Iteration 1

• Phase: Elaboration
• Dates: December 16 - January 27, 2010
• Finished:

Tasks to be Completed

+ Source Code

• Review existing v6.0 code on the repository. How much was completed? Can it all be reused?

+ Project Infrastructure

• Setup new project on the LANDIS-II web site (Rob)
• Revise to-do list on the LANDIS-II web site (Rob first, then Jimm)

+ Unified Process

• Start the list of project risks (Rob)

Assessment

+ Tasks Completed

• Code review on repository: No major changes to the core since the version-6 branch was created in 2007 as part of the dual-scale work. Merging the version-6 branch into the trunk (= main codeline) will be straight-forward. However, a few libraries (e.g., biomass cohorts) will need some preparation before merging their version-6 branches into their respective trunks. Existing extensions have changed so much since their branches were created that we determined that the most expedient way to proceed is to discard those branches and instead, update the current code on their trunks to work with the version 6 core.

• How to manage code-lines? How do we do ongoing maintenance while actively maintaining code? Not a problem with the core because of its demonstrated stability. More of an issue with extensions which undergo more rapid evolution. We developed a codeline policy: after work begins on an extension's main codeline to update it for the v6 core, if a critical bug in discovered in the extension's current version (i.e., the version that works the v5 core), we will create a v5 maintenance branch. We'll use this branch to promptly fix the bug and release an update of the extension's current version.
• New project set up on web site as evidenced by this iteration assessment.
• Project Risks: Listed.
• Feature list: Rob edited, Jimm edited, edited together. Each item given a priority (generally high, medium, low). Many items deleted as they were outdated.

+ Iteration 1 Hits and Misses

• Huge Hit: The Core has been remarkably stable. Our plan worked!
• Miss: Our notes (to-do list) were a bit out-of-date. Hardly surprising.

+ Iteration 1 Lessons Learned

• Iterations may need to be longer due to our respective changes in positions. May not be able to do 3 week iterations. We will need to be more flexible.
• Going over the v6 features list (the to-do list) together was extremely valuable. Many ideas were outdated or outmoded. We now have a much stronger sense of what needs to be done, i.e., we have ranked everything very accurately.