So here are my IMHO thoughts inline on the questions Jeff poses above:
What would cloning projects involve?
Cloning projects allows users to utilize JIRA in a application lifecycle management sense in that they can determine what feature items can be associated with typical "project" that parallels their SDLC to some extent. Each release cycle can automattically generate a "basic project" or template with all the typical issues automatically populated with default values that can be customized based on what is going on in the next release. This would help us in terms of software release where the development is more like an assembly line such as in information systems related to clinical research orgs (CRO). Same basic software is released multiple times for multiple clients, some with customization some not.
- Cloning just the project info (obviously the key would need to change)
At the very least the project info and base level issues. Perhaps a "batched, clone and move" event could create a fresh project from an existing "template" project. Perhaps maybe something more elegant if needed.
- Cloning associated permission, notification, workflow schemes?
Make this configurable. Ideally perms, notification and workflow is abstracted somewhat from the project level objects so that projects could "subscribe" to perm, notification and workflow schemes already in existence, or possibly subscribe to new ones. The batch clone and move idea could potentially just grab what ever linkages exist in the template project if the user selects this level of cloning.
- Cloning associated custom fields?
Make this configurable. Ideally custom fields are portable enough to be repository material that can be drawn into a particular project. The batch "clone and move" idea could potentially just grab what ever is in the template project if the user selects this level of cloning.
What are the problems that you're experiencing, that you think project cloning will solve?
Some of us want to be able to create an SDLC that is more like an assembly line VS. product leveling, or extension. So, more like banging out cookie-cutter versions of the same system with minor configuration changes vs a whole new version of an exsting product.
JIRA would be awesome to manage this, but needs to be able to repeat projects with base configuration without having to "hand create" the basic beginning state (by adding issues etc...) of the assembly line prior to a starting a new cycle.