I mentioned earlier that Procedural Generation in Game Design is available, and that I have a chapter in it. I’ve now had a chance to look at a few of the chapters that I didn’t read previously during the publishing stage, and wanted to highlight a few of these as especially relevant to IF readers.
Joris Dormans’ chapter on Cyclic Generation talks about design patterns for procedural dungeons, including the most systematic section on the deployment of locks and keys I’ve recall seeing anywhere. He identifies concepts such as single- or multipurpose keys, consumable and reusable keys, asymmetrical and “valve” doors, safe and unsafe locks and keys, and other concepts; if you’re looking for design patterns for puzzle games gated on geography, this has a lot of ideas you might want to raid. You can get some of the same material from Dormans’ lecture at PROCJAM last year in Falmouth, but the article gives you more reference material.
Ben Kybartas writes about story and plot generation using expansion and rewriting grammars. The rewriting rules (“secondary rewrite rules” in Kybartas’ terminology) take a simple plot and then add complications to it based on what elements already exist in the game world: for instance, a simple plot about someone cheating at poker could be expanded with a complication that they have an accomplice in cheating — but only if there is someone with emotional ties that would make them willing to participate in such a deception. Rewrite rules could even add nodes to the story that provide player choice. I would have welcomed more information from finished games about how this method goes down in practice.
Jason Grinblat’s article on Emergent Narratives and Story Volumes talks about how procedurality can be used to define the themes of all the possible stories to emerge from that system; it ends in a close study of the tabletop storygame Fiasco, but also includes examples from Caves of Qud.
Mark R. Johnson writes about several aspects of Ultima Ratio Regum, but in particular about the procedural generation of dialogue for different character types and personality styles — something that’s obviously of strong interest and ties into some of the work I’m doing at Spirit AI as well as in my own practice.
(PS: 2017 PROCJAM Kickstarter fundraising is in its last days; you still have an option to help kickstart it.)