I think I could make a case that documenting the design (at least at a high level) and getting a review of that should come between showing community support and all the work to start on the code. The point being that changing direction is easier the sooner you do it. This is especially true when it comes to issues of storage and scalability that are inherent in deploying something new into enwiki or commons. Unfortunately off the top of my head I don't have examples of such a thing having happened or the types of artifacts that would be generated to accomplish this.
User talk:Dan-nl/Developing extensions for Wikimedia wikis
It looks like this point is covered in Writing_an_extension_for_deployment. Maybe effort should be focused on updating that page and linking it from more pages?
I think the complete roadmap should include discussion of the post-deploy support plan as well. Part of the difficulty in getting a new Extension pushed out to the Wikimedia projects is getting commitment for long term maintenance support from someone. This is even a problem for work that is started from within the Foundation itself.
There are no older topics