First fix!

Took me a while to get to it, but I finally got to return to my list of post-jam improvements based on the feedback received here (thank you to everyone who chimed in!). The single most important change that needed to be made was to ensure that it would *always* be possible to win. Indeed, there was a flaw with the build that was submitted during the game jam: since the preferences of each character are randomly generated on an individual basis, there was no mechanism in place to ensure that they would not have conflicting preferences (which would make the game impossible to win). Given the time constraints and the fact that this game was my first attempt at exploring puzzle game concepts, I was not able to figure out how to implement a mechanism to ensure all games can be won, so I settled for a compromise, tweaking the design to ensure that the impossible combinations would be, at least, kept to a minimum.

Figuring out how to exclude impossible combinations was no easy task, especially given that this game was made during a game jam, and thus its code doesn't have the most flexible architecture to begin with. I ended up devising a solution which uses coroutines and hashsets to keep generating new combinations as long as the previous combination is not possible. Once a possible combination has been generated, UI controls are enabled and the player can start playing. In terms of performance, I have noticed no significant impact, so I would say so far this solution is a success! Next up, I will work on redesigning the UI to make it more intuitive, and to allow users to see which combinations have been tried as well as the results for these attempts. I expect this one will take a while since more design is involved, but it is the next order of priority as it was requested by many people. At the same time, I will implement a message which informs the player of the characters' preferences upon losing the game.