Hi,
Dual dialogue has never been implemented, so it can't be described as having been "yanked out".
Many things in the world are "useful", yet do not exist because on the balance their benefits do not outweigh the costs. Dual dialogue is one of them. It would take a long time to implement and make the program's code much more complex and harder to maintain. So it would lead to fewer other new features, more bugs, make it harder for new developers to join the project, and make ongoing maintenance much harder. The benefit would be that an extremely rarely used feature would be supported. So, in the end, it's an easy decision not to do it.
"Casual users", as you call them, don't have to weigh these tradeoffs, so of course they want every imaginable feature: why wouldn't they? We as the developers have the unenviable task of having to decide where to allocate our meager resources to get the most bang for the buck, which in the end comes down to saying "no" to 90% of feature requests.
--
Osku Salerma