Posted On: 2022-01-24
At the start of each calendar year, I like to look back on the past year, and plan the next one. This time around, I'll be solely focused on looking forward, as I recently did two retrospectives - one for the year through November and one for the last couple months of the year.
Last year featured two different writing goals: the ambiguous "write a lot" goal, and the concrete "write for one month" goal. Though the former didn't really materialize, the latter was not only met but exceeded - as I was able to accomplish more than I'd planned to within that single month. Learning from that success, I intend to set similar goals for this year: while I hope to "write a lot", I will set aside more specific intervals: at a minimum, 2 month-long writing sessions and 4 two-week writing sessions. As a stretch goal, I'll also aim to write more if time/availability permits, but so long as I can hit those concrete goals I will consider my writing a success.
One important deviation from last year's writing goals is that I am intentionally omitting the work it takes to get that writing in-game. Most of last year's delays were predicated on the writing demanding more of the code than I could reasonably achieve, but, as I found in November, using Ink as my writing engine makes getting it in-game a bit less important. Ink's editor has basic playtesting functionality built-in, allowing me to vet the flow of the text and the feel of making choices regardless of the state of the game proper. What's more, I am finding that having a written script is actually serving as a de-facto specification for missing features: I often have to use choices to hand-wave a missing feature (ie. the choice of "go to woods" or "go to swamp" in place of actually moving the character to said location), making it clear what exactly the game needs, versus what would be nice (but not strictly necessary for the story.)
Of the missing features that my script has revealed, none is more important (or troublesome) than autonomous agent movement. Often my script calls out for characters to perform certain bits of acting (move here, gesture that way, etc.), but I currently have no way of having the in-game character perform the described actions. This is something I've wanted to have for a long time - but, I've only worked on it piecemeal, and made little progress doing so. This year, however, I intend to change that - I have only one programming goal for the entire year: get autonomous agents working. Realistically, I will probably accomplish other programming tasks as well (a year is a long time for only one feature), but, since I don't want anything to distract from this most-important feature, I am treating all other programming work as "stretch" goals - to be considered only when I am finishing up work on the one that matters most.
I think I've got a pretty realistic set of goals for the coming year. By learning from the mistakes of last year (ie. ambiguous writing goals), I've got a lot more confidence in this year's plan. The simplicity of the plan belies its importance, however: autonomous agents is one of the few foundational systems yet unfinished - and, as I've seen from my writing drafts so far - it's one that will open many doors for the story (both figuratively and literally). I hope you'll continue with me in this coming year, and read along as I document this strange journey of making a (rather ambitious) game.