A Performing Medium for Working Group Graphics
 

Fred Lakin

Center for the Study of Language and Information, Stanford University
       lakin@csli.stanford.edu
Center for Design Research, Stanford University
       lakin@cdr.stanford.edu

Rehabilitation R&D Center, Palo Alto Veterans Hospital
3801 Miranda Ave, Palo Alto, California 94304



Abstract

Writing and drawing together on a common display often assist a working group in a task. For example, face to face groups have long enjoyed the richness of graphic communication found on blackboards. The spontaneous image manipulations which take place over time on a blackboard can be viewed as a text graphic performance. A human performer generates and manipulates text and graphics for the purpose of assisting the working group in their task.

The phenomenon of performed text graphics presents opportunities for research in the area of computer supported cooperative work. 1] Spontaneous generation demands a performing medium where the focus is on live manipulation of text and graphics. Design of a computer based medium with enough agility and generality to support blackboard like activity is a challenge for interface design. 2] Agility and generality must not be achieved at the expense of specializability. After a group has initially sketched an idea in text and graphics, then that same medium should also support refining the sketch according to formal schema. 3] The performing medium can also be used as a recording medium for studying image manipulation as part of the working group process.

This paper presents a stepwise approach to the design of a performing medium for working group graphics. First, examples of non computer text graphics for groups are examined to get a preliminary idea of the underlying phenomenon: the performing of text graphic manipulation to assist working groups. Next key features of that kind of text graphic manipulation are isolated. Then, third, the architecture and behavior of a graphics editor providing those features is described.


 

vmacs transcription of paper graphics performance by David Sibbet


1.2 The phenomenon   First, the subjective impression after ten years of observing working groups is what we might call text graphic dance: chunks of text and graphics created at one location on the display, lingering there for a while, and then moved or changed or erased; continually shifting patterns forming and reforming as the group goes about developing and displaying their concepts. An expressive performing art, the meaning of which is in all of the intermediate imagery, where the final frame may not be any more meaningful than the final position in a ballet. And finally, the impression is also of a performing art so horribly artifact bound that text graphic dance as it was `meant' to be is only glimpsed now and then through the clumsy media that circumstances force the performers to employ.

Next, to more objectively characterize the phenomenon underlying the three examples: a text graphic performance to serve a task oriented group. The performance is text graphic in that it involves manipulating text and graphics over a set period of time. The performance serves the working group in that the images displayed during the performance relate to their task (present, represent, express, explain, diagram, show the structure of, mean). The performance is by an operator (or operators) whose purpose it is to insure that the performance serves the group.

We will call this kind of text graphic performance "working group graphics". There are many other kinds of possible text graphic performances that we won't discuss in this paper (a single operator for his own consumption, multiple operators for a play oriented group, etc.). There are many different possible styles of text graphic peforming within the kind we have designated working group graphics ...


8.4 The most agility imaginable   In order to get a better feeling for all aspects of agility working in concert, let's see if we can imagine the most responsive possible medium for text graphic dance. This imaginary performing instrument will represent the limits of agility which the present medium tries to approach. We might describe this medium in operation as hand powered performance animation. Not only are text and graphics moving as fast as the operator desires, but even the very quality of their movement in under his or her complete control. Admittedly such a medium is hard to imagine      complete animation of text graphics generated live in performance      but perhaps part of the problem lies with our ability to imagine ...

Consider the following parable. A cargo plane is flying over the jungle enroute to a very remote Club Med. Through a freak accident, the player piano meant for the night club accidentally falls out of the cargo bay along with hundreds of paper rolls, some with music and some blank. Miraculously the piano survives intact, the only damage being the total destruction of the keyboard. Discovered by a local tribe, one of their musicians figures out how the player piano can be operated in order to make music. It takes about two days to punch a roll which then produces two minutes of music. Years later an explorer happens upon the tribe and they demonstrate piano music for him. He listens politely and thinks somebody ought to tune the thing. Then he tries to explain the concept of a keyboard and its use in jazz improvisation. The tribe listens politely and thinks he is crazy. They explain patiently that his idea would never work: piano music is far too complex for human beings to make up and control in real time      and even if they could, no interface could possibly handle the bandwidth.


11.3 On What Can't Be Measured   The most important part of a text graphic performance may be the hardest to measure. Text graphics for working groups is important at all because it is expressive of the group's task. Some of the expression takes place in the static image content, and is relatively easy to measure. And yet finally, to return to the initial characterization of the phenomenon of working group graphics as being like dance, the most effective expressive quality may be the hardest to isolate scientifically: the quality of movement of the text graphics *. If each visual event is under full expressive control of the performer, then nuances of phrasing and timing may strike the audience very forcibly, but be very hard to measure. To algorithmically measure the full effect of a performance may be like trying to write a program to measure the effectiveness of a piece of music.

* This is point is directly due to conversations with Gayle Curtis and Margo Apostolos on robot ballet.




Contents

1. Introduction: Working Group Graphics On Non Computer Media
1.1 Examples
1.2 The phenomenon
1.3 Methodological assumption

2. Features Of Text Graphic Manipulation For Working Groups
2.1 Chronological
2.2 Text Graphic
2.3 Manipulatory
2.4 Performed
2.5 Fast
2.6 Unstructured
2.7 Structured
2.8 Reflective

3. A Computer Medium For Performing Text Graphics

4. Chronological

5. Text Graphic

6. Manipulations

7. Performing Medium

8. Agility

9. General Purpose

10. Specializable
10.1 Formal Visual Languages (Statics)
10.2 Formal Visual Dynamics

11. Reflection
11.1 Storing & Retrieving Performances
11.2 Measuring Performances
11.3 On What Can't Be Measured

12. Conclusion

13. Acknowledgments

14. Appendix: On the Measurement of Working Group Graphics

15. Footnotes

16. References


Published in the proceedings of the CONFERENCE ON COMPUTER-SUPPORTED COOPERATIVE WORK, Austin, Texas, December 3-5, 1986; reprinted as a chapter in COMPUTER-SUPPORTED COOPERATIVE WORK: A Book of Readings, edited by Greif, Irene, Morgan Kaufman Publishers, Palo Alto, 1988.


Return to papers list

(C) Copyright 1994 PGC