This is not so much a coherent postmortem as just some scattered thoughts after having released the game into the wild. Of course, if you want to ask me anything about the game, please do so. I should note, however, that these are just by own opinions, not necessarily those of my coauthor— not that we aren’t on the same page about the game, but I’m not going to presume to speak for anyone else.
- This was a fun game to write, particularly in the narrative voice of the protagonist and the conversations throughout the game. My coauthor and I enjoyed the brainstorming and development process, and working in a collaboration reduced the tediousness and frustration of coding and describing all the dead ends in a game this large.
- I wanted to do a story about espionage that wasn’t a goofy satire. The Bond media have become almost synonymous with the genre, almost to the point of parody. I wanted to write a lightly comedic game, but I also didn’t want to write an episode of Archer. Our main character was a spy, but he was a pretentious lawyer and career bureaucrat rather than an adventurous womanizer. We had a Q stand-in, but he only gave out a few modest devices; he was more useful for general intel. Our villain had the standard Bond-plot of escalating the cold war, but he didn’t have any doomsday lasers, hijacked nuclear weapons, or sharks with laser beams attached to their heads. We also wanted to give the player at least a taste of more routine intelligence work with coded messages, dead drops, mission dossiers, and so on. (Unfortunately, I never found a place in the game for a numbers station.)
- Most of the reviewers suggested that the game was too difficult, and they were probably right. The feedback from playtesters suggested that the game was hard, but the odd thing was that they got stuck on different puzzles. We wound up tweaking some of the puzzles and giving the player access to the endgame with 7 out of 8, rather than all 8, puzzles solved. (The exact mechanic is a bit more complicated than that, but the criterion reduces to having solved 7 out of 8 puzzles in most scenarios). In retrospect, though, it might have been better to lower to endgame condition to 6 out of 8, as well as to cut the second of the three endgame puzzles.
- We probably should have reduced the list of verbs a bit. I tried to reuse verbs to avoid having a single eigenverb attached to each puzzle, and that also meant that it would be impossible to adopt the conceit of having a verb spontaneously destruct and fall out of your inventory on solving its relevant puzzle (which would be unmotivated anyway). As a result, the narrator would usually be lugging around a dozen or so verbs by the end of the game. I also toyed with the idea of cleaning out verbs and adverbs that were no longer necessary at various checkpoints in the game, but that didn’t seem natural either. Besides, I liked the idea that the demon was effectively gaining power or leveling up as he advanced through the game via the verbs and adverbs he collected, and I didn’t want to take that sense of accomplishment away from the player.
- We deliberately made most verbs and adverbs fail quickly and unremarkably. The mechanics in the game are subject to a combinatorial explosion, and we wanted to avoid giving red herrings or distractions to the player. Furthermore, adverbs have (with a few exceptions) no effect unless they’re required for a puzzle, and then they’re clued by the fact that the bare verb does something nontrivial but not quite useful. In short, we kept the runtime complexity of the game linear: try all the verbs in succession, then try the adverbs in succession for the verb that produces something interesting. (Ideally, of course, there would be clues in the scene to obviate the need for a brute-force search.) On the positive side, this reduced the size of the space of actions for the player to try and made the puzzles more tractable. On the negative side, it did mean that there were some unrealized opportunities to add some flavor to the game.
- Along similar lines of simplifying the puzzles, we considered the idea of having the adverb mechanic involved in some bonus puzzles in the first area, or even as alternative solutions to the existing ones. Ultimately, we decided that we had enough puzzles. (We had a lot of puzzles.) I’m not sure it was communicated well enough to the players, though, that the four main areas of the game (the Casino, the Hotel, the Conference, and the endgame) were mostly isolated.
- The (code)names of the demons in the game are kinds of reptiles, and those of the angels in the game are kinds of birds. There’s no significance to that other than the fact that I like reptiles and birds.
- Since a number of reviewers mentioned it, I feel obliged to point out that I do know the meaning and etymology of the word “anthropomorphize.”