TOPIC: dependant actions

As I understand it, there were quite a few ppl already, who asked for a "dependency" function for actions.

Currently, the built-in functionality allows TR to automatically set an "Inactive" action to "Do ASAP", as soon as the previous one is marked as done. While this is a great thing, there are many cases, where you'd want an action to depend on several actions/projects to become "Do ASAP".

Considering the already built-in functionality, it would be almost trivial to build in this addition and it would be a great addition imo, because it would allow ThinkingRock to automise the search for next actions. Often I am able to tell, that action C can only be done after actions A and B, but I don't really care if I do B first or A first.

My idea how to implement that:

the tr.model.Action.Action might have a List of tr.model.Item.Doable objects on which it depends. If all the Doable in the list are done (and the list is not empty), then the action can go from INACTIVE to DOASAP.

In that way, you don't change behaviour if you didn't have any depends-on objects, but if you set any, it will automatically update the ActionState.

All that it would take more, would be some way of selecting the Doable's in the form and then you'd have a very nice addition to TR.

Dunno if you can do that, but I'd really love this feature.

P.S.: If you'd like me as a beta tester or you'd like me to help with it, i'd gladly help.

Yours Icey

It is not that easy to implement. As the moment for the auto-sequencing, you don't have a link between action mechanism. The software just pick up the next action within the same project.

To have the dependency, you will need to select the action you are dependent on. So users might want more than one actions. How do you select the action? From a project tree type window? What happens if the action you depend on gets deleted, etc... They will be many aspects to consider.

This is how another user solved the problem:

I solve this problem by enlarging the idea of "project" along the lines of the quotes from David Allen:

If the statuses of projects P1, P2, depend on completing project B, I create a "last' Action for project B:

"When completed, change the statuses of projects P1 and P2 by:...."

In this view, project B is not officially "done" until the statuses of the projects which depend on it are all revised. The changes usually consist of removing "waiting" actions and designating next actions.

The process ordinarily takes less than a minute per waiting project. It has the virtue of letting me reconsider next actions before activating them.

I do not want TR to load up with project management functions, even if it is feasible. There is other good software for handling project dependencies, resource availability, and scheduling conflicts.