AC 2007-346: DOCUMENT MANAGEMENT IN TEAM-ORIENTED, PROJECT-BASED COURSES: EVALUATING A LATEX/SUBVERSION-BASED APPROACH

1 AbstractThis paper discusses a low-cost approach to the implementation of a document versioning systemfor technical reports. Several alternatives have been considered, including commercial documentcollaboration services such as NextPage 2™ (NextPage, Inc.) and SharePoint™ (Microsoft Inc.),open-source versioning applications such as Subversion or CVS, wikis, and free web-basedservices such as Google Docs & Spreadsheets (formerly Writely). This paper explores thesealternatives and then focuses on a versioning system-based solution as the approach judged mostappropriate for our requirements.2 Introduction and BackgroundMost engineering and technology programs place a high value on team-based assignments andprojects. At the University of Detroit Mercy, a written project report is often one of the requireddeliverables from each team. When the size of the team exceeds two or three, collaborative reportwriting becomes problematic. Ad hoc processes aimed to keep track of who is working on what,and which version is actually the correct current version often break down, and chaos ensues. It isalso difficult for an instructor to find clear evidence of an individual’s contribution to the report.The typical document revision cycle for a student team goes something like this:1. Student A creates a draft outline of the document in Word and includes a draft of the sectionfor which he/she is responsible.2. Student A e-mails the document to teammates, usually with some indication as to whoseturn it is to revise the document (Student B). Problems arise if more than one person tries torevise the original document. In such cases, merging the changes from several reviseddocuments is not always straightforward.3. After adding other content and/or revising Student A’s content, Student B e-mails thedocument to the next team member.4. Eventually, the document comes back to Student A, and another cycle of revision can begin.Students often wait until the last minute to begin project reports, and this ad hoc system oftenbreaks down. A team member may go missing just when teammates are waiting for an e-mailwith revisions from her/him. It often falls to the team leader (or Student A in the above scenario)to plug as many gaps as possible at the last minute, based on what may be a limited knowledge ofthe details of the part of the project in question. This often results in project reports that read asthough they were slapped together at the last minute (because they were).Changing the revision cycle so that all team members can work simultaneously on the documentmay not prevent the last-minute scramble to write the report, but it can address the delays that areintroduced in waiting for the document to travel sequentially among the team members.