As much as never-ending MMOs appeal to me as a consumer and a designer, some game designers speculate that while most adventure-based media intentionally or unintentionally mirror the Hero's Journey/Monomyth (at least partially), MMOs tend to take players three-quarters of the way through the Hero's Journey and then prevent them from finishing it, to hold players hostage for more revenue, which I think is very interesting food for thought.

Obviously some players may refuse to complete the monomyth cycle and live there, just as some mythological heroes do, but actively blocking them from leaving seems almost malicious, at least fridge-logically. Kinda like gambling: Playing blackjack or slots or poker is one thing, even for money, but casinos intentionally taking advantage of people with gambling addictions (or microtransaction addictions) is another matter.

Note: I don't fully agree with the Hero's Journey. I think it partially confuses cause and effect; also, the risk of thinking too much in monomythic terms is the same as the risk of thinking too much in trope terms: that of focusing so much on generalities that we lose sight of the distinctive attributes that makes individual games (and other media) poor or great. Nevertheless the monomyth is a brilliant insight into the nature of human literature (and thus, game design and plot), and that lens is worth peering through from time to time.

How can MMO game designers recognize when their MMO design is artificially holding players captive by false promises of entertainment that no longer fully arrive, verses giving players a satisfying conclusion and then releasing them to find a new game/adventure or releasing them to get back to their life in-general?

This applies to open world games, whether single-player or online; how can we provide players a satisfying "conclusion" to an open-world game, once they've had their fill?

How can we provide an exit-door in an otherwise dark theater, for when they are ready to leave?

Morrowind and the Elder Scrolls games in general had a 'main quest' in addition to the open world. Did you beat Morrowind's final boss first, then explore the world? Or explore the world before finishing off the main quest? Once you beat the main quest, did that give you a satisfying sense of conclusion? Did you put down the game afterward (or shortly afterward) and cease playing it?

Economic incentives aside, what do you think design-wise of providing a "main quest" in MMOs (and maybe a separate "main whole-guild quest") that players can beat when they are ready to be done with the MMO?

I don't know, I think these games do tend to have a conclusion if you're looking for one. MMOs tend to have main questlines with milestones when you reach max level or beat the highest tier raids, it's just that there's almost always more to do afterwards (and expansions to extend the storyline) if you want to keep playing. There is actually a sizable chunk of players that stop playing once they hit max level since they aren't interested in raiding or pvp.

Even ignoring economics, the nature of MMOs usually requires them to maintain a healthy population so players have other people to group with, fight, or otherwise interact with, so I don't think there's any incentive to provide a definitive "ending" so players would be encouraged to stop playing.

Additionally, I think story is just less important to most open world games. For most people, the main appeal is going to be the gameplay itself, and as long as there are new and fun things to do, there will be a reason to keep playing. (I for one never finished the main questline in Skyrim, still played it much more than most RPGs and had fun exploring the world) The real "end" for a lot of players is either when you get bored or when you can no longer improve your character because you've beaten all the dungeons and you have all the best gear.

Besides, just from a narrative perspective, it's harder to use a standard RPG main quest in an MMO: "You are the chosen hero who is destined to save the kingdom… except for those thousands of other people doing the same thing." Not to mention that you can't ever really defeat the bad guys for good because in MMOs mobs have to respawn and instances reset so you and other players can fight them again. MMO stories are more believable when they focus on your character's personal journey of self improvement, and that kind of story ties in well with leveling up.

The problem I've found with MMOs is they resemble the same mechanics -- this is the biggest bore for me when it comes to MMOs. You know exactly how it will play before playing. Is that good, or bad?!

In respect of story in MMOs, I've never considered how they construct the story (despite being a lover of good stories) or implement it as it always seemed a means to an end. And thinking now, not one MMO I've played sticks in my mind for story - let alone what story structure they use. Although oddly enough, open-world games (which can be quite similar to MMOs but single-player) tend to have stories that are memorable.

But saying that, there is one MMO that I'll never forget. Good memories, yes, but what stuck was the concept: Face of Mankind (FPS and no levelling). A universe of several factions (headed by a GM) that included a political system, law and order, factions for the economy, mercenaries and an underground lawless faction. The designers of the games and GMs generated events, so with that and the player-base, created the story.

Anyway, in respect of a main quests and such, the story could be (with thought) a very long story. Just consider, say, Game of Thrones. But however you implement a story, I think the big thing is making it a social event. Isn't one of the biggest draws to MMOs the ability to play with friends?

I'm not a huge fan of the hero's journey, but if you want to substitute the idea of the rags-to-riches bildungsroman, I'm a big fan of that in RPGs and MMOs. For a single-player RPG, I think the satisfying ending is to give the player a lot of praise/recognition/affection from NPCs and bonuses for completing various collections or achievements. Then, though the linear/exploratory gameplay is over, leave the player with a minigame or two that they may want to come back and toy with every few weeks, at least until the sequel of your game comes out. Minigame examples would include chocobo racing and other golden saucer games like snowboarding in FF7, Vasebreaker endless and I, Zombie endless in Plants vs. Zombies 1, really any good solitaire or match 3 game or a sim-game like Fish Tycoon/Plant Tycoon would work. But I'm also a big fan of time loop single player RPGs where the end of any playthrough is a good stopping point but you try to provide the player with enough rewards for them to complete at least a 2nd playthrough, and earn a better or at least different ending.

For a single-player RPG, I think the satisfying ending is to give the player a lot of praise/recognition/affection from RPGs and bonuses for completing various collections or achievements.

I'm kinda a fan of the "well, the whole world's in ruins now, but we survived, and enough of us survived that we can rebuild" type of endings - though not exclusively (examples that come to mind are Final Fantasy 7 and the anime Fullmetal Alchemist: Brotherhood).

But both your single-player example and the world-in-ruins example, are still conclusions - they are still 'exits' even if they offer more sidequest stuff and offer Game+ modes, there is still a clear moment of arrival at the exit door (very few linear singleplayer games don't offer that, and some offer more than one - though I have displeasure with the 'more than one' and 'none' games).

The problem I've found with MMOs is they resemble the same mechanics -- this is the biggest bore for me when it comes to MMOs. You know exactly how it will play before playing. Is that good, or bad?!
In respect of story in MMOs, I've never considered how they construct the story (despite being a lover of good stories) or implement it as it always seemed a means to an end. And thinking now, not one MMO I've played sticks in my mind for story - let alone what story structure they use. Although oddly enough, open-world games (which can be quite similar to MMOs but single-player) tend to have stories that are memorable.

But saying that, there is one MMO that I'll never forget. Good memories, yes, but what stuck was the concept: Face of Mankind (FPS and no levelling). A universe of several factions (headed by a GM) that included a political system, law and order, factions for the economy, mercenaries and an underground lawless faction. The designers of the games and GMs generated events, so with that and the player-base, created the story.

Anyway, in respect of a main quests and such, the story could be (with thought) a very long story. Just consider, say, Game of Thrones. But however you implement a story, I think the big thing is making it a social event. Isn't one of the biggest draws to MMOs the ability to play with friends?

You bring up some other interesting MMO flaws; they have more than enough flaws to keep us in discussion for years. ^_^

In referencing the Hero's Journey, I accidentally gave the impression that I'm talking about stories and plot, but I'm actually talking about player experience and post-game sanctification, even if no plot exists, or regardless of whether the plot is good or bad.

We do consume media for story, we do consume media for characters, but we also consume media for experiences. Exits help reinforce (and sometimes cap off) the experiences.

I don't know, I think these games do tend to have a conclusion if you're looking for one. MMOs tend to have main questlines with milestones when you reach max level or beat the highest tier raids, it's just that there's almost always more to do afterwards (and expansions to extend the storyline) if you want to keep playing. There is actually a sizable chunk of players that stop playing once they hit max level since they aren't interested in raiding or pvp.

The difference is, those milestones are, "Well, I'm bored now, time for something else!" - which is fine! But, it'd be great if players left the game not saying "I'm bored", but left saying, "That was incredible!".

I don't want MMOs to force players into an "ending scene" prematurely - the easiest way is players consciously choosing to trigger it. Like playing Chrono Trigger, a mostly linear RPG, it "ends" when you say, 'Time to go finally face Lavos'. Up until then, you're running around doing side-quests and leveling up.

Even ignoring economics, the nature of MMOs usually requires them to maintain a healthy population so players have other people to group with, fight, or otherwise interact with, so I don't think there's any incentive to provide a definitive "ending" so players would be encouraged to stop playing.

Yes, definitely huge numbers of players are critical for MMOs. Many designers have correctly stated that "players ARE the content" of MMOs (at least, part of the content).

However, players don't stay forever. Players will exit eventually (if only from the servers eventually being shutdown), if we can make it so when they do exit, it reinforces their memories of pleasure in that game, it helps that brand for when the sequel MMO comes out, or for other games by the same studio it'd enhance the brand. Further, by identifying the exit (through the player taking some concrete action in-game), it'd be the perfect time to wait a week and email them a link to another one of your games, or an offer to transition their account credits towards your Sci-Fi MMO instead of your fantasy one.
That's the business incentive - but not my goal; my goal in exploring this is design-wise.

Additionally, I think story is just less important to most open world games. For most people, the main appeal is going to be the gameplay itself, and as long as there are new and fun things to do, there will be a reason to keep playing. (I for one never finished the main questline in Skyrim, still played it much more than most RPGs and had fun exploring the world)

For me the main appeal is the world, and the exploration of it. I'm honestly not trying to push stories. The idea of the Hero's Journey goes beyond the plot of the book/game, and in a meta way, has been applied by some game designers to the process of players themselves playing the game (not merely the characters they are controlling).

I feel - and this is my personal view - most good media is strong in at least one of three areas: Story, World, or Characters. Ideally all three, but if you take Morrowind, for example, it's World was strong enough that it more than made up for the sub-par characters and story.

I also feel - and this is my personal view - that consumers have different preferences for those three things. We all, to some extent, want great worlds, great stories, and great characters, but some of us care more about great characters, and some care more about great stories. (and with games, there's also gameplay as well, and so on)

As for myself, I care most about the World aspect. I'm huge on atmosphere and exploration and immersion. That is what is most important to me (above gameplay, stories, characters, visuals, etc... - but sure I enjoy those also).

So when I'm talking about an 'exit' for MMOs, I'm actually not coming from a story-focused background. I'm thinking about the concept of 'exits', because I intuitively think it might enhance the 'World' aspect. I think it might enhance atmosphere and immersion, to a subtle extent.

The real "end" for a lot of players is either when you get bored or when you can no longer improve your character because you've beaten all the dungeons and you have all the best gear.

Absolutely! I just have this nagging question about whether providing some symbolic "exit" that players can choose to activate when they are ready, might provide an extra "oomph" of satisfying conclusion. But only when they are ready to activate it.

This doesn't have to be a boss fight, and doesn't have to be anything specific - it'd have to fit in with the theme of the game.

But imagine the theme of the game is that you are in a different world, but aren't native to it. What if "the exit" was a big stone archway that you walk through to "return" to "your world", or even not your original world, but one that leads to "the next great adventure". I mean, it's kinda cheesy when I say it like this, but I think in-context, in-atmosphere, in-world, it could work well and provide the difference between laying down a book you're reading that you're still reading, and the satisfying finality of laying down a book that you've just finished.

Or if the game is about being stranded on an alien planet with no way off, what if, unlocked sometime through playing and exploring the world, you find a way to finally signal activate a beacon so a ship comes by and rescues you, so you literally 'exit' the world?

My examples/suggestions are symbolic methods of "leaving" the "world", but it doesn't have to be so on-the-nose. It could be dying, or ascending to godhood, or starting a family, or whatever else. I mean, but bookends are nice too, which is why I think the symbolic method is appealing to me.

In Trespasser, a linear Jurrassic Park-licensed survival game, you play a women who crashes on the island (the Site B research island, if you know your Jurassic Park lore); after trying multiple methods of trying to signal for help, the game ends with you raising an antenna, making contact, with a rescue helicopter going to pick you up on a helipad on a small mountain, and you have to make your way up the mountain, hunted by velociraptors, before you reach the helipad and are flown off the island. It was a very satisfying ending. It was exciting to be introduced to that world, it was enjoyable to explore that new world, and (equally importantly) it was satisfying in the way you exited that world. I see the first two done well in MMOs (beginning and middle), but not so much the last one (exit).

Some people won't want to leave, and that is fine (I imagine the 'Cheers' themesong playing here). But for those who want to leave, and are ready to leave, how can we make their exit satisfying?
Sure, anyone can just log out and not come back - I'm not seeking to prevent that. Nor am I saying my "exit" would be required. Nor would I even want the "exit" to prevent the player from logging back in again - though I imagine it might be similar to Game+'ing or Prestiging your character.

The worst problem would be someone accidentally activating it before they are ready to be done with the game, but that's a communication issue that is case-specific that can be solved.

Have you ever watched a TV show or movie series that is really really great, but it goes on so long that, though you keep watching, the final season is (even just slightly) at a lower standard than the earlier seasons, and it psychologically mars even your previous enjoyment?

If someone is playing an openworld game (singleplayer or MMO), and what point can we help the player exit the world satisfyingly, immediately after she jumps the shark, while still at an euphoric high point? I mean, if she's going to leave anyway, but would otherwise play for another four or five hours of boringness before she realizes it's time to quit, I'd rather save the player those five hours, and enhance the lasting perception of enjoyment of my game, by ending on a high note. Ofcourse, I don't want to artificially exit a player while there's still net enjoyment to be had (net enjoyment meaning, more pleasure than boredom).

Ideally, I'd like to *accurately* know when it's time for a player to quit before she herself knows it - or at least as reasonably accurate as possible.

Besides, just from a narrative perspective, it's harder to use a standard RPG main quest in an MMO: "You are the chosen hero who is destined to save the kingdom… except for those thousands of other people doing the same thing." Not to mention that you can't ever really defeat the bad guys for good because in MMOs mobs have to respawn and instances reset so you and other players can fight them again. MMO stories are more believable when they focus on your character's personal journey of self improvement, and that kind of story ties in well with leveling up.

Not everybody, but alot of people, are tired of the whole "chosen one" cliche. You can have satisfying stories without making the player the chosen one. Nor am I suggesting that an MMO needs a fixed linear story that every player goes to. In my own design thoughts, I'm leaning in the direction that players kinda form their own story through what happens to them in-game (and Director-style AI brings 'experiences' to them, as well as adventure-quests they choose to engage in). My 'exit' suggestion is less story-related and more emotional-high/exhilaration/euphoria brought to a conclusion rather than keep trying to drag the player back in with promises of more pleasure (more expansions more DLC more content updates, etc...) if those promises can't actually be met for a specific player. There should still be updates, expansions, DLC, and a continuing-to-change non-static world (if we're talking MMO), but though the train goes on, some players want to get off at the current station, and I want it to be a pleasurable departure, and one that leaves them with a good overall taste of the game (and the company), rather than "It was great... for awhile, and then it got boring.". When it ceases to be great (which will vary between players), I want there to be a way to exit while the overall impression is still "It was incredible! /FULL STOP/" without any 'buts', asterisks, or qualifiers.

Does that make sense? Does that sound like a good idea to explore, or would it ruin MMO experiences?

There's really several separate aspects to it:
- Is it a good idea to make sure that when players do exit (which is inevitable), it's a satisfying exit?
- How can we determine when a player is ready to exit, so we can present the exit, before she continues playing beyond her point of enjoyment? (ideally, we want to encourage use of the exit before the game gets boring, but after he's had as much enjoyment as can be mined out of it)
- How can we present the exit in a way that players comprehend well?
- Is this actually beneficial for the developer?

So, I used to play a game called "shattered galaxy" all the time, it's an old mmortsrpg. The game has (It's still up!) persistent territory. The game's almost entirely PVP, however, there's a PVE mechanic that was never an issue where Aliens would come out of caves, take territory, and slowly expand.

This never happened in practice, because the game was hyper competitive, and alien invasions usually died within minutes.

However, as the game got older, and population got smaller, the aliens usually managed to hold a few pieces of territory (They got weaker as they got away from their "capital").

I logged in a few years ago, and the aliens owned most of the map, and the remaining players usually were trying to fight off the alien invasion, which dominated 80%+ of the surface.

A few weeks ago I spoke with a friend who logged in that day and found that Humanity was completely exterminated by aliens. Yes, the aliens have expanded over every single piece of terrain, making it virtually impossible to play. In other words, humanity lost.

Probably the most satisfying end to a game I've ever seen, really.

Actually, the game's still up running in a zombie state with no players/much of the functionality broken, but you can still log in, go to the main planet, and see that you have literally no options. Just a map that says "Alien territory #1-250"

The best part is, if I were to put it into a novel-like form, it'd probably be a great read. Different political structure occurred (Democratic/oligarchy/anarchy), each with their own interesting periods (For example, massive voting blocs formed during democracy, and the community rioted until the developers removed the ability to vote), and different factions rose and fall during multi-year power struggles. People banded together into a tight knit community sucking in new talent/training them/becoming friends with them, to the point where nuetral groups were set up purely to help out new players, then during human-induced drama massive wars that broke apart the status of the game occurred where people were ran out of the game and empires were crushed. Different groups had their time in the spotlight being top tier and dominating the world, but eventually all of them crumbled to dust with time to the enemy everyone ignored, the aliens.

The question is, if this was compacted to 1-2 years, would this ending even be satisfying? I'd think not. It's only neat because it's been almost 2 decades since the whole thing began.

>> - Is it a good idea to make sure that when players do exit (which is inevitable), it's a satisfying exit?

no harm in it.

>> - How can we determine when a player is ready to exit, so we can present the exit, before she continues playing beyond her point of enjoyment? (ideally, we want to encourage use of the exit before the game gets boring, but after he's had as much enjoyment as can be mined out of it)

nigh on impossible. its hard to say when they will get bored. you might assume they won't get bored until they run out of content, but that's not even a safe bet. they may not be into the content that remains, or they may not have even found it.

- How can we present the exit in a way that players comprehend well?

shouldn't be that hard. reminds me of Sol (Edward G Robinson in his last film) "going home" in the movie Soylent Green with Charelton Heston.

just be sure to make it obvious. i accidentally ended fallout 3 twice by completing the rather short main quest line. and the second time it was unintentional. i found myself at the final quest without ever actively pursuing it.

- Is this actually beneficial for the developer?

for the developer, no. its more work. and its not enough of selling point to significantly impact sales.

it seems you like closure. there's no harm in adding optional "closure" features to a game. but then again, in real life - the ultimate rpg - there's no closure, no exit for the bored (if there was i would have left this god forsaken rock long ago - i'm just doing time on planet earth). the only exit is suicide. so you keep playing til you drop or you're so bored / disgusted / without hope you cap yourself.

I can see the appeal of an ending, although that's not something that I personally look for in games. I think the biggest problem in designing this conclusion/ending mechanic is making it feel significant, but still allowing the player to come back later on (especially if there's an expansion to the game later.) MMOs really have an interest in keeping players playing, so I think any kind of ending is also going to have to leave the door open for players to come back, and probably even encourage some players to replay the game more.

Perhaps some sort of highscore entry could work? Each character gets one opportunity to "finish" at a point of their choosing and they see a cutscene and get a special in game title depednding on their achievements (lore wise, the idea of "finishing" could be the player retiring, or maybe it marks their graduation from training adventurer, maybe it's signified by receiving knighthood or induction in some special guild, or so on.) In any case, when a player chooses to finish, all their achievements are tallied and they can see how well they did according to various metrics: total wealth, all the dungeons they've beaten, PvP K/D ratio, and so on.

This way you could have a way to put a capstone on your character if you want to quit, but it would also encourage other kinds of players to make multiple alts: go for a highscore of fastest to max level, lowest level to beat a certain dungeon, make a pure crafting character, and so on. It could even use a roguelike inspired feature where each character that finishes would give the player's next character certain starting bonuses, or maybe some kind of account wide flair for your other characters.

A bit of a tangent, but as a side benefit to the roguelike idea above; one of the other problems this might address is that in MMOs that fall out of popularity, the population tends to get concentrated at max level endgame content, so players who pick up the game years after release don't have anyone to group with until they get to the endgame too. That was one of the problems I saw a lot of in Age of Conan where as the population dropped it became impossible to find groups for all but the highest tier dungeons; they eventually added some achievements but there was never really enough of an incentive for most players to collect them. Encouraging the hardcore players to properly play through low level content again would probably be good for the community in this case.

One option is to ease up restrictions on linear time. A mythic hero might have an overarching narrative that follows the classic journey structure, but around a campfire you were probably as likely to hear a tale of some side adventure or exploit. Thus although the narrative followed an arc, it's retelling was piecemeal with new details being added generations after the story was first told to "completion". It's harder to accomplish in an interconnected MMO, but you could take the Ragnarok approach and foretell a great cataclysm. At the end of the main quest the player is shown a prophecy of a coming endtime and there role in it. The journey is complete. But that's still in the future, so you can keep questing.

Another is serial heroes journeys. You often see this sort of approach in television. Problem arises =>hero sets out to improve himself and face down the problem => does so => all is well => new, greater problem arises. The observation here would be to avoid continuously ratcheting up the adventure, but instead build in happy lulls where the hero rejoins society and is praised. Depending on how you structure the story you could end the current narrative on a cliffhanger with the player perpetually improving until new content allows the narrative to push on. Each lull is a natural place for players to slip away satisfied.

You could also position the endgame as denouement. The victorious hero returns home to the praise of his people. He retires to an idyllic life of mindless dungeon raiding and guild politics. Nothing threatens the land anymore, the fruits of his labor are visible in the renaissance his homeland is experiencing. The player eventually logs out for the last time, perhaps not realizing it is the last time. I think treating the endgame as catharsis also fits the heroes journey arc.

A voluntary exit is an interesting alternative. I'd be curious to see how players reacted. I think some would love it, some would find the very option stressful, some would exit and regret the choice. You might also put the exit on a timer. After 4 months, your player exits the world. Depending on how well they did, this might be a glorious death in battle, or a happy death-while-sleeping amid family, or ascension to the plane of Gods. Some of the nice byproducts is that it would allow you to build a wider world (many mutually exclusive quests rather than a really long singular ones) and that it creates turnover at the top of the player pool: If you join years into the game, you still have the opportunity for your day in the sun where you are among the strongest characters.

Another is serial heroes journeys. You often see this sort of approach in television. Problem arises =>hero sets out to improve himself and face down the problem => does so => all is well => new, greater problem arises. The observation here would be to avoid continuously ratcheting up the adventure, but instead build in happy lulls where the hero rejoins society and is praised.

That's really interesting. "Seasons" of gameplay, just like anime has seasons which are usually stand-alone arcs. That'd fit in well with a 4-month expansion release cycle.

That's going in a different direction than I was originally thinking of (because it'd affect the entire population of the MMO at the same time), but an interesting idea to explore on its own. Some MMOs do that a bit already, but others could definitely go further in that direction.

For it only effecting a single player within an MMO, player 'experiences' could be triggered and managed by Director AI, which is something I've thought about in the past that would work well with this.

It's harder to accomplish in an interconnected MMO, but you could take the Ragnarok approach and foretell a great cataclysm. At the end of the main quest the player is shown a prophecy of a coming endtime and there role in it. The journey is complete. But that's still in the future, so you can keep questing.

Thematically, that'd be great. The player could even, when ready, "embrace" the prophecy, and 'enter into it' (fast-forwarding time to that day, with the player avatar having aged 20 years) and participate in some final end-game content.
Or the same can be done without a great "chosen-one" battle, and show the player with children who he raises in a series of montages, ending with him dying peacefully in bed (as you mentioned elsewhere). Some roguelikes even do the thing where you can then start over, but as one of your now-adult kids.

You could also position the endgame as denouement. The victorious hero returns home to the praise of his people. He retires to an idyllic life of mindless dungeon raiding and guild politics. Nothing threatens the land anymore, the fruits of his labor are visible in the renaissance his homeland is experiencing. The player eventually logs out for the last time, perhaps not realizing it is the last time. I think treating the endgame as catharsis also fits the heroes journey arc.

It'd be interesting to figure out how to provide and encourage catharsis through post-game events. That'd be really interesting.

You might also put the exit on a timer. After 4 months, your player exits the world.

Some players (probably of the roguelike mindset or competitive-gaming scenes) might enjoy that, but that wouldn't be my personal cup of tea.

I'd rather go perma-death than timer and scoring, personally.

Unless it was more like Legend of Zelda: Majora's Mask, where chronic partial-resets of the world were an integral part of the game. That'd be interesting - the entire gameworld resets (for all players) every 45 days, with different events (and different endings) happening based on player actions, and players' progression mostly doesn't reset, just the world itself and quest and dungeon progression.

A voluntary exit is an interesting alternative. I'd be curious to see how players reacted. I think some would love it, some would find the very option stressful, some would exit and regret the choice.

Player stress over the decision is an interesting aspect I hadn't considered.

As for exiting and regretting it, that's a problem there are ways of handling, I'm just not sure of the best way.
First, it's important that the decision to exit is taken intentionally - so communicating it is clear. But communication and theming aside, I see two mechanic-wise effects for it:

A) The 'exit' could not be final gameplay-wise, just emotionally 'final' for the player, with nothing preventing him from logging back in.
B) It could be similar to prestiging / New Game+'ing, which would actually work well in some aspects of my MMO design* - you have (randomized) genetic character traits, and so NewGaming could let you carry over your existing traits, and get additional random traits added on. I had forgotten (it being before my time) that some older MUDs used to do this (calling it 'remorting').

*I'm not currently making an MMO. I just have folders of design notes for one that I'd like to make in ten or so years.

If it goes in the New Game+ route, then it could even unlock new abilities, access to special guilds or in-town lounge areas, and so on.
I wouldn't want to unlock new dungeons (unless they are very end-game dungeons), because I don't want to divide the playerbase and confuse players, and I wouldn't want to unlock new classes, because I'd rather players have access to whatever class they want, right from the get-go, but I'm fine with unlocking skills or, even having each class have an extra hardcore variation of that class that could be unlocked, with tweaked skill trees, slightly tweaked playstyles, and extra difficulty.

One thing that the Etrian Odyssey 4 single-player RPG does is let players choose a second class (so they are dual-classed), after a certain point in-game. I could do the same, but only unlock the dual-classing after the player prestiges.