So, over the past week I've created several prototypes, all of which were quite dull. I learned a lot!
Let's talk about boring gameplay. First things first:
Boring gameplay is not bad
There are many genres which rely on boring gameplay - oldschool RPGs, strategy games, train sims, life sims, etc. The moment-to-moment gameplay of these games is incredibly dull. Rather than boring gameplay making boring games, the gameplay simply turns transparent and moves the play density deeper into the world.
These genres have more action-oriented equivalents. Zelda games have action-packed fights, navigation challenges, and even observation challenges. In turn, the gameplay density lies closer to those experiences and Zelda games do not have much statistical complexity. They focus on an endless supply of interesting, unique places and encounters.
Without analyzing it too deeply, it's possible to fall anywhere on the spectrum. I think there's some kind of conservation of complexity, though: Elder Scrolls games have really mediocre action combined with mediocre statistics. I can't think of any game which has both good action and good statistics, perhaps simply because it'd be asking too much of the player to focus on so many things at once.
In the end, it seems to be all about the pacing system. In Zelda, players are paced by the layout of the level and the fluid, attrition-filled fights. In most RPGs, players are paced by the stats - how fast and flexibly you can level and improve gear. In an Elder Scrolls game, the player is paced a bit by each, but it's relatively flat in both cases.
Elder Scrolls games are interesting to analyze, because you can see them steadily drift more towards the action side. Each new game is a little more action-oriented, and the leveling/equipping becomes steadily less complex. They also allow the player to choose where on the spectrum to fall - a more statistically complex mage, or a less complex warrior.
The same holds true of tactical/strategy games. Something like Civ has transparent moment-to-moment play and extremely complex statistical play. Something like Starcraft has a bit of a balance, with much stronger moment-to-moment play but much less extensive statistical play.
The reason I started to think about this is because I created several very boring prototypes. My problem with the prototypes is that I hadn't properly thought of the early-game pacing mechanic. That is, I hadn't bothered to think of what the player would push against. Everything was either easy or impossible.
This got me thinking about diving games.
One of the big problem with games about being underwater is that being underwater is pretty dull. Most modern diving games try to fight this off by injecting action into the moment-by-moment play. OH A SHARK DAAAAMN!
But it never seems to work, because adding in moment-to-moment action doesn't actually provide a pacing mechanic. It just introduces an arbitrary challenge. It comes off as somewhere between annoying, dull, and petty.
What you need to think of, if you're creating a diving game, is what the player will push against. This is how they will express themself, how they will pace themself. It's not quite the same as progression: progression is often a gating mechanic, not a pacing mechanic. IE, the plot of an RPG is not the pacing mechanic, because the player cannot play around with it. The leveling is.
Let's assume a transparent diving mechanic. That is, the player never faces any real interface challenges. They are diving, and hopefully enjoying the experience, but they aren't fighting sharks or being swept around by difficult currents in real time.
Because the moment-to-moment play is transparent, there needs to be a deeper kind of play.
You might consider something like analyzing fish or searching for downed ships or chasing dolphins. While these are underwater activities, they do not allow a player to express themselves. It is difficult to play around with the concept of identifying a fish.
You might consider costume/avatar changes. While this is self-expression, it isn't pacing control because it has no feedback mechanism. The player cannot play around with it because there's not any real interactivity. It's just a bonus.
As I consider an underwater game, I'm starting to consider ways the player can express themself - a system that feeds back as time goes on.
Fundamentally, it's probably not about any particular area of the ocean. Our moment-to-moment gameplay is too transparent to make levels expressive, in the same way that interacting with a town is not a particularly expressive part of an RPG. But, on the other hand, the levels do have to have the kernels which lead to self-expression, as a town has shops and quests and things that make us feel like the world exists.
In an RPG, the progression lies within the growth of the characters. However, we have nothing for our characters to grow relative to, not in any complex manner. There's no fights, so improving our stats and performing special moves is not important. Well, some stats can affect how we might explore, but it's not enough to really count as "expressive".
We could still use characters, but we'd have to change it from statistical growth to personal growth.
We can have both diving buddies and "surface" NPCs. Over time, you can do things to make them happier or better off, and they will come to like you more, their personalities might improve, and they might do interesting things. In order to properly link them to the world, they need to interact with it: these characters cannot simply be talking faces. Diving buddies might help you spot things or extend your air. Landlocked buddies might give you equipment, new locations, tend a reef, send you work, or clean up an area before you go diving. the line between landlocked and diving might be blurry, too.
That could actually be enough, as long as the method this happens is properly paced. That means it'd probably have to be largely done in-level. You might collect science samples, observations, pick up trash, scan/repair infrastructure, mark possible locations, and so on in a level, which would make specific people friendlier. Rather than simply being "friendlier" on some kind of flat scale, it's important for the player to be given options here, to allow them to express themself. So the player would probably spend these "friendship points" on some kind of upgrade tree that affects activities and reactions.
I also like the concept of front-loading it: promising people that you'll do X thing on this dive. If you accomplish it, you get more points. If you fail, you lose points. Choosing what to promise which people on any given dive is a great opportunity to self-express, and it can be done independently of the game's opinions on what a dive should be about.
You could add in a bunch of other things. Cash, sidequests, jobs, "bounties" to locate specific rare species, time limits, team coordination, revisiting locations that you have improved, etc. But I think that if you wanted to take this approach, the core element would definitely and obviously be the people around you.
There are other options. You could make upgrading your rig the pacing mechanic, for example. Or make the moment-to-moment play the focus. But I don't think that has as much punch.
Anyway, just thinking out loud.