The last week of April came and went with the ultimate result being a simple one: If I spend all day brainstorming a design on paper, I can paint myself into corners nicely.

When an artist paints, s/he doesn't spend all day deciding what to paint. That's what I've been doing, so very paralyzed by what I could create that actually creating something is never going to happen.

Finding myself in a corner from which no fun can escape, I've basically been re-examining core concepts of what's fun about games.

Fun results from a mind very well occupied, neither frustrated from lack of progress nor bored from no apparent challenge. This much we know, with little deviations depending on if you're Chen or Koster.

The problem I'm attempting to conquer is a little more complicated than that. You might be meaningfully occupied with the activity in a game, but if you think about the why of it, you can derail yourself handily:

The trouble is that there's rarely any compelling reason to commit to any activity in a virtual environment. Accumulating virtual loot and levels is only an imaginary achievement. PvP tends to be pointless when people just respawn anyway. Dynamic content prevents persistence, so why build an empire?

I'm a persistent-state online game fan at heart (BYOND being an excellent platform for persistent-state online games) and it's this little crux that is bothering me. How can there truly be meaningful dynamic content if it's just one "why bother" away from being rendered null and void?

Apparently it's "in the air" right now, because (as Scott Jennings made me aware) Richard Bartle's recent lecture was about this even as I unknowingly whined impotently about it the day before.

Bartle's suggestion is that we look to accommodate different player types. Those who want their hand held typically not welcome dynamic content, while those who are expert at these kinds of games have already asked "why bother" when confronted with such silly activities such as raiding or PvP when everything and everybody respawns anyway.

As for me, I'm looking at the roguelike genre. Call me crazy, but I think the concept of randomly generated worlds can create quests of infinite replayability rather easily. All the developers have to do is keep introducing more and more interesting possibilities to the random generation trees.

For a proof of concept, look at Nethack. You can go on a quest to receive the Amulet of Yendor many, many times and it would take a very long time until you get bored of it. This isn't a unique thing, you can repeat the result with ADOM or Dwarf Fortress.

The main counteracting trouble behind this randomly generated approach is that pattern recognition is a function of intelligence. Thus, hand-crafted quests and maps would seem to beat out randomly generated quests and maps simply because content made by people will likely be aware of pre-existing patterns and seek to deviate from them.

So this is pretty much what I've been grinding my mental gears on lately. A massively multiplayer roguelike with aspects of player-created content. Oh, and I'd like it to be fun to play, too.

This might take awhile. The devil is in the details.