The original objectives and deliverables for this project were:
- To migrate the Student Tracking, and SAGE applications to ColdFusion 9
- To update supporting technical documentation for the migrated application
- Migrated versions of Student Tracking and SAGE in DEV, TEST and LIVE
- Updated technical and support documentation
Not all of the above has been achieved. The CCTS application (Student Tracking) was eventually deployed into LIVE in September 2012, but the SAGE programs were eventually dropped as it was thought that the need to migrate them would be superseded by the adoption of a replacement system, and the fact that extra effort was required for the migration of CCTS. However, it has transpired that SAGE will now need to be migrated to CF9 and this will be covered in 12-13 by an in-year project
The project sponsor and stakeholders have agreed that the project can now be signed off.
Analysis of Resource Usage:
Staff Usage Estimate: 44 days
Staff Usage Actual: 47 days
Staff Usage Variance: -6.81818%
Other Resource Estimate: 0 days
Other Resource Actual: 0 days
Other Resource Variance: 0%
Explanation for variance:
The extra days that have been required have simply been used on project management time as the original project ended up being done over a longer period than originally planned. The main reason for this extended duration was that the migration of CCTS took longer due to the limited number of people fully familiar/trained on CF9, and then the testing was delayed because of inconsistent codebase versions between DEV and TEST. There were further issues leading to the extended length of the project, including a change of lead developer and problems with testing accounts.
Key Learning Points:
This was another ColdFusion migration project that experienced problems around planning and resourcing but it also suffered from issues that had not previously arisen on the other migration projects, leading to a 'different' experience for all involved and one that took longer to resolve.
The main difference was that the developer undertaking the upgrade had not previously been exposed to the way in which a couple of CF6 features (e.g. local scoping) needed to be properly converted to CF9, and was not familiar with some of the foibles of the CF conversion helper tools provided by Adobe. Only a few developers would have had the necessary CF9 experience to catch these problems but they were already resourced to other projects.
The resolution of these problems may only have led to lessons learned about particular aspects of CF9 that may not be applicable elsewhere but even subjective lessons such as these are of value.
There are no outstanding issues. The migration of SAGE is to be undertaken as a 'simple' migration as part of an in-year project as discussed with the Programme Manager.