Subject: a meta-comment
I am not concerned (or less than before :- ) about the "risk" that resolving a future issue may undo parts of a solution to a current issue. Taking an example: i019 and i028 have a proposal on the table that - I believe - works. It could be that the solution to a future issue calls for a consolidation that may undo parts of what has been accepted before (as long as all issues are still addressed) - we just do not know, and cannot know today.
I consider this as inherent to the incremental nature of the way we work. It may not look optimal, but its just part of the process. There is no way we can be sure that all we design today is final, as long as the set of issues itself is still not final...