The 90% feature that we are looking for here is the ability to use drag-and-drop to set strict issue order on issues across projects. We don't really have as strong of a use case for the rest of GH across projects, but the GH implementation of drag-and-drop issue ordering is pretty close to ideal except for the lack of cross-project functionality. If the only way the GH team would implement drog-and-drop issue order across projects would be to properly integrate all GH functionality as cross-project capable, then I offer the following opinions:
- How do you schedule issues in to a fixVersion? Do we require the same fixVersions for all projects? Components?
- Do we require the same Master/Child relationship for versions/components? What about differing Start and Release Dates?
These problems would really not exist if a single fixVersion entity could exist on multiple projects, which is a separate but related missing feature in JIRA. The workaround for GH would be to implement a mechanism that tracks "sameness" between fixVersion entities. (Such a feature would be fantastically awesome on its own for searching purposes, even if you didn't use that work as a platform to implement the feature described in this issue.)
With that "sameness" mapping, it would be possible for GH to always display the union of all fixVersions on projects represented by the displayed issues. GH would also then be able to create a new fixVersion entity on a project that is the "same" as a fixVersion from another project when GH is used to assign an issue to one of those fixVersions that doesn't already exist on that issue's project. Master/Child relationships and start and release dates would automatically match because they would already be known to be the "same" fixVersions, therefore living in the same hierarchy.
- What happens when there are different issue type schemes for different projects? For instance, Bug vs Defect?
I'm not sure I quite follow where this matters, but I think you would just need to set the behavior globally instead of for each project. If you're talking about two entities with different names but having the same meaning (as in "Bug" and "Defect"), then I think that's a problem for the JIRA administrator, not the GH developers.
- What permission scheme do we use? What is displayed if some users have permission to view/edit/schedule all issues from all projects and other cannot?
You just display what the current user has permissions to see. This could be confusing if a user is not configured to see the right issues, but again, that is a problem for the JIRA administrator of that system, not the GH developers.
- GreenHopper charts are generated dynamically for the most part, how do we ensure everyone is looking at the correct information?
You're talking in terms of permissions? That's just a call you will have to make. Either the charts reflect what the user has permissions to see (see my answer to the previous question) or you override the permissions and make the graphs reflect the state of the world. This problem is not unique to GH... JIRA's built-in charting has the same problem and they went with the former solution, which seems fine.
Folks,
This has been addressed with the Rapid Board in GreenHopper 5.8. Rapid Board development continues, specifically around the scrum use-case and improved epics in the short term.
Thanks,
Nicholas Muldoon