DICE: Distributed Interactive Collaboration Environment
DICE - What Is It?
The Distributed Interactive Collaboration Environment (DICE) is a general-purpose template for collaborative training, planning and information sharing environments. It can be customized for particular domains, combining specific scenarios and content with appropriate collaboration tools. For training applications it can be augmented with human- and/or computer-driven simulations.
Key DICE template features include:
| |
 |
Network architecture - running on the Internet, WAN or LAN. |
| |
|
Persistent workspace - allowing users or trainees to share documents and materials they produce. |
| |
|
Collaboration tools - commercial-off-the-shelf (COTS) and/or shareware tools selected and customized to meet user requirements, including the capability for enhanced automated data collection. |
| |
|
Human- or computer-driven simulation - to incorporate the reactions and interactions arising from users' real time decisions. |
| |
|
Realistic scenario - evolving on the web site over time, instead of situation summaries |
DICE can support audiences of varying sizes in multiple distributed locations or co-located exercises involving distributed work. The collaboration tools are custom-fit to the specific requirements of the users and their application.
 Back to the top of the page
The DICE Experiment
The DICE project began in 1999 and its objectives were to develop a prototype DICE and assess its effectiveness in an experiment with an operational user. The project's sponsors were the Department of Defense (DoD) C4ISR Cooperative Research Program (CCRP) and the Defense Advanced Research Projects Agency (DARPA).
The operational users helped define the requirements for DICE and served as a testbed for the experiment. The operational users were National Defense University, the Foreign Service Institute, and the US Army Peacekeeping Institute. These three learning institutions are responsible for training Presidential Decision Directive 56 (PDD 56) which mandates that, in the case of complex contingency operations (CCO), all involved US government (USG) agencies will collaboratively develop a coordinated political-military (pol-mil) plan. In the first phase of the project, a DICE/PDD-56 prototype was developed, combining free and low-cost collaboration tools with PDD 56 content. In the second phase, an experiment was conducted in October 1999 to assess the prototype's effectiveness for PDD 56 training, and to gain an understanding of the operational differences between virtual and co-located teams - how they work, the quality of their work, and how they develop shared awareness.
The experiment measured performance differences between two groups: one that trained with DICE/PDD 56 technology, and a No DICE (Control) group that met face-to-face as in previous PDD 56 exercises. The presentation of the Korea 2000 scenario varied for each group but both groups used similar curricula and role-players. Data was collected on the DICE/PDD 56 web site and through surveys, tests, and observations of the participants.
Sample results indicate the following:
| |
|
Output. Both DICE and No DICE groups were tasked to write two documents: a short paper describing US interests in Korea and a detailed pol-mil plan responding to the crisis. The No DICE group wrote a better paper but the DICE group created the superior pol-mil plan. |
| |
|
Process. The DICE group applied a divide-and-conquer strategy to develop its papers: in brief face-to-face meetings, they assigned different sections to different participants. Back at their offices, they completed their assignments individually and asynchronously and posted them on the web site for feedback. The No DICE group did all its work together as a single co-located group. |
| |
|
Training time. The DICE group spent more time training than the No DICE group, which spent 9 hours in meetings producing its work products. The DICE group's time, which varied by participant, was spent on-line at the PDD 56 web site (instrumented to capture individual's on-line time) and working at their desks (participants estimated time spent at their desks). DICE group training time averaged 10.6 hours: 6.6 hours on-line and 4 hours working at desks. |
Back to the top of the page
The DICE Prototype
An instance of DICE was created and customized for PDD 56. At the time of the experiment, the PDD 56 web site included training content, a Korea 2000 scenario, and synchronous and asynchronous collaboration tools. To test the collaboration capabilities used in the DICE experiment, test out the DICE Sandbox.
The original DICE/PDD 56 web site is at http://www.pdd56.com; note that the training part of the site password-protected to maintain the privacy of the training audience. This site will be used in the future by the PDD 56 community and will change over time.
We've created a sanitized version of the original PDD 56 site by changing the names of the training audience and limiting some of the functionality. This version is not password-protected and is at http://www.thoughtlink.com/PDD56
Using DICE, the training audience could:
| |
|
Collaborate with each other in real-time or asynchronously, |
| |
|
View the unfolding scenario (in this instance, economic and political deterioration in North Korea, followed by a coup and the possibility of reconciliation with South Korea), |
| |
|
Find additional information about North and South Korea, |
| |
|
Look at information on Presidential Decision Directive (PDD) 56 (the focus of the training), and |
| |
|
Find information about other members of the training audience. |
The real time collaboration was done through a custom interface to Microsoft's NetMeeting 2.11. When the training audience logged on to the PDD 56 site, they saw a "presence window" listing the other training audience members who were logged on at that time. Users could select one or more people from the list of those logged on and initiate a conference. In the conference, they could share applications, use text chat, or use audio conferencing over the Internet.
Asynchronous collaboration was done via e-mail and file sharing. An e-mail capability was incorporated onto the web site - solely due to the requirements for extensive instrumentation of the site. Without the requirement for instrumentation, the training audience would have been able to use their regular email packages.
Using the DICE bulletin-board style e-mail, users could reply to, or post new, messages. When they received a new message, their regular office e-mail accounts were sent the following message: "You have received an e-mail from John Doe at http://....". They would then click on the URL and see their DICE e-mail account.
In addition to e-mail, there was a file sharing portion of the web site in which users could upload and download files. These files were accessible by all members of the training audience. Back to the top of the page
DICE Benefits
DICE offers a number of benefits for training, planning and information sharing environments.
For training, DICE can provide benefits as a stand-alone tool, or by augmenting more intensive and expensive traditional methods. It can:
| |
|
Lower travel costs |
| |
|
Allow for just-in-time training |
| |
|
Encourage participation from multiple external organizations |
| |
|
Provide a ubiquitous tool for access to training resources |
| |
|
Support remote feedback from web site and collaboration tool instrumentation, and Provide a template that can be customized to fit specific training needs. |
Back to the top of the page
Analysis and Visualization Tool
As part of the web site and collaboration tool instrumentation performed under our FY99 work for the government, we have developed a cgi script to collect and read data and make it available for visualization.
Back to the top of the page
DICE-related Publications
Back to the top of the page |
