Well, my concern is pretty simple; being limited to three basic character builds feels needlessly restrictive in the context of the rest of the game, and my strong kneejerk is that it'll make for a boring system.
As for why, it took me a while to find the right way to explain it that would make sense at a more intellectual level. So... okay. To start with, as I alluded to, if we take these as strict rules to follow for character design, we've essentially made a system in which one set of characters are only useful in short and furious battles, another set only for long, epic battles, and a third that get to pick when to blow their wad. Unless we give characters a lot of out-of-archetype moves to round things out, it boils down to half the cast being, in any given fight, dead weight. Since we're not making an SRPG (where all three of these situations would routinely show up in every conflict!), that strikes me as entirely boring.
So, even if you are really dedicated to a simple, easily workable system and focusing on enemy design/character balance to make things interesting, I don't think this is the right system.
More broadly, I don't think simple, easily workable archetypes are the right move for this game at all. This is more complicated.
It's pretty well a given that we're going to have predefined characters and skills. At no point in this topic have people expressed the slightest interest in designing a gameplay system akin to Materia, nor the use of a chanable Job system, or anything of the sort. While it's not been decided what form, precisely, learning skills will take, the desire for each character to do so in a different way, without simple level progression, also kinda rules out things like a skill tree or the like. In summary, we are not going to have a lot of character customization. It's still possible we might have robust equipment options (that's something that really hasn't been discussed yet, and is for another topic), so there may be twinking, but real decision making in the progress of our characters is going to be at the design end, not the player.
What that means is, we've already closed off an avenue of player choice in the battle system. This isn't to say it's the wrong decision; on the whole, the DL has long valued character uniqueness, and such a system detracts from it. However, because we have made this design decision, we need to be sure to put the player's ability to make meaningful choices and play around with the system elsewhere. Thusly, if we also have a limited pool of characters to choose from, and force those characters to conform to a very limited number of base character builds, we've basically designed a system in which the player makes no gameplay choices at all and can work out how to use each character within five minutes. This is a terrible decision. While gameplay choices can be given by expanding the cast from the bare minimum, and perhaps by giving opportunities to mix the various casts/having significant joined cast sections, if each character falls into a minor variation on a theme this is not terribly meaningful.
However, if we put effort into making characters highly different from each other, each with their own quirks related to fighting style, how they learn skills, and how to best play characters off each other and make use of every aspect of the system, this choice becomes one requiring careful consideration. I think the best example of what I mean is, funnily enough, Meeple Fantasy 6. Each character has their niche, and a lot of time was spent to make sure that everyone always had times to shine and tools to make them worth considering, even if the game wasn't perfectly balanced. But since MF6 is bound by the mechanics and rules of FFVI, and we're literally making things up, we can do even better!
Oddly, though, as I think on it I wonder if we're making the right decision including as strong of SRPG elements as we are. While having a mid-sized battle field, something expanded from the WA4 model, is fine in principle, I'm not sure if it'll work correctly in practice. I mean, we're looking at what, 10 minutes per battle on average, just due to positioning? Sure, this is short of what you'd see in FFT or WAXF or a lot of full-blooded SRPGs, but... they have 50-60 battles in the entire game. How many are we gonna have? 100? 200, 500? That's gonna really wear people down.
I could be overthinking this one, but it's something to consider.
Opposed to plot branching. Quite apart from personal views on games as a storytelling medium in the first place... at present we seem to be implicitly agreed that Andy's doing the real story work, moderating his ideas with suggestions from the peanut gallery. And while he indicates he likes the idea and is willing to do the work... well. At minimum, something like this would neccessitate writing multiple versions of every scene with a conversation branch. In other words, writing 2, 3, whatever versions of a single scene repeatedly. I mean, I imagine trying to do something like that and I know I'd really get bored with it, lose focus, and quality would suffer, y'know?
And at most, of course, we essentially write multiple games with the same premise, which is bound to either take ages upon ages or get a lot sloppier than a single coherent story.
Now... if we had multiple writers, yeah, a lot of the fatigue would probably be avoidable and, broken up correctly, we could essentially have people writing different views of things pretty effectively. But I'm kinda skeptical on multiple people volunteering for such a thing, to say the least.
