I like emergent play. I enjoy games in which surprising things can happen when different aspects of complex systems bump into each other.
I believe there can also be commercial value in games whose content emerges from system-interactions. Not only is interacting with such content fun for gamers like me, it's also valuable to smaller programming teams who simply can't crank out the vast amount of content needed by a game with pre-determined events and narrative.
But having this long-standing interest in emergent-content play means that I can also see some of the potential difficulties of this kind of play. There are benefits to designing a game around emergent play. But there are risks and gotchas that need to be understood and addressed as well.
1. Emergent content is not for everyone.
Not everyone likes emergent play. A lot of people --maybe most people -- playing games today prefer well-understood rules and outcomes. They don't want to be surprised.
For example, consider the players who think of "crafting" in a MMORPG as manufacturing lots of identical widgets to compete in a sales game. If you implement crafting to have emergence, you're saying that you won't always know exactly what you're going to get... but that's pure evil to the manufacturing/sales-oriented player because it means that they're "losing" resources every time something gets made that isn't exactly what they expected.
Another example is Minecraft. Minecraft certainly has emergence -- the time my niece rode a pig into a lava stream and set both of them on fire wasn't something I ever thought I'd see. So Minecraft is great fun for people (like me) who enjoy being surprised. It's also fun for the gamers who enjoy the sensation-oriented survival challenge. What it isn't so much fun for are the gamers who prefer clearly rules-based games with clear win conditions. These are the players who, since Minecraft launched, have expressed unhappiness that "I don't know what I'm supposed to do" and wanted things like character levels and "adventure mode" rules-based play. As Minecraft's developers have added those, Minecraft has now been said to have passed WoW in terms of total revenue... but it might not have done so without understanding that emergent play alone would not be enough to satisfy the gamers who like clear rules and win conditions.
It's OK to make games with emergence. But it's important to recognize that by doing so, you're limiting the audience for your game. If you're fine with that, awesome; if you think that emergence by itself will make your game broadly popular, though, that might need a re-think.
2. Emergent content is hard to balance.
Emergence is optimal for exploratory play. It can also be good for cooperative play. It's not good for competitive play.
Competitive play demands fairness, or at least the perception of fairness. Emergence works against that because it allows one player to experience content that another player probably won't. Emergent content lets one player "get stuff" that another can't. That creates a perception of unfairness, and that snuffs out any willingness to compete.
So it's important to understand that if you're determined to make a competitive game, then you either should not try to include emergent content, or at least be extremely careful in how you include it. If the world can generate things that one player might get that another might not, unbalancing the level of challenge for different players, then the typical competitive human player will probably find that extremely annoying.
There are ways to address this; the important thing is to recognize that it's a potential issue. Then you can think carefully about how to help your players feel they're playing a fair game.
3. Emergence isn't enough.
I'm not convinced that increasing emergence is sufficient to increase player engagement.
The word I usually use for this effect is "investment." Players who stay -- and keep paying -- are those who become invested in the world of a game. It becomes a place they enjoy being in, and want to come back to, and want to see more of. I think emergence can contribute to creating that sense of place; I'm just not sure it's the primary cause of investment. The appearance and sounds of a world also matter, as does the plausibility of the AI of non-player characters.
So, as one way of increasing the "feeling of place," I could agree with a judicious enhancement to emergent play. But I would suggest looking at emergent content as just one component among several for helping the world of a game feel more dynamic in a distinctive way, and thus increasing its ability to foster investment by more players.
4. You only pay once for an emergent game.
By making an emergent game, you're choosing to put all your money-generating eggs in one basket.
You can either make multiple non-emergent games with static content, or you can make one (or a very few) emergent games with dynamic content. Making just one game that players can happily play in for years (because new content keeps emerging) means you get one shot at their money as an initial sale, versus multiple opportunities if you make a larger number of smaller, fixed-content games.
This is why I feel pretty strongly that monetizing an emergent-play game needs to be done by putting a price tag not on the base game itself but on additional developer-created system expansions (things that add more dynamic elements into the world) and on player-created content that can be purchased by other players (for which you as the developer get a cut of each transaction).
If you're just going to sell the one game -- because it has emergent content -- then you need to plan to make your money on long-term, ongoing improvements to that game, not just on initial sales of the game itself. Frankly, I'm inclined to think that the best way to monetize a game like this is to give away the base game for free (to seed it as widely as possible into the general game-playing public) and plan to earn all revenue from nominal charges to download new dynamic elements and scenarios created by you and by other players.
I hope it's clear that these comments aren't just objections to designing games to have emergent content. I like emergent content!
I do think it carries some potential gotchas, though. Those need to be considered along with the possible benefits of making a highly emergent game.