I originally came up for this idea last month as part of Thing-a-Day for a game that I'd orchestrate for/with my family. I've linked to the original context just above but I'm also going to just cut and paste the entire text here so that I have it in a location closer to under my control. So here's the concept document:
A game for my family
February 23rd 6:29 :42 am by Christopher Weeks
tagged game LEGO plan RPG (Daily thing), Getting organized, Little people
This is a project plan and a design document for a game that I'm considering setting up for my family to play collaboratively over time. It is a sort of hybrid simulation/miniatures/role-playing game that will be played over time through face-to-face meetings of some or all of the players and solitaire play, all of which will be rendered in LEGO models and spoken/written narration and conflict resolution. Further, many facets of the game will be computer-assisted or facilitated. The roles taken by a traditional RPG-GM will be apportioned to the software tools and to the players on an ad-hoc basis. Models for game play that I have in mind include: Universalis (story-game, GMless play), HeroQuest (RPG, personal characterization/general abilities), Monster Island (PBM, vague theme), Evil-Stevie's Pirate Game (miniatures, rendering only), Castle World (collaborative fiction, style)
In this game, the characters will be shipwrecked and wash ashore on an island. The island will be comprised of "squares" including an elevation, a biome and features (monsters, people, mysteries, +?). These attributes will be generated by a computer before the game starts. As the island is explored, each square will be modeled as a 4×4 stud LEGO structure that grows with our game-play. A "square" should be understood to be an arbitrary large area: an acre, a square mile, something like that. On a weekly(?) basis, the family will decide what the characters are doing during that turn. We might have some in-character play or maybe just hang out and moderate one another in author-stance. The mechanical systems will be flexible enough to be useful in any kind of play that evolves. While we'll have an island model, we may also model scenes, locations etc, or write, what amounts to fan-fiction. I'm thinking that we'll set up a blog or a wiki to allow posts from all the players as well as a repository for record-keeping, etc.
There are two starting places. The first of these is behind the scenes: generation of the map, building facilitating software, etc. The other starting place is how play begins: creating characters, landing on the island, etc.
I'm imagining that the island will occupy six grey baseplates and thus be 96×144 studs or 24×36 = 864 squares. I'll write an application that generates a grid of squares with heights based on proximity to two "logical peak" locations — that is, the random elevations will tend to be higher as proximity to either of these locations grows and lower as proximity to the outside edge grows. Then it'll crawl across the land and deposit biomes based on elevation, etc. Then it will randomly distribute monsters, people and mysteries.
Regular resources — to the extent that we decide to play a resource game, will be dependent on the biomes expressed on the map. E.g. wood can be harvested from forest or jungle squares, peat from swamp, rock from rock, whatever. Monsters, people and mysteries are special, ill-defined resources independent of elevation and biome. As we play and encounter these special resources, we'll just decide what they are and how we play with them. I'm thinking that each of those should have a 1/60 chance of appearing in any square — giving about 43 special squares (on average).
Because I want knowledge of the island to be blind, the generator will output a data file that is read by an explorer application. All the explorer does is allow us to punch in a coordinate and reveal complete details about that square and elevation/biome of all surrounding squares. My default assumption is that we won't know the special resources of any place that we haven't visited (though that could change in game play — through player preference or special character abilities, etc).
I'm thinking of using either HeroQuest or a mod of a game that Mike Holmes is working on and I got to playtest for character representation and mechanical systems. So we'd start out play by creating characters with a pool of abilities from each of three categories: homeland, occupation and personal. An optional step is to have a fourth "trademark" category spring up as a result of the shipwreck or something else shortly after arrival on the island. This final thing could be entirely fantastic.
Once characters are created, we jointly decide where we're washing ashore and build up the base model to represent the water and land that we encounter. After that, the plan kind of ends. We play. Part of play will be figuring out what play looks like.