Using group decision support systems (GDSS) as a backdrop, this paper looks
at the issues surrounding anonymity, with particular emphasis on how it ma
y be achieved. Ensuring anonymity requires more than simple encryption. Ano
nymity is viewed as a composite of three types of anonymity-environmental,
content-based and procedural. Each type is investigated and procedures deve
loped to improve the anonymity of the system. Behavioral and operational co
sts are identified for each of the three types of anonymity. For some meeti
ngs, the benefits of supporting anonymous communication may not justify the
costs. Mechanisms providing complete procedural anonymity in various syste
m configurations are presented. It is shown that five separate mechanisms c
an be used to provide procedural anonymity. The impact of relaxing each of
these mechanisms is also investigated. Even though a system provides anonym
ity, participants may be skeptical. This skepticism can reduce the overall
efficiency of the group. For this reason it is important to be able to conv
ince groups that their comments are indeed anonymous. The issue of how to c
onvince individuals that their anonymity is secure is discussed and a parti
al solution to this problem is presented. (C) 1998 Elsevier Science B.V. Al
l rights reserved.