Jonathan D'Andries
gte498e@prism.gatech.edu
April 28, 2002

CS 4903-G03 with Dr. Mark Guzdial, Dr. Spencer Rugaber, and Kathy Gray

Ectropic Research Semester Report

From the beginning, my goals for this semester were to learn more about software 
engineering, gain research experience, and have an excuse to work with Kathy Gray 
and Dr. Guzdial.  These goals narrowed as the semester progressed and as I began 
to take a more active role in my own education.  In fact, my semester can be 
divided into three distinct phases of progress and vision:

1. release the ECoDE prototype and documentation
2. explore Ectrospace concepts
3. develop a model for the Ectrospace server.

In this report, I will analyze each of these phases to point out my role and what 
I have learned from each experience.  I will then discuss a "wish list" of things 
I could have accomplished and what I could do to improve in the future.

1.  ECoDE Release

During the first five weeks of the semester, we scrambled to release ECoDE and 
the corresponding documentation.  Kathy did a great job of finishing the user 
interface and implementing key functionality including logging, color coding, and 
improved responsibility management.  I spent a few weeks working with Kathy until 
I felt comfortable enough to begin work on the saving and loading mechanism.  Three 
iterations later, we settled on a system that allows users to save their work either 
as a squeek project (.prj) or as an ectropic design (.ect) file.  I then completed 
the following documents to support the release of ECoDE:

The users manual can be found at 
http://swiki.cc.gatech.edu:8080/ectropic/55.
The alarmclock example in PowerPoint can be found at 
http://coweb.cc.gatech.edu:8888/cs2340/uploads/Ecode-AlarmClock-Example.ppt.

The success of our early work can be measured by the number of patches released (one), 
the number of coweb questions regarding how to use ECoDE (zero), and the usability 
responses from the students (mixed).  With only one patch and the apparent ease of 
learning the system, ECoDE seems to validate itself well.  Through discussions online 
and with me personally, however, students expressed difficulty with the interface.  
They found the multiple window interface cumbersome and wished ECoDE could import the 
work they had already completed on the project.  This feedback is valuable and can be 
used to create a better product in the future.  I am proud of my work on ECoDE and 
believe it will provide interesting results that will help to build a truly Ectropic 
environment.

2.  Ectrospace Exploration

We hit a lull after the patch was released and students had submitted their first ECoDE 
design.  During this time Kathy reminded me of what I could do to contribute more to 
the Ectropic project.  She looked for me to ask questions, perform product research, 
and document our discussions and findings.  This was the turning point in my semester.  
I began by researching other design programs that were built for open source 
environments.  These products included

ArguUML - http://argouml.tigris.org/servlets/ProjectHome
Poseiden - http://www.gentleware.com/

Both the products and processes used to develop them were influential in my analysis of 
Ectrospace.  I documented all our correspondence, and our meetings became more focused 
as we set real goals and made design decisions about Ectrospace.  This led into the 
third stage of research - developing the model.

3.  Model Development

My final step was to create a well-defined description of Ectrospace.  Along these lines, 
I made three attempts at a task breakdown chart.  The first turned into a graph of basic 
navigation, the second a chart of system functionality, and the third a rough draft of 
user tasks.  My hope is to pass on a vision of the Ectrospace model to those students 
who will continue to build the Ectropic project.  A finished product may be several years 
and a few graduate students down the road.  My contribution to the Ectropic vision is 
to reduce the amount of research that must be repeated by the next person in line.

"Wish List" - Room for Improvement

If life were perfect, I would like to have documented the model of Ectrospace with a 
goal tree, implemented a limited prototype built off the Swiki, and presented a list 
of future research questions.  As it stands, I have documented much information to 
contribute to these artifacts, but have not created a finished product.  Looking back 
at this semester, I could have started researching earlier, produced diagrams earlier, 
and worked more with the Swiki group.  These may be limits on the usefulness of my 
semester of work, but they also serve as lessons for future research.  I am better off 
for my experience and hope to continue to grow as I look toward the future and graduate 
school at Georgia Tech.