Hi,
my thinking in regard to Mylyn Docs is that it has its own integrator community (such as Intent) and is used independently of the task-focused interface so that it makes sense to keep it as an independent project. I would even consider reversing dependencies on the code level and moving the 2 bundles in Mylyn Docs that have dependencies on Tasks and Context into TFI to decouple targets, builds etc. but that's a separate discussion.
My initial thought was to leave the Mylyn Docs structure untouched and keep Intent and Vex as sub-sub projects. That would require fewer changes and as you point out makes sense in terms of reflecting dependencies. On the other hand, having a flat structure makes it a bit easier for users to discover Intent (at least in the PMI) and would provide more consistency.
I don't have a strong opinion on whether Mylyn Docs should be restructured. It's decoupled from merging the other sub-projects and we could always do it later if it makes sense and proceed with the creation of the TFI project (which is the key driver).
Great to hear that Intent has plans to mature. I agree that we should revive the discussion around creating a Mylyn repository that aggregates all components under the top-level project.
Does anyone else have thoughts on the structure of Mylyn Docs?
Steffen
_______________________________________________