I've a fairly complete design document together, and the project I'm putting together ("Vehicle Wars" is just the project title) looks completable. I shouldn't be getting stumped this time because, for the first time, the game as designed is pretty much fully playable. I've also endeavored to keep things simple, which certainly helps bridge the gap from impossible to possible.

I'm a bit rusty - too much time spent away from Dream Maker from the duress of encountering designer's block - but I've still enough coding experience under my belt that once I get going I can realize things pretty quickly. About 8 hours in, starting from ground zero, I've established:

• A somewhat complete 4.0x skin, without much code behind it but I'm not anticipating any changes to the layout. (Again, I'm keeping the game simple - no more multiple tab controls on swapping panes for me.)
• A customizable vehicle framework. (With ugly placeholder art - I'll need to add some beautification code at some point.)
• The ability to leave the map and enter a base room that allows vehicle customization.
• The turfs types I will be using. (Also with ugly placeholder art.)
• A 700x700 map that will function as the persistent world space in which the action takes place
• The map cut into 49 separate (yet freely traveled between) areas of roughly 100x100 spaces.

The next milestone will be dynamic map generation using those 49 separate areas. Right now, I'm pretty much stuck with an empty map - I guess once you draw the area on it, the world.turf is no longer applied to it. (Future versions of the game might include elevation, but for now I'm sticking with a flat map.)

After that, I'll actually have a backdrop in which to operate, and it will be on to the milestone of actually fleshing out the GUI enough to have my little player-controlled vehicles driving about.

These are both milestones I've accomplished before in previous prototypes, and I don't expect any major snags there.

The final milestone is relatively new territory for me: introducing dynamically generated conflict. I have some interesting plans here emphasizing player cooperation in a predominantly PvE conflict. I'm moving the focus away from personal accumulation and greed to a more communal progress focus because I'm thinking that may prove to be a more interesting game mechanic than Haves and Have Nots.

However, there will be a PvP element to it. I can draw from previous experience with online games that some players just want to raise havoc. Rather than forcing them to try to undermine other players from within, I think I've an even better means that will really let people flex their destructive muscles without stabbing anyone in the back. (Incentive alone isn't enough for griefers, unfortunately - backstabbing will be prohibited by code whenever possible. I have a goal of a game that maintains itself without requiring anyone is banned, but this may be too optimistic - I've learned never to underestimate the creativity of one hellbent on getting your goat.)

Even though the scope is persistent-world focused, the design current has a victory condition and a losing condition. Winning or losing will not be accomplished in one evening - this persistent-state game shoots for allowing the players to develop an epic story. Both lead to the players starting over on a new world, but only winning grants a persistent reward.

Three milestones isn't bad, factoring in a day or two for each. I'd like to say this will be ready for testing in about a week... but I've got class coming up on Tuesday and Thursday and it's hard to say what sudden death problems will manifest during a senior capstone project.