClearBoard

Iterative Design of Seamless Collaboration Media
Hiroshi Ishii, Minoru Kobayashi, Kazuho Arita

This paper presented several interesting prototypes of computers that you can look "through" to see the person you are collaborating with (imagine standing on opposite sides of a glass wall with markers).

- Focus in on new application concepts, rather than technology itself
- Buxton's Design Principle: "Let's do smart things with stupid technology today, rather than wait and do stupid things with smart technology tomorrow."
- Seamlessness (1): projects applications are seamless/compatible with current work practices. Continuity.
- Seamlessness (2): seamless transition between functional spaces (e.g. personal->shared, physical->digital).
- Each group member choice of environment should be independent
- heterogenous set of toolset should be supported
- each group member is free to use whichever tool(s) fits their style (computer vs. physical desktop)
- tools should be usable simultaneously
- Facial expressions and gestures are important for building trust (Buxton)
- Videophone and video conferencing focus on "interpersonal spaces" (spaces that provide information on facial expressions/gestures, "telepresence")


Iterative design from original TeamWorkStation project to ClearBoard project. The first TeamWorkStation prototype operated on a shared workspace model, where electronic and physical desktops could be video composited onto a screen. The mixed media allowed participants to interact with the system using traditional physical input (e.g. normal pen/pencil) and also display electronic and paper documents in the same space, but it did not allow for collaborative document editing directly (not a shared data system). This first prototype was not field tested due to the complex wiring.

The second prototype switched the backbone to N-ISDN and also added "ClearFace" which superimposed the participants images over the shared desktop instead of in separate windows (in order to save screen real estate). Experimental results focused mainly on the video performance, which was somewhat jerky and also not high
enough resolution for easy character reading (352x288). The paper includes some comparisons with videophones, with the insight that the desktop image tends to help ground the converstation and enhance the productivity of the system, because it is relatively static, as opposed to the jerky image of the other person.

The next prototype, ClearBoard, was motivated by the authors observation that there is a lack of
seamlessness between shared workspaces and interpersonal spaces. The interpersonal space does not allow participants to use non-verbal gestures to communicate information in the
shared workspace (e.g. talking head in a separate window).

The ClearBoard 1 prototype was dual vertical boards that participants could draw on. Displayed on the drafting board was the shared drawing space, as well as a mirror-image of their partner ("through the looking glass"). The experimental results of ClearBoard showed that it enhanced "gaze awareness." Participants were aware
of the other's gaze and could use it to resolve ambiguities or to have their attention drawn.

One failure they noticed in the ClearBoard system was that participants had a WYSWIS (why you see is what I see) expectation, so they would (for example) draw glasses and moustache's on the other participant's face, not realizing that the participant could not see their own image.

The ClearBoard 2 prototype focused on using TeamPaint as the drawing program so that participants could save results. Each participant had their own drawing layer and could not modify another. Also, the prototype was made more horizontal to decrease arm fatigue.

The authors felt that the ClearBoard prototypes represented a shift to HHI (Human-Human-Interaction) rather than HCI, as the participants were essentially looking through the computer and less aware of its presence.

Post a comment

related entries.

what is this?

This page contains a single entry from kwc blog posted on May 20, 2003 2:20 PM.

The previous post was Freaky numerical coincidences.

The next post is Paper: mediaBLOCKs.

Current entries can be found on the main page.