I wouldn't say that a complete abandonment of PvE would be a good idea for any game. Very few MMO players want to PvP *all* the time, non-stop, and I think this is why totally PvP games like WW2O or Planetside have limited appeal. Many like it as a sideline, to greater or lesser degrees. And most of those want the ability to say "I am *not* getting ganked today, I'll just whack mobs."
I agree to an extent. In my ideal mmorpg most of the PvE happens "somewhere else" even if there are parts of it strongly tied with PvP. But at the same time I love what happened in WoW. I love how PvE melts with PvP in a cohesive world and I love a lot more this model than the one in DAoC where the PvP is always "external" and all the lore revolves around an "invasion" that never actually happens.
There are parts of PvE that are necessary to give PvP a depth. Even DAoC could have used more development in the scope and ambition of its PvP by adding to it more activites that aren't just fighting and killing all the time. So a simulation of other mechanics that are integrated with the combat. But not just combat.
These games with a "world" component should start to put aside the full focus on just the combat mechanics and explore new forms of interaction.
So I would say that what I would be a proponent of would be shorter treadmills, more use of AI-based content creation tools for the filler, and where content is being hand-built that content is high-quality, well thought out, and highly polished. Spending a great deal of time and effort building minor variations on what you've already got, in order to fill out an extended treadmill, not my idea of a great plan.
I strongly believe that should be the content to define the treadmill, and not the opposite. So firstly you look at what you have available and then figure out the pace it can support. More content available = possibility to stretch the treadmill. I mean, first you plan the content before you even start to think at the container (which is exactly what even
Jeff Freeman wrote). And I believe that WoW reached a perfect balance here, with content in excess to even support a degree of variation for the alts.
Instead I completely disagree on the AI. I consider that path completely pointless and irrelevant for this genre. Mmorpgs do not need to implement AI. A slightly more scripted system could already do EVERYTHING you need and look like generations above what is already available. It would be already a dream to have some complex encounters where the monsters follow simple schedules and basic tactics that can be learnt. The fact is that here I have my own strong theory.
PvE is "authorship". It's the possibility to tell a story imagined by someone. It's a direct form of communication that goes from an "artist" to an audience. This is why it's stupid to suppose that your players will create the content themselves. There's a need of quality here, something that only a dedicated writer paid to do so will be able to deliver. So, as in a book, a movie, or similar things, all the control must be in the hands of the dev. Good PvE will never be randomly generated and will never be based on AI. Good PvE needs history, it needs personality, persistence. It cannot be contingent, it cannot vary, it cannot adapt. Everything that is vaguely relative is going to fail in PvE because what the players want is something fixed, something that has an interesting story to tell. In that precise point in the space and in that precise moment. It's the concept of "identity" opposed to "contingent". Identity as something that cannot change and that cannot be elsewhere.
This is why the players love when a zone isn't built randomly as in SWG but carefully handcrafted in all its details. This is why they learn how to pull, how each single encounter works and must be tackled. It's all about mini-puzzles, or the act of "chunking" as Raph Koster would define it. If the lesson to learn goes directly out of control and isn't predictable to an extent, it will become simply frustrating and look absolutely generic, a-specific, relative, not tied to the history and life of THAT particular place.
(
Reference)
Then if you add in that your PvP game requires your players to be within shouting distance of the top of the treadmill, telling large numbers of them that already achieved that stage and are now happily PvP'ing that they have to climb another treadmill to stay competitive is even less my idea of a great plan. Especially since it may give them an impression of a nightmare "infinite hallway", feeling that every time they do reach the end, you're just going to move the goalposts again.
Here I can be an heretic but I disagree on an even more unpopular topic. ToA didn't fail because it had more required content to go through. Players LOVE content and they LOVE when you give them something more to advance, new places to discover and so on. EverQuest was successful for a reason and this shouldn't be ignored (as Anyuzer wrote in his critics to you). An enriching experience is ALWAYS positive, no matter how long it is. Again it's not the "length" of quantity of the treadmill to be the REAL issues, but its quality.
The real problem of ToA was and still is today its *accessibility*. And the same happens with the criticized endgame in WoW. What the players refuse and push back is exactly what they cannot effectively experience. Both the problem of accessibility and the boringness of repetition.
ALL of ToA content is planned to not be easily accessible. Often through insane grinds that are completely unacceptable. The point isn't that the players don't want to go through ToA to be able to compete again in PvP, the point is that going through that is way too hard, demanding and boring. It's not accessible for everyone. It's a process of selection that not everyone can endure. So, again, not always accessible and desirable.