An Idea I have been investigating (for a long time) is the use of 'bubble' scenarios, where the quest/missions lead the player(s) to their own section of the map (server bubble) which has other (non-party) players blocked from entering (and interfering).

In what ways does this idea differ from regular "instancing" (having dungeons/areas that only you and your party are in)?

So I think the future of MMO's need to give you a starting area that basically is filling out an online "friend" profile based on the decisions you make. Take the online dating profile questions (a subset anyway ), and hide them in the choices you make at the start of the game. Then based on how you played the starting area (filled out your questionnaire), the game will find the best matches for you when you want to group up.

You'll have to be careful not to just surround people with a bunch of people who think identically to them. I wouldn't want to be surrounded by a bunch of 'yes men', and if I was surrounded by alot of people like myself, I'd stop playing. I'd annoy myself way too much.

Alot of enjoyability in real-world relationships come from each others' differences, not just their similarities. Differences that compliment each others' strengths or weaknesses (not talking RPG strengths/weaknesses, but real-life human ones). A great deal of enjoyability in relationships is our conflicting ideas and what mindsets and experiences and the lens we look through that we can share with each other.

Yes, this forces people to not be shallow jerks and to actual put real effort into their relationships, but the alternative is blanket conformance and just following the pack leader who is the loudest or most-popular.

Yes, this forces people to not be shallow jerks

Have you played a recent MMO lately? What I speak of would more be about play style. People of all personalities can still enjoy the same play style. Are you a run and gun sort of guy or do you enjoy solving a puzzle? Hell, the quests could also be taylored to your groups play style preference. Quests can generally be defined different ways.

Alot of enjoyability in real-world relationships come from each others' differences, not just their similarities.

There is almost always something similar that brought 2 people together. I'd venture to say often times in MMO's you are playing with many people where possibly nothing could bring you together.

Are you against online dating then because it's pretty popular and proven to be useful.

Have you played a recent MMO lately? What I speak of would more be about play style. People of all personalities can still enjoy the same play style. Are you a run and gun sort of guy or do you enjoy solving a puzzle? Hell, the quests could also be taylored to your groups play style preference. Quests can generally be defined different ways.

Ah, I thought you were talking about personality types. I haven't gone to any online dating websites lately, or ever, so I assumed they matched by personalities.

If you're talking about the Bartle's Test, or something similar, then that makes much more sense.

[Hearts, Clubs, Diamonds, Spades: Players Who Suit Muds]

(This is Bartle's revised 8-playtype test that adds an additional axis)

The accuracy or relevance of Bartle's archetypes are debatable, but it provides a nice starting point for discussion on the subject of gamer playstyles.

Others have ran with the idea, and have taken it further. [Personality And Play Styles: A Unified Model]

There is also a definite overlap between personalities and play styles, though, and you also run into some of the same problems I mentioned before. Having three players who are all "explorer"-type players is fun some of the time, but the different gamer-types can be used to reinforce and compliment each other's gaming experience. To give a class-based analogy: Miners mine ore, blacksmiths make the armor, enchanters enchant it, and the warrior wears it. If the Miners only hang out with miners, they can't give ore to the smiths. If the smiths and enchanters don't hang out, how will the armor get enchanted?

Bartle* theorizes that the different player types increase or decrease the enjoyability of other player-types, and that the increase of one player type can increase or decrease the community size by attracting or driving away the players of other types.

*For the record, I don't agree with everything Bartle says about MMOs; I feel I have to clarify that, because I point to his test alot in these type of discussions.

Grouping people by type is a great idea, but I feel like there'd need to be tools to keep people from being silo'd into a single camp, and that the game should also actively work to cross-pollinate a player with players from other types. People don't fit fully into a box - at least not in my family - so there would need to be precautions to keep an automated system from "boxing" players without them realizing it, into playing with people of a certain type and giving them a wrong impression of the type of MMO.

Definitely a great idea to explore further.

Administering a personality assessment within gameplay is an idea I've liked for years. :)

If you're talking about the Bartle's Test, or something similar, then that makes much more sense.

[Hearts, Clubs, Diamonds, Spades: Players Who Suit Muds]

(This is Bartle's revised 8-playtype test that adds an additional axis)

I like this graph. If I'm interpreting it correctly I guess I'd be halfway between planner and scientist. I'm not 100% sure I agree with the Keirsey type correspondence in the linked article, though I fit ok as a conqueror in this other graph. Conqueror is a strange term choice since it's not about conquest, but about completism and mastery of skills; 'conquering' the game's mechanics and content, not territory or opponents.

Grouping people by type is a great idea, but I feel like there'd need to be tools to keep people from being silo'd into a single camp, and that the game should also actively work to cross-pollinate a player with players from other types. People don't fit fully into a box - at least not in my family - so there would need to be precautions to keep an automated system from "boxing" players without them realizing it, into playing with people of a certain type and giving them a wrong impression of the type of MMO.

Definitely a great idea to explore further.

Would it really matter if players are getting boxed as long as the data supports that the majority of players like playing with the type they like playing with and not at all a certain other type? I mean they wouldn't really know anyway. Right now when you get auto grouped we never know what kind of experience we'll have and it's a lot more hit and miss, which over time kills our experience and makes us frustrated. Often to the point of quitting the game.

Would knowing that this algo is happening be the thing that ruins it for someone? If so, then I think it's fine to not leak what's happening. As far the player is concerned they are having more pleasant experiences, which in the end is all that matters. After all this is just a game, not a government takeover

I think this added with a "I'd like to group with this person again" button that remembers people who want to group with each other to increase the odds that you group with each other more often if you both agree. Yes, there is a Add Friend in almost every MMO, but people are complex and just because I enjoyed one run with a person doesn't mean I want to be friends with them. I don't even really know them yet. Like real world relationships, friendships take time to progress and that progression in games generally means seeing the player a bunch of times and having light conversations while clearing dungeons. People are lazy and trying to change people isn't going to work. Instead the system needs to nudge them along and make making friends easier. Systems in games can do this if they are smart.

I look at it this way. If the numbers show that play style A enjoys play style B 75% of the time and play style C 25% of the time, then match up A with B. Yeah, there could be some possible friendships to be had in C, but the numbers are too low that it doesn't make sense in a game to match them up just for such a small possibility. That option should be a last resort. It's all about the odds. Using analytic's to better the odds of grouping people successfully. It's not perfect, but then again it's a video game

Honestly, a game might even give the players the option to use this smart system or to just be random. Let the players decide which they enjoy more. Who am I to tell someone they should look for in-game friends that are different from them? If that's what they want, then so be it. We are making video games here, not trying to teach life lessons.

Would it really matter if players are getting boxed as long as the data supports that the majority of players like playing with the type they like playing with and not at all a certain other type? I mean they wouldn't really know anyway.

Good point. I just feel intuitively that with a large group of players, they won't fit as perfectly into pigeonholes as we might hope and I'd worry about putting a large percentage of players into the wrong group. Take, for example, the United States armed forces - a huge dictatorship (what they say is law for people within it) of an organization (and they have to be very organized, and classify everyone into some fixed category or another). A friend joined a branch of the military early this year, and finished boot camp, and while the armed forces were doing background checks on him, they were really confused, because he didn't fit their pre-planned pigeonholes. He mentioned later, "That's no big deal, nobody fits their templates perfectly."

There's a famous thought-experiment showing that you can't categorize everything, there are always things that don't fit perfectly into your categories no matter how hard you try (even when your categories are as simple as "A" and "!A").

Right now when you get auto grouped we never know what kind of experience we'll have and it's a lot more hit and miss, which over time kills our experience and makes us frustrated. Often to the point of quitting the game.

I didn't realize it was that big of a deal - I haven't played an MMO in quite a long time. But that makes sense - a similar thing occurs on a smaller scale in online FPSs (where my experience is also limited, but at least more recent): You join a server, if it's full of idiots (griefers, or cheaters, or people not coordinating well, or whatever) then you leave and join a different server. If that's full of idiots, then you continue repeating until you find a good server. The more 'repeats' you have to do, the poorer opinion you form of the community, and the more likely you'll quit the game permanently.

What's nice is when you find a server that you know is a great server, and you frequent it. Just like MMOs have guilds, you kinda "join" that server, either by unofficially frequenting it, or by joining the clan that runs it.

A game, especially an MMO, being able to "autodetect" what type of "server" (group of people) I'd enjoy, and invisibly steer me to it, is a fantastic idea. Just so long as if it turns out to actually not be a good match, the game doesn't lock me out of still trying to find a group on my own.

Would knowing that this algo is happening be the thing that ruins it for someone? If so, then I think it's fine to not leak what's happening. As far the player is concerned they are having more pleasant experiences, which in the end is all that matters. After all this is just a game, not a government takeover

I fully support MMOs being run as stasi dictatorships.
My problem is that algorithms or templates operating on real people rarely ever fit me or my family or my friends perfectly. They almost never do.

Algorithms occasionally make good suggestions but rarely give good directions, when operating on and trying to analyze humans.

Even their suggestions get easily confused:

My current Netflix suggestions. I'm not interested in any of those... but boy is that an eclectic set. Random suggestions could've had better results.

You like My Little Pony? Maybe you'd like Peep! -> Nope.

You like Lost? Maybe you'd like 'Once Upon a Time'! -> Uh, no.

You like Lady Killers? Maybe you'd like 'Bernie'! -> *sigh*, no.

You like Mission Impossible? Maybe you'd like 'Skyfall'! -> No...

You like Aristocats? Maybe you'd like 'House of Mouse'! -> Noperz. But if you'd instant-stream the Aladdin TV series, I'd be all over it.

You like Ghost in the Darkness? Maybe you'd like 'A Haunted House'! -> What? That's not even remotely related! Now you're just reaching...

I think this added with a "I'd like to group with this person again" button that remembers people who want to group with each other to increase the odds that you group with each other more often if you both agree. Yes, there is a Add Friend in almost every MMO, but people are complex and just because I enjoyed one run with a person doesn't mean I want to be friends with them. I don't even really know them yet. Like real world relationships, friendships take time to progress and that progression in games generally means seeing the player a bunch of times and having light conversations while clearing dungeons. People are lazy and trying to change people isn't going to work. Instead the system needs to nudge them along and make making friends easier. Systems in games can do this if they are smart.

Yes, that makes alot more sense than what I initially thought you were saying.
Giving players more tools, and making the game more intelligent to make recommendations, would be good.

I look at it this way. If the numbers show that play style A enjoys play style B 75% of the time and play style C 25% of the time, then match up A with B. Yeah, there could be some possible friendships to be had in C, but the numbers are too low that it doesn't make sense in a game to match them up just for such a small possibility.

Yes, that makes sense, as long as you are matching up cross-playstyles, and not just matching playstyle A with A, B with B, and C with C.

Take Team Fortress 2 or League of Legends, for example. People's different play styles compliment each other's. Not just by what class they pick (because people can pick a medic and then run straight into the enemy lines using the 'wrong' playstyle for that class), but because certain playstyles reinforce each other. Supporter-type playstyles (regardless of what class they are playing), more run-in-gunning playstyles, cautious tanked/armored damage-absorbing playstyles, and etc...

By matching players together by personalities or by playstyles, don't you run the risk of accidentally matching too many similar playstyles together rather than the diversity that is necessary for overcoming difficult dungeons?
And if you automatically steer them towards a specific pre-set template of diversity (one tank, one healer, one dps, one ranged, or whatever), that might reduce the likelihood of players coming up with their own class roles and playstyle combinations better suited to certain challenges or overall better suited to your specific game?

Developers can't accurately predict what the players will enjoy or dislike, and what the players will or won't do, or how the players will or won't play. All they can do is adjust their thinking after the game goes live, and continue tweaking the game as it continues to run.
If, even a few years into the game being live, you apply algorithms to steer players to certain other players or to certain playstyles, it might reduce or eliminate future change in playstyles that are better fits for your game.

Honestly, a game might even give the players the option to use this smart system or to just be random. Let the players decide which they enjoy more. Who am I to tell someone they should look for in-game friends that are different from them? If that's what they want, then so be it. We are making video games here, not trying to teach life lessons.

My point wasn't that we should force players to join up with people different from them because "it's good for them", but because I think that being surrounded by people that are too similar ends up being less enjoyable in the long-term.

Match me up with friendly people, match me up with people who might share similar interests, match me up with people in the same guild as me or people who I already know when possible...

I would much rather effort be put into cultivating a friendly community, matching jerks with jerks and friendlies with friendlies, instead of matching players by playstyle, level, class, gender, community reputation, or personality.

I don't like the divide between old players and new players - why can't a level 5 player run a dungeon with a level 92 and both be challenged and have a good time? Power (level or equipment) divides are artificial divides created and enforced by the game mechanics. By removing that divide and getting experienced players to play with inexperienced players, it might make the inexperienced players stay around longer and become long-term members. This is a divide that already exists in most MMOs that I'd like to see removed.

I don't like the idea of adding additional divides (invisible to players or not, enforced or just nudged/suggested/steered-towards). Community is strongest when there is diversity, in my opinion. Diversity in community will hopefully lead to longevity in how long your MMO stays profitable and enjoyable to a wider range or deeper niche of people.

But I do like the idea of adding a divide between those who contribute a net positive playing experience for others and those who contribute a net negative playing experience to others.

I like this graph. If I'm interpreting it correctly I guess I'd be halfway between planner and scientist. I'm not 100% sure I agree with the Keirsey type correspondence in the linked article, though I fit ok as a conqueror in this other graph. Conqueror is a strange term choice since it's not about conquest, but about completism and mastery of skills; 'conquering' the game's mechanics and content, not territory or opponents.

I'm partial to Bartle's cube-graph as well. I fall somewhere in the Scientist/Hacker areas, since I'm all about the game world and exploring the environments and the "natural laws" of the world.

I don't like the idea of adding additional divides (invisible to players or not, enforced or just nudged/suggested/steered-towards). Community is strongest when there is diversity, in my opinion.

We almost never see the person behind the keyboard and even someone who is like me or is a person that met my "likes" as a gamer, will have drastically different life experiences and seem very different than me. Of course race, gender, etc wouldn't be a consideration in any of this though.

I think it's a fallacy to think that just because someone meets certain high level traits that you wouldn't still have diversity and they would all be just like you. Nobody is even remotely close to anybody else, unless they are siblings perhaps. I have friends who are just like me in these high level traits but man are they still very different people with very different backgrounds and experiences.

I would love to trash levels and break down those divides in MMO's. That's the single biggest divide that exists and makes the pool of players to choose from so small. However, it's the cash cow for MMO's. It's how they make you play longer. We'd have to find another way that helps make the same amount of money by keeping players playing longer to get to that next level but doesn't severely reduce matching people up.

Grouping people by type is a great idea, but I feel like there'd need to be tools to keep people from being silo'd into a single camp, and that the game should also actively work to cross-pollinate a player with players from other types. People don't fit fully into a box - at least not in my family - so there would need to be precautions to keep an automated system from "boxing" players without them realizing it, into playing with people of a certain type and giving them a wrong impression of the type of MMO.

I think that Wizard101 has effectively implemented this concept. (Yes, I played it with my sons for a while there. It's not shabby.) The whole game is optional in parts. Excepting the early levels where the game mechanics are taught and practiced, the entire story can almost be ignored in favor of pet training, PvP arenas, crafting, farming, and house decorating. It's a "do what you feel like" kind of world. Then the birds-of-a-feather axiom happens naturally.

The idea to take from that game in regards to this discussion is that the better design for an MMO is to have the boxed playstyles in their own boxes while allowing the players to pick which box in which they wish to participate. It'll self-regulate that way and not seemed forced. I feel that it's the right way to go for a high-population environment.

I think there are a few different things we are talking about. Meatsack seems to be talking about playstyles like PvP, Crafting, Farming, etc. Servant seems to be talking about styles like griefers, planners, etc. I'm talking about personality traits like, realist, social, introvert, etc.

See I think you can have the same personality traits that can be across all of those other styles. An introvert can be a griefer or a planner. They can like PvP or crafting. But I'm not saying group all introverts together, I'm more saying find out what other types each player likes to play with and try to match them up with players of that type that like the other style too. So if person A is an introvert but likes to play with optimistic people, then try and group them with optimistic players who like playing with introverts. Now I use one trait as an example but the reality, I think, is that most people have a few traits that they enjoy in people more than other traits. Some people hate blindly optimistic people and it irritates then to no end to play a game with someone like that. So if they specify that in some game mechanic way (if possible) then the system needs to do it's best to avoid matching those players up. However, this doesn't limit diversity because people with the same traits all still have very different backgrounds and experiences. It will still be very enjoyable I think.

I didn't realize it was that big of a deal

I think it's a silent game killer. The thing is that MMO's are addictive in their own right that people put up with it longer than they would normally. I still say this is about trying to help make the best possible experience in the game, and the truth is grouping with a personality that you just hate can kill this experience for you. MMO's are all about the community. If your experience with the "community" (which is really just the players you run into in game, by pure luck a this point in time) ends up being bad, then you most likely won't keep playing the game, or you'll have bad things to say about the community, which give the game a bad rap, and over time can kill it. So how do you solve that problem? You present each player with the "community" (again in an mmo the community is really a small pool of players you happen to run into) that they will enjoy the most to give a higher chance (again it's all about odds) that they will have an enjoyable time.

If you ask each MMO how the community is for that MMO, you generally get bad feedback first all the time. Then they'll say, well not all people are like that, some are good. The good they speak of are the ones that they found that matched their preferred traits in a person. They lead off with the bad feedback first because luck/chance delivered them traits they don't enjoy grouping with more often than not most likely (again, odds. some people will have been lucky and have had more better traits or are more tolerable with others).

Again, this comes down to the whole online dating thing. It works because it increases your odds of finding someone that has the traits you enjoy. That doesn't mean they are just like you, it just means the traits you like and the traits you have matched with these other people giving you both a better chance to enjoy each others company. Doesn't that seem like a better way than to just go down to your local bar?

The cool part about a system like this is that you could allow players to change their preferences at any time. I think this could surprise some people in them finding out that they like certain traits that they didn't think they liked in people.