The amount of money that Square has spent on Bandaid fixes at this point has to be astronomical. Massively expanding wards has upfront hardware costs to keep these persistent housing districts up and running at all times. A lottery system was engineering time spent by developers, coders, and the UI team.
But... adding in an entirely new housing district only gives you 1000 or so houses, in a game that has exploded in popularity. The lottery system, in many cases for older housing districts, presented an 0.5% chance to win a home.
Both cost money and have costed money. Both cost and have costed development time. Both end up only applying a bandaid to the problem.
I struggle to think about why we haven't gotten instancing yet for the housing system. To my knowledge, aside from sandbox games where you're meant to compete for land, I don't know of a major MMO that has implemented a housing system and not had it instanced so everyone can have one. Heck, even EverQuest II had one back in 2004, and they were running on their own internal data centers in San Diego. Their system is also much more robust than Final Fantasies. That was coming on 20 years ago.
Major counterpoints i've already heard are:
1.) It is server intensive: I'm not sure I buy this, unless the netcode is just.. wonky. Again, a ten year old game managed this. Elder Scrolls Online manages this and lets your character own every house in the game. I'm not even suggesting you go that far. One instanced house per character shouldn't bloat the server's database too much, right?
2.) It hampers community interaction. And... eh... go to a housing ward, and I don't see neighbors talking to neighbors, people hanging out like a true "neighborhood" would allow. Their FC's will sometimes come out to the market board together, but that's non-inclusive interaction. Finding and visiting other houses to look around is also doable with instancing.
So.... why are we still here? Why have we not seen a reliable fix to this done? Why is anything other than a realistic, long term solution to what is coming on a ten year problem being green lit by management at CBU3? I'm genuinely curious.