I'm trying to be pragmatic more than anything. There would be an insane amount of resistance to shaking up the existing formula completely even if it was something that was objectively better in every way. Reworking the game like this, while satisfying to people like us, would be a massive turn off to people who are invested in the game as it stands. I genuinely don't think it's a question of whether things like this should split off or the game should be reworked. I think at this point it's a question of splitting it in some way or having no change at all because it would cause too many problems. The exception being waiting until XIV ends completely and they next MMO Final Fantasy is produced with a new system from the ground up.

I also agree that focus on accessibility shouldn't be a sole focus and that gamers are more than capable of learning and adapting. However by that same token don't underestimate how resistant to any kind of learning curve some people can be. Square doesn't want the money of just the people who are capable of and want to figure out new or complex systems. They want everyone's money. Complexity will inherently and unfortunately alienate people and that is the last thing a business, especially in the modern world, wants to do. The only reason I even think an idea like this is viable at all is because it would please a large number of people with different opinions on how the game should be while simultaniously solving a growing problem that is itself causing player disengagement.

Finally this idea addresses the disengagement problem many players have while limiting the actual workload. Reworking the game for a new system would require reworking just that; the entire game. By intergrating a solo mode into the game as it stands you're limiting the increased workload to creating the solo variation of the class itself instead of the whole game. This is a much simpler task as it's only balanced against itself in the first place which reduces the amount of consideration that needs to be done. Truthfully and sadly, though, if anything would shotgun this idea down it's that even implementing this solo idea would require a massive increase in playtesting and thus time/cost as content would now need to be playtested not just against one class but a class with a large amount of build diversity. Not for balance purposes, mind you, but for bugs and exploits. The only upside there is, again, you would be playtesting based around the solo varient of the class and not the content as a whole.

An idea like this is still not perfect and I'm not going to pretend it wouldn't take a lot of work to implement. Still I hope the precedent set by Blue Mage can be followed up on in a positive way. A lot of this will depend on whether or not Blue Mage is even going to be done well in the first place and... I have my doubts. That is beside the point though. I'm simply hoping for the best and presenting ideas using the concept as a jumping off point.