Position paper for CHI97 workshop: Awareness in Collaborative Systems

Jean Scholtz
National Institute of Standards and Technology
Gaithersburg, MD
jcscholtz@aol.com
301-320-7898

(Please use home mailing address:
6315 Bannockburn Drive
Bethesda, MD 20817
Note: I am currently working for UserWorks, Inc. but will have started my position at NIST by the time of CHI.)

Bio:

Dr. Scholtz is just starting in a position with NIST where she will be involved in evaluation of interfaces for collaborative systems among other projects. She is currently a consultant in user interface design and evaluation and has worked with several collaborative systems, ranging from large on-line web based products to an interesting main frame product with distributed users. Previously, she worked as the co-manager of the human factors group at Intel, concentrating on desktop video products. These products included ProShare desktop video conferencing and a one to many ProShare add-on, ProShare Presenter.

Dr. Scholtz has a Ph. D. in computer science from the university of Nebraska. She was on the faculty of the computer science department at Portland State University for four years prior to joining Intel.

Position statement:

I am particularly interested in evaluation methods for collaborative systems. In this respect, I am interested in developing dimensions of awareness in order to evaluate the suitability of collaborative products and group tasks.

My work at Intel involved desktop video conferencing products and a one to many collaborative product. Both of these product types were synchronous as they involved video, some of the awareness problems were solved. However, we found that participants still desired the following awareness information: where in the shared editor others were working, what pieces of information in the shared were contributed by other participants. In some instances, participants needed to know exactly where on the page others were looking - having the page number alone was not adequate. The multipoint version of the system could only display the live video of the person currently speaking, the other participants were displayed as still pictures. There were cases when people were busy working and the video was hidden. In these instances, participants were also interested in the question of who was speaking.

In the one to many system, a presentation could be given to others who would watch from their desktop. Awareness issues here were slightly different. Presenters wanted to know who was there (handled by an attendance list) but presenters were also extremely interested in knowing that everyone was receiving the audio and video as well as any presentation materials. The same is true of live presentations - most presenters inquire of the audience if they can see and hear. We provided both an audio and text backchannel for asking questions. Names were automatically filled in beside of submitted questions and requests for the microphone. However, many discussions were held about allowing anonymous questions to be submitted. This was not done, but this brings up an interesting question of whether awareness should replicate current possibilities or add new dimensions to current situations. If so, to what extent is behavior "controlled" by the dimensions of awareness present in a situation?

Recent work on evaluating a virtual places system has added to my concerns about awareness. Are there benefits of letting a person observe a situation but not participate? Outside of technical benefits what are the social implications of situations such as this?

I have also been involved in the evaluation of a main frame product with workers sitting at distributed terminals. They have expressed desires to know if collaborators are on-line, thereby increasing the chance that the information they have just submitted on-line will be observed soon.,br>

Although, I had originally been thinking that perhaps there were factors of awareness, I am now beginning to suspect that a high to low scale of awareness would be useful, with "factors" actually being points on the scale. For example, a low level of awareness might just be that others are there, increasing to the number of others present, increasing to the actual identification of who is there. Adding on factors of who is doing what with what artifact would add increasing points on the high end of the awareness scale. Two other scales would be useful in evaluating the control of awareness, one from the point of view of what access others had to information about you. The second scale would be what amount of control you had over what others could see and do. I indicate these as separate scales because an application could give others the ability to see everything you do, but you could be allowed the ability to control the amount of information revealed.

My intention in starting the collaborative evaluation project next month is to look at proposals for evaluating awareness, along with other dimensions of collaborative systems. An interesting question would be where along the suggested scales would a collaborative system have to be in order to be considered "as good as" face to face methods? Would moving up or down on the scale be considered an advantage over current face to face methods? This has been shown to be true for brainstorming techniques where anonymity improves quality of the brainstorming output. By using scenarios and different collaborative products, I intend to do some experiments to look at the effect of different degrees of awareness based on various group activities. I am hoping that the strategies for evaluation developed in the workshop can be integrated into the evaluation work we have planned.



Back to the Awareness Workshop Page