How to handle naming schemas for timelines? Best Practices?
I was wondering if anyone would be willing to share their experience and best practices with respect to naming for timelines / development lines and iterations.
The ootb MTM example uses a schema that is not very scalable. Even with the Scrum template, I find my self wondering if the used schema really scales as the number of developmentlines grows.
When setting up projects I ended up using schemas similar to:
-
Project 1
-
Release 1
-
Iteration 1 (R1, Project1)
- Iteration 2 (R1, Project1)
- ....
- Release 2
-
Iteration 1 (R2, Project1)
- Iteration 2 (R2, Project1)
- ....
One of the main reasons for a schema like that is that it makes using the planned for picker much easier to use.
I also find editing the timelines a bit tedious and error prone and have thought about creating a command line tool or eclipse extension that would help me doing it. For example create a timeline with name, start date, schema Line/Release/Sprint, number of releases, number of sprints, duration of sprints. Or to add a release to a timeline etc. Another enhancement I mused about is something to keep the iteration dates consecutive (e.g. if you are not so agile and want to extend an iteration in the middle).
Would be great if anyone could share their experience. I would also be very interested in your thoughts on improvements in the UI especially when maintaining development lines.
Any feedback would be greatly appreciated.
Accepted answer
we have a slightly different approach. First to understand our approach you need to understand our definition of Backlog.
Backlog ist a collection of all NOT planned items.
Based on that we would like to be able to create Plan for the Backlog as Plan for the Release (Backlog + all Sprints). So we need to create Helper Iterations.
- Program (MyPrj)
- Program_Backlog (MyPrj)
- MP1_Release (MyPrj)
- MP391_Backlog (MyPrj)
- MP391_S01 (MyPrj)
- MP391_S02 (MyPrj)
We have build today a small java script to create iterations. As input we use a csv file. With the same parameters you propose, but as one line per iterations.
One of the idea for future enhancement was to create a Project-Definition-File with:
- Prj Name
- Abbreviation
- Release ID
- nbr of sprints
- start date of release
- sprint duration
and enhance our tool to create the csv file based on this file. Sure that a good part of the reult is hardcoded (e.g this backlog iteration)
Next step would be to add some hierachy design in the definition file
That's about it
Hope this gives you an idea how we do it