Surf Solar said:
I'm on some creative dead end at the moment
staring at an empty design document all day isn't really funny
Chances are, you get overwhelmed by all the possibilities of what you COULD DO, and find yourself lost trying to find what you SHOULD DO right now. It seems that you can go wrong in so many ways, so you get stuck trying to theorize the right way, and then the fear comes in.
If that is the case, let me offer some advice distilled from a river of tears.
Creating something terrible, RIGHT NOW, is 1000 times better than creating nothing at all.
Let's say you create a maze full of tight corridors and a premise that's entirely underwhelming. It's not just "a lot better" than nothing, it's MONUMENTALLY better.
You're struck by a sharp realization that you have opinions. Like a sonar ping, it lets you know where you are.
For example, now you know that your level has to have areas that can be easily identified, because you get lost, and to go easy on narrow corridors because your NPCs kill themselves with rocket launchers in them.
This gets you momentum. You make a new map, faster, and it's 5 times better. You change the premise to work with the map. Make areas more distinctive. Then you tweak it as new ideas hit you. And it keeps going, unless you let the momentum die.
--------------
Regarding design documents.
FUCK the documents. Here's why.
For example, you want an assassin to teleport behind a door when you do a dialogue option, and you document this and several similar occurrences as part of a whole quest chain, until you actually do it, and realize that Teleport() function in the engine was conceived with a different purpose in mind, and upon doing a Teleport(), the NPCs' home faction is set to the faction of the place he gets teleported to, so you can't make him attack anyone!
Then you go, oh fuck, now I have to make that NPC be always present in the room, ready to stab people in the back, and you go and update all the similar Teleport() occurrences in the document.
Don't start a mountain of theoretical documentation, because you will have trouble connecting it to gameplay. You can't tell in advance how it will work, whether the engine has the right tools to do a quest that way, etc etc. Update the document, update the script, update the document in every place that your new discovery affects... You will get lost in a constant feedback loop between theory and practice.
Put something in, make it work, and document it briefly. One quest.
Then another.
Make your "brief world description+whole narrative arc of the main quest" fit on one printed piece of paper in a normal readable font size. Put that piece of paper near you.
Go.
P.S. never make quests tied to other quests. If you have:
quest 1: bring NPC some petrol
quest 2: bring him matches so he sets himself on fire
DO NOT make quest 2 look up the "mission accomplished" status of quest 1. Instead, make it look up whether NPC has petrol (and thus is ready to ask for matches).
This will make the structure a lot more robust. If later you decide that the quest chain needs shortening, you delete quest 1 and plant some petrol on the NPC. You don't even have to touch quest 2.
And since you all seemed to like this foggy thing, here's another one on the same map:
But I hope this is just temporary.
