MMOs as they exist today bear the curse of DikuMUD unto the seventh generation. It's interminable iterations on "Kill X of Y" until you get to the endgame, where legend has it that the good stuff accumulates in a paper-thin layer that can be scraped off of the walls of dark, dank, dreary, drama-infested caves.
Or, I mean, that's what it looks like to an outsider.
And not that there's anything wrong with that! Sometimes all you really want is to kill X of Y. But let's assume that those who wish to grind will find a way to do so regardless of whether it's what the game is optimized for. How do you take the good stuff, the bits of planning and strategy and cooperation and arcane combination of rare ingredients that keeps people playing long after the novelty of killing boars has worn off, and give it to them throughout and from the beginning, without the pretense that killing boars was a novelty to begin with?
So here's what I'm thinking. Take the resources that go into making a vast world full of quest hooks that nobody ever reads, and epic bosses that 90% of your subscribers will not have the patience to see, and colorful animals that can be safely ignored once you've determined the amount of HP and mana you must lose in order to kill one, and put it all into making bosses. The bulk of the gameplay must be in unique, tactically interesting encounters, balanced for a variety of team sizes. The first time you kill it, you get EXP. Once per day, you can kill it to get one of its good drops (killing it more than once in a day does not yield additional rewards). Every day after the first time, you get one of its bad drops automatically. Drops are ingredients which NPCs or players of various profession can combine to make the various items that influence advancement in other ways, like gear.
Basically, if a summary of what the player is doing resembles a quest in Progress Quest, it should never be in a real game.