I could go on but it will fall on deaf ears, I am not a game designer, and I'm sure game developers are busy enough as it is. The point I was trying to make is there are various shades you could add to each quest to differentiate them. Does every quest have to be unique? No, but most quest shouldn't fall into the 1-2 dimension category. The wheel doesn't have to be re-invented, but use every wheel that has been invented not just three brands or types.
Thing is, all you've really touched upon are derivatives of the "Kill X, Fetch Y, Go To Z, and Defend/Escort Q" dynamic. Whether it's talking to an NPC directly, overhearing them, or basically stepping into an "event" you're including a bare bones level of the Go To Z element. Quests started by finding a random drop included a mix of the Kill X and Fetch Y unless maybe it's a random interactable in the world (Thus, Go To Z) or harvested from a node. As well, I'd be exceptionally cautious of what you basically outlined in a reputation system. These do exist in MMOs today, perhaps not to the intricacy you hope for, but at the same time, knowing how people play these games, when it's determined the rewards of Faction A are better than those of Faction B, you will never see people do Faction B if it's an either/or situation. From the dev perspective, I also dislike the notion of deliberately gating people from content you've created because it risks eliminating a percentage of Things To Do(tm) with their playing.
I've spent a fair amount of time contemplating how I'd do things had I the means to put a game out there. While I wouldn't call my ideas superior, innovative, or likely for the hardcore, I know one thing I'd definitely like to emphasize is the players having the ability to affect the world in tangible ways. For example, I'd like to establish a good monster eco-system where there's a predator/prey relationship between mobs. If, for example, a certain prey's predator is killed so much within a given span, you'll see more of that prey spawn while less of the predator. On the other hand, if the prey is over-hunted, you'll see them less often while the predators might start popping up in other nearby locations taking on different tastes and potentially influencing that little eco-bubble. I'm on the fence with this possibly affecting harvesting potential, but it's something I haven't drowned out. Either way, the hope would be to create an ebb and flow. I also entertained the thought of actual mob extinction, but in the end I think I take more to the thought of there always being a place where, say, wolves would lurk and could always have a chance to expand their territory and take everything back to "normal" if player intervention was minimal or non-existent. This is also perhaps a point where it's could to simplify crafting materials, like instead of having red, black, and white wolf furs, you'd just have a single wolf fur drop all sub-species would share.
Anyway, this could actually apply the holds NPCs have on areas, too. While I'd probably never make a PvP game for reasons of balancing difficulty, weighing hadcore contribution against casual to avoid that feeling of helplessness, and so on, the thought of players doing things to help an NPC "hold" an area, and thus allowing access to a dungeon, new/special NPCs, and so on does appeal. And while GW2 tried this to a degree, I feel like it was both fleeting and not really... epic enough. You could even tie this into the monster eco-system where if a particular region is suffering from a high pest level, then crops are either more expensive or not available at all from vendors. Overall, the world shouldn't be terribly static and "nature" would always be on the offensive trying to reclaim things toward what one could probably call a launch state.
I also realize this doesn't deviate much from the hated quest tropes, but if done right, you could have no two days of doing "quests" in a particular area being the same. I quoted that because I would indeed like to see FATE-like things going on "out in the field" that could tie into this twisted web of progression and regression.
I'm also somewhat quirky in the fact I wouldn't have our characters run off a general EXP level, but instead have the ability to learn, improve, and later modify abilities. Just to crudely correlate to XI here. Let's say you fight with a sword enough to learn Fast Blade. You also take the time to learn Fire and Banish. With all three "mastered" you could then learn Fire Blade and Banish Blade. With those then capped, you could learn Fusion Blade. Yeah, this basically shifts EXP away from the concept of character to skills, but I've always hated the the aspect of XI and XIV where you "forget" things you've learned just because you talked to a moogle or put on a different weapon. As well, I'd probably tie equipment restrictions to masteries. Like, you couldn't use a 2nd tier sword until you're level 2 in Fast Blade. Stuff like that. Given enough time, someone could potentially master everything, sure, but if MMOs have taught me anything, you have some people who will never
touch certain classes for whatever reason. And that's only toxic when they turn around and insist their particular preferred class be top dog and those who invest in others should be deliberately weaker. Anyway, I know this system would need a few more refinements, but my main goal is striving for that good mix of character (skills, gear, crafts) and world growth. I'd also probably try and keep the stories more modular, avoiding END OF TEH WORLD tropes because of the whole hero meets errand boy mis-match that tends to happen with the curious absence of reference to those other players who help you do whatever. So while there could come a point where you'd slay a dragon to protect whatever, you'd never be revered as THE dragonslayer, but more pragmatically amongst those who helped do the deed.