THE IMPACT OF AWARENESS IN

COLLABORATIVE TEXT EDITING

MARTIN LNSMANN HANSEN

MORTEN MLLENBERG BISBO

PELLE BJRN NIELSEN

AARHUS UNIVERSITY

AARHUS UNIVERSITY

AARHUS UNIVERSITY

[email protected]

[email protected]

[email protected]

ABSTRACT
Collaboration in multi-display groupware is often
loosely defined in terms of actually collaborating. In
this paper we examine how factors, such as awareness
and turn-taking affect the quality of the content we
create in a multi-display and single-display text editing
program. Through a design-oriented research approach
we have used an ongoing project as initiator to get into
researching the field of single-display groupware and
multi-display groupware. This paper sets up to two
different paradigms to describe the collaborative
differences, advantages and disadvantages.
We try to determine the roles in collaborative textediting, the matter of awareness and the importance of
turn-taking - depending on the collaborative platform in
use. All of it to raise the question: What is collaboration
in text editing and what makes co-working in a shared
space to be defined as collaboration? By examining the
art of collaboration, from two self-constructed
paradigms, this papers evaluates on the pros and cons in
those types of collaborative text editing.

KEYWORDS
Groupware, collaboration, awareness, multi-display
groupware (MDG), single-display groupware (SDG),
multi-user interaction, social interaction.

INTRODUCTION
In 1990 Greenberg sparked the discussion about
collaborative computing. Most software is developed
for single-user purposes and social interaction is a major
stakeholder in the collaborative work in terms of turntaking, awareness and active participants. When
designing a collaborative digital workstation it is
therefore important to be aware of roles, location and
purposes for the software. The goal in digital

collaboration is to eliminate physical restrictions and

distance between people working together (Greenberg
1990). But how important is physical presence in
collaboration? Are we actually creating collaborative
content - or are we just remixing content without
awareness? In this paper we introduce two paradigms,
we call the turn-taking paradigm and the Google Docs
paradigm. These two paradigms have different
approaches to collaboration - and different advantages
and disadvantages.
Our main contribution is an overview of different types
of collaboration - what we call different collaborative
paradigms and how these different ways of
collaborating affect the end product and especially the
quality of it.

RELATED WORK
Greenberg discussed the use of single-user application
in multi-user situations. He introduces the use of turntaking in view-sharing software and the different roles
in cooperating on a computer. He describes four major
human roles, which he call The View Manager, The
Chair Manager, The Registrar and The Meta Manager.
Technology has however changed since Greenberg
published his work in 1990, but it is still relevant to use
these roles as a way of thinking the design decisions in
our system, but not as human roles (Greenberg 1990).
Especially The View Managers tasks can be managed by
the software today, but the problems he as a human role
faced are still applicable, such as: What if one user
changes the size of a window? Closes a window?
Changes font-size and et cetera.
Greenberg furthermore discusses what he calls 'The
Human Component' in his paragraph about obstacles
and issues - and the concept of WYSIWIS (What You
See Is What I See). He claims that sharing the same
view creates a variety of obstacles, both in settings but
also in terms of possibilities for personalizing. The
system needs to be collaboration-aware, as he frames it.
(Greenberg 1990).
McGill, Williamson & Brewster (2014) also discuss the
roles of participants. By monitoring who is writing, this
writer also takes on a role of leadership and they
experienced that the other participants sometimes
adopted the role of bystanders. (McGill, Williamson &
Brewster 2014)

You could argue that this can be avoided by integrating

turn-taking through social interaction, this way no one
(hopefully) would adopt the role of a bystander, since
everyone have their own segment of work. This
however is not possible in all collaborative tasks and
one may argue that it removes the point of
collaboration.
In further relation to turn-taking in collaborative
systems McGill, Williamson & Brewster (2014) found
that "participants showed a strong inclination toward
taking the display .." and ".. using their social
capabilities to determine the acceptability of taking the
display .." In turn-taking there is a game, as Nofsinger
also describes it, where each participant want to
attribute and an ongoing competition to actually take the
turn (Nofsinger 1991). Having a shared display
furthermore created a shared focal point, which makes
creativity sprout (McGill, Williamson & Brewster
2014).
AWARENESS:

Awareness within collaborative systems is a key issue

that has been discussed for many years (McGill,
Williamson & Brewster 2014). Greenberg, as cited by
McGill, Williamson & Brewster (2014), defines three
types of awareness: workspace awareness, which
defines a persons understanding of another person's
interaction with a shared workspace. Artifact awareness,
defining what objects they are working on and action
awareness, what are they doing.
McGill, Williamson & Brewster researched the topic of
small groups collaborating using a shared view, in their
case a TV, to facilitate collaborative work. Their results
showed that flexibility in functionality is a key concern
when developing groupware systems. They experienced
issues with the participants ability to be aware of the
other participants activity. The collaboration in a shared
view also had a consequence of compromising with the
participants independence, which the participants meant
they had more of using a personal device, such as a
mobile device.
Wallace et al. made similar awareness experiments with
SDG and MDG systems. In both the SDG and MDG
systems they had a shared display, but in the MDG
systems they also had a personal display. This resulted
that the participants rarely used the shared display, even
though the results showed that using the shared display
created a much higher level of awareness. (Wallace et
al. 2009)

thoughts and findings of this paper. The project has led

us to discuss interface design and collaboration, which
created interesting discussions on how to best
collaborate in text editing software (and highly expert
driven software such as Photoshop). Using our project
and the process of developing it as a way to explore the
field is a well-known method. The approach has close
resemblance to Fallmans (2007) way of thinking designoriented research in interaction design, where research
is the area, but the design is the means. In that sense,
you could argue that we have used our project as a way
of gaining new knowledge within a field of research, but
creating a final product has never been the main purpose
of the project, hence the use of the word project and not
product about our specific project (Fallman 2007).
Furthermore, Fallman (2008) introduced the design
triangle as a way of thinking interaction design (as
illustrated below). As until we started discussing our
contribution for this paper our project was mainly
carried out in the design practice triangle. We had
performed interviews and observation in our domain
and our concept was driven much around the specific
domain. Our research approach by that time could
arguably also be determined as research-oriented design.
However when we started creating an interface for our
concept we discovered a general problem which we
started researching further: The problem of designing
for multi-display groupware for multi-users. Looking
further into this field of research we could not find
anything about the actually quality of the content that
was created in these collaborative contexts. It
furthermore raised the question of how do you actually
determine the term 'collaboration' and how big a part
awareness was in this discussion. By going from a very
context driven design process with the concept as main
focus, we have now moved into a more general problem
and thereby into the design studies triangle. You could
therefore argue that we have performed what Fallman
calls a trajectory and moved between two of the activity
areas. As our project is still ongoing this trajectory could
however end up being a loop, by going back and forth
between the design practice and design studies triangles,
as they can become inseparable (Fallman 2008).
Fallman strongly recommend to move around inbetween activity areas, which what "makes interaction
design research fresh, innovative, and unique."

METHODS
In the work with this paper we chose to use the
discussions we had as part of our process in developing
a collaborative blended interaction space that enhances
creativity. Our project 'Portal' itself is therefore
irrelevant to this specific paper, which is why it will
receive no further introduction. However it is important
to note that it has been the main initiator behind the

!2

Figure 1: Fallman 2008, p. 5

RESULTS

DISCUSSION

Greenberg (1990) notes that a voice channel is

important for collaboration and coordination and when
discussing text editing collaboration we also work under
the assumption that you have at least a voice channel for
communication during the collaborative process; either
through VOIP or by sitting in the same room.

In Google Docs there is no turn taking system built into

the system as everybody using the software can interact
with it simultaneously - unlike using single-user
applications like Microsoft and using a third party
screen sharing software. It is however possible to assign
these roles, as Google Docs has a editor, a commenter
and a spectator setting for each user. Wallace et al.'s
(2009) studies showed a higher level of coordination
problems in these type of MDG systems. You could
argue that there is no need for turn taking as users can
write simultaneously, but there needs be to a level of
awareness for the collaboration to be effective. Wallace
et al.'s (2009) work did on the contrary also showed that
the solutions developed in MDG systems were more
accurate than the ones developed in SDG systems.
Depending on how you work together it is possible to
work simultaneously - probably not on the same line of
text, but in the same document. Seeing it in Norman's
terms of affordances (Norman 2002) you could argue
that Google Docs affords you to use it in this way
(because it is possible even though you could use it in a
writer/reviewer fashion). But if e.g. you have
coordinated your work and you are working with
someone on two sections at the same time, you are not
aware exactly what the other person is doing and you
are working on the same document, but not on the exact
same thing at the same time - can it still be considered
collaboration?

When it comes to MDG collaborative text editing we

see there being two (somewhat contrasting) paradigms:
The first is what we call the turn-taking paradigm
where one user has the role of writer and shares her
screen to allow others to follow her work and give
feedback during the process and possibly switch roles at
some point. This user can hand off control of her system
allowing others to assume the role of writer and
conversely she would become reviewer or commenter.
This type of work necessitates turn-taking as you need
to establish who has the role of writer and when/if you
should switch roles. This type of work can be done
using e.g. Microsoft Word and a screen sharing and/or
remote control software like TeamViewer.
The second type of collaboration we refer to as the
Google Docs paradigm. In Googles Docs software,
which is a part of their cloud platform Google Drive,
multiple people can write simultaneously; the interface
is devided into a personal space and a shared space; the
toolbar is personal and can have different settings for
different users (one user can write bold text and the
other one can write non-bold as the toolbar and its
settings are not mirrored across screens) and the paper
writing area is shared (and everything happening in this
space is mirrored across screens).

Figure 2: Google Docs in Google Drive (in the webbrowser) - shared

(mirrored) vs. individual space (non-mirrored). 1. The productive area:
a shared workspace that is the collaborative space. 2. The static
toolbar area that is individual to each user (non-mirrored).

In Google Docs you do not need to hand off and

turntaking as multiple users can write simultaneously.
As a consequence of this, we have found that instead of
assuming a writer/reviewer role, in Google Docs you
often coordinate your work at the beginning and perhaps
underway and hence you often write different sections
of the same document simultaneously. Although this can
be seen as a more effective way of working it might not
lead to the highest quality work as this means that you
often do not review each others work during the process
and hence the quality of the work may suffer as the
writers lack the critical input from others (each other).

This begs the question: What exactly is collaboration? Is

it still true collaboration if you are not fully aware of
what the other people you are working with are doing?
This question of participant awareness is a key issue in
the field as noted by McGill, Williamson & Brewster
(2014).
We hypothesize that there is a difference in the quality
of the end product depending on which type of
collaboration is used (which paradigm). Although the
Google Docs paradigm seems to be more time efficient,
we would argue that the turn-taking, often review
inspiring paradigm would produce better quality work.
This would make sense seeing as you constantly review
the writers work and give her feedback and negotiate
about the work being done whereas the Google Docs
paradigm often inspires working on different sections
independently (although in the same document/
workspace) and thus you are not as aware of what the
other person is doing thereby missing a potentially
important feedback and negotiation process that could
otherwise lead to higher quality work - at least that is
our hypothesis and we feel that this could be interesting
to test in further studies. It is still important to note that
awareness is still a key issue though, also in the turntaking paradigm, as the quality of the content will only
be high, if the actually reviewer is aware.

CONCLUSION
By reviewing earlier collaborative research projects, this
paper tried to set up two different paradigms of
collaboration in text editing. We did this to enable the
discussion about the matter of turn-taking and adopting

different roles as a participant in collaborative text

editing. Our key finding, besides setting up two
different paradigms and ways to understand this field of
research, shows that awareness is, as noted by McGill,
Williamson & Brewster, extremely important when
working with collaborative systems. Even though the
turn-taking paradigm seems to create a higher quality
content it is a single-user application used in a multiuser situation. Therefore you could argue that social
interaction is also of extreme importance for the quality
of the content created. The constraints nested in the
turn-taking paradigm enhances creativity, whereas the
openness of the Google Docs paradigm enhances
delegation of work and a more effective, yet less quality
controlled way of working.

REFERENCES
Fallman, D. (2007). Why Research-Oriented
Design Isnt Design-Oriented Research: On the
Tensions Between Design and Research in an
Implicit Design Discipline. Knowledge,
Technology & Policy, 20(3), pp.193-200.

!4

Fallman, D. (2008). The Interaction Design

Research Triangle of Design Practice, Design
Studies, and Design Exploration. Design Issues,
24(3), pp.4-18.
Greenberg, S. (1990). Sharing views and
interactions with single-user applications.
SIGOIS Bull., 11(2-3), pp.227-237.
McGill, M., Williamson, J. & Brewster S.A.
(2014). Mirror Mirror, On The Wall:
Collaborative Screen-Mirroring in Small
Groups. Proceedings of the 2014 ACM
international conference on Interactive
experiences for TV and online video, pp. 87-94.
Nofsinger, R. (1991). Everyday conversation.
Chapter 4. Newbury Park, Calif.: Sage.
Norman, D. (2002). The design of everyday things.
New York: Basic Books.
Wallace, J., Scott, S., Stutz, T., Enns, T. and
Inkpen, K. (2009). Investigating teamwork and
taskwork in single- and multi-display
groupware systems. Pers Ubiquit Comput,
13(8), pp.569-581.