On Tue, Feb 5, 2013 at 4:41 PM, David Nalley <da...@gnsa.us> wrote: > On Tue, Feb 5, 2013 at 6:22 PM, Marcus Sorensen <shadow...@gmail.com> wrote: >> On Tue, Feb 5, 2013 at 4:15 PM, David Nalley <da...@gnsa.us> wrote: >>> On Tue, Feb 5, 2013 at 6:12 PM, Chip Childers <chip.child...@sungard.com> >>> wrote: >>>> On Feb 5, 2013, at 6:10 PM, Marcus Sorensen <shadow...@gmail.com> wrote: >>>> >>>>> Well, for example, take >>>>> https://issues.apache.org/jira/browse/CLOUDSTACK-644. It is >>>>> 'unresolved', but the code was merged a few weeks back. It has had two >>>>> subtasks for awhile. Do I close it, or do I wait because those >>>>> subtasks are dependencies of the whole package? >>>> >>>> Please to keep it open. >>>> >>>> Sorry for the confusion, but I've been talking about maintaining >>>> comments about status, as well as keeping the fix version accurate. >>>> >>> >>> So closing the ticket doesn't close the sub-tasks - they are >>> sub-tasks, not dependencies (or vice versa) >>> >>> --David >> >> Yes, I didn't think it would close the sub-tasks, but sub-tasks by >> definition are things that need to be finished before completing the >> task, no? > > Hmm, I think they are all distinct, and only really 'linked' to the > issue of getting code in which seems complete, but perhaps I am wrong. > > --David
I think in Jira the coupling is relatively loose, I was talking more about sub-tasks as a concept. If I had a task to bring a birthday cake to your party, and I had sub tasks of 'baking the cake', 'decorating the cake', and 'delivering the cake', then the task wouldn't be completed properly until I had done all three. We have a cake, the code has been committed, but it's not decorated. We could bring it to the party, but it wouldn't be pretty. Perhaps there should have been a sub-task to code/commit the solution, although I'm not sure how granular we want to get.
