A couple months ago I wrote about realizing "enough was enough" with my feature-creeped game. Along with working on other projects, I spent some time in the next couple months cutting out features and polishing existing ones, all working under a "late-May" deadline.
Now, my new (much smaller) vision for Sawdust is feature complete... and ready to roll in a few days! I'm not 100% happy with how it turned out... but it is something, and it definitely was a great learning experience.
Lesson learned: Always prototype a few games or variations at a time
While I read various advice on game prototyping, and did build a Sawdust prototype, I failed to see the warning signs that the prototype itself has to be actually pretty fun. I thought the core concept seemed solid, so I'd add features to make it fun. Like software, "more features" does not mean better... one simple concept that is fun is better
It's completely possible to complete an actually fun prototype without a huge time investment. Now I feel I need to complete at least one or two of these before I should invest time in one of them. Otherwise, you could dump days into gameplay, and the game will never quite feel like a game. While I feel that Sawdust did eventually turn into a game (only at the end, really), it was a meandering journey that could have been a lot shorter if I forced myself under time constraints to discover some kernel of "funness" at the heart of the concept.
Lesson learned: Keep in mind what could be re-used
This is one I mostly did right. A lot of stuff I wrote for sawdust I wrote as re-usable packages. I also wrote about 30 minutes of a fun soundtrack which no longer fits Sawdust's smaller scope, so that can easily be reused for my next game project.
I did however make a lot of content -- notably a partially finished storyline and puzzle campaign -- that won't fit into the smaller-scoped Sawdust. If I ever get the urge, I might wrap that up into a bigger game that more closely reflects the original vision.